Molecular Analyses of Alpha 1-Antitrypsin Variation and Deficiency

n 2.13

MOLECULAR ANALYSES OF ALPHA l.ANTITRYPSIN VARIATION AND DEFICIENCY

A thesis submitted for the Degree of DOCTOR OF MEDICINE

tn

The Depafiment of Medicine

The University of Adelaide, South Australia

by i Mark D. Holmes, MBBS, FRACP

July, 1992. Delicøtel to

Alrienne, E[eøtLor ønl Aftco[ø, CONTENTS

Page

ABSTRACT vilt

DECLARATION ix

ACKNOWLEDGEMENTS X

LIST OF ILLUSTRATIONS xi

LIST OF TABLES xvi

AIMS AND GENERAL INTRODUCTION xvii

1. LITERATURE REVIEW

1.1. Early History 1

1.2. The Normal crlAT Gene 4

1.2.1 al AT Gene Locus 4

1 .2.2. crl AT Gene Structure 5

1 .2.3 crl AT Gene Expression I

1 .3 The Normal crl AT Protein 10

1.3.1 crlAT Protein Structure 10 1.3.2 crl AT Biosynthesis 12 1 .3.3 al AT Function 13

1.4 c¡1AT Variation 16

1.4.1 Classification of crlAT Variants 17 1.4.2 Detection of alAT Variation 18 Page

1.4.2.1 GlAT Serum Levels 18 1.4.2.2 Electrophoretic Variation of a1 AT 19 1.4.2.3 Analysis of cr,1AT Gene Mutations 20

1.4.3 Normal alAT Variants 22 1.4.4 ct1 AT Deficiency Variants 25

1.4.5 Dysfunctional cr,1 AT Variants 29

1.5 Molecular and Cellular Basis of crl AT Deficiency 30

1.6 olAT Variation and Disease 33

1.6.1 Lung Disease 33 1.6.2 Liver Disease 36 1.6.3 Other Disease Associations 37

1.7 Treatment of ol AT Deficiency 38

2. MATERIALS AND METHODS

2.1 lntroduction 41

2.2 Reagents, Radiolabels, and DNA Restriction Enzymes 42

2.3 Study Population 42

2.3.1 Source 42

2.3.2 Cli n ical C haracteristics 43 2.3.3 Bronchoalveolar Lavage 43 2.3.4 Monocytapheresis 44 2.3.5 Liver Bioosv 45 2.3.6 Drug Therapy 45 Page

2.4 Analysis of olAT Phenotype 45

2.4.1 General 45 2.4.2 cr,1AT Levels 45 2.4.3 al AT Function 45 2.4.4 lsoelectric Focusing of Serum 46

2.5 ol AT Gene Sequence Analysis 47

2.5 1 General 47 2.5 2 Extraction of High Molecular Weight DNA 48 2.5 3 lsolation of the Human cx,lAT Gene by Cosmid Cloning 49 2.5 4 Single Stranded DNA Generation by the Polymerase Chain Reaction 53 2.5 5 DNA Sequence Analysis of ol AT Variants 56

2.6 ldentification of crlAT Gene Point Mutations 58

2.6.1 General 58 2.6.2 Allele Specific Oligonucleotide Analysis 59 2.6.3 Allele Specific Amplification with the Polymerase Chain Reaction 59

2.7 Analysis of ln Vivo cll AT Biosynthesis 60

2.7.1 olAT mRNA Analysis 60 2.7.2 crlAT Protein Biosynthesis 61

2.8 Analysis of slAT Biosynthesis Using ln Vitro Svstems 63

2.8.1 General 63 2.8.2 Production of Recombinant Plasmids Containing Variant crl AT Genes 63 Page

2.8.3 crl AT mRNA Cell FreeTranslation ln Vitro ...... 66 2.8.4 Transient Expression of Human olAT cDNAs in COS I Monkey Kidney Cells 66 2.8.5 Permanent Production of Human crlAT in Murine NIH-3T3 Fibroblasts by lnsertion of a Human alAT cDNA by Retroviral Gene Transfer 68

3 ALPHA 1-ANTITRYPSIN NULLsran¡1s 1¿¡¡e: THE INTRACELLULAR MECHANISM CAUSING THE ABSENCE OF DETECTABLE ALPHA l.ANITRYPSIN

3.1 lntroduction 70

3.2 Methods 70

3.3 Results 75

3.3.1 Patient Characteristics ...... 75 3.3.2 cr1 AT Phenotype Analysis 75 3.3.3 Analysis of crl AT Genotype 75 3.3.4 Evaluation of a Nullgranire rals Homozygote for cl1 AT mRNA Transcripts ...... 77

3.4 Discussion 77

4 THE MOLECULAR BASIS OF THE LUNG AND LIVER DISEASE ASSOCIATED WITH THE ALPHA l.ANTITRYPSIN DEFICIENCY ALLELE Mmalton

4.1 lntroduction 82

iv Page

4.2 Methods 83

4.3 Results 84

4.3.1 Patient Characteristics 84

4.3.2 ldentification of the ø1AT M¡¡s¡1en Protein . 87 4.3.3 Functional Activity of the crl AT Mmatton...... 87 4.3.4 ldentification of the M¡¿¡sn Mutation 87 4.3.5 lnheritance of the Mmahon Gene...... 92 4.3.6 Evaluation of alAT mRNA Transcripts from Cells Expressing the Mmatron Gene 92 4.3.7 Synthesis and Secretion of olAT by Blood Monocytes Expressing the Mmatron Gene...... 97

4.4. Discussion 99

5 ANALYSIS OF THE MOLECULAR BASIS OF THE ALPHA l.ANTITRYPSIN DEFICIENCY VARIANT Wbethesda

5.1 lntroduction 104

5-2 Methods 105

5.3 Results 106

5.3.1 Patie nt Characteri stics 106 5.3.2 Phenotypic Analysis of u1 AT W6s1¡ssda ...... 106 5.3.3 Sequence of the crl AT Wbethesda Gene 106 5.3.4 Confirmation of lnheritance of crl AT Wbethesda 109 5.3.5 Biosynthesis of olAT Directed by the c¡1 AT Wbethesda Allele...... 109

V Page

5.4 Discussion 114

6 MOLECULAR HETEROGENEITY AMONGST THE P.FAMILY OF ALPHA l.ANTITRYPSIN VARIANTS

6.1 lntroduction 120

6.2 Methods 121

6.3 Results 126

6.3.1 Patient Characteri stics 126 6.3.2 cr1 AT Phenotype Analysis 126 6.3.3 Genetic Basis of al AT Pbwel and ol AT

Psainr atbans Alleles 129 6.3.4 Confirmation and lnheritance of the Ptowe¡ and Ps¿¡.1 atbans crl AT cDNAs 132 6.3.5 Biosynthesis of crlAT Directed by Prowe¡l

and Ps¿¡n1 atbans crl AT cDNAs ...... 135 6.3.6 olAT mRNA Transcripts in the PtowellZ lndividual 139 6.3.7 Effect of Tamoxifen Therapy on crlAT Serum

Levels in a P¡e,¡,s¡¡Z lndividual ...... _ 139

6.4 Discussion 139

7 CHARACTERIZATION OF THE NORMAL ALPHA 1. ANTITRYPSIN VARIANT Vmunich : A VARIANT ASSOCIATED WITH A UNIQUE PROTEIN ISOELECTRIC FOCUSING PATTERN

7.1 lntroduction 146

VI Page

7.2 Methods 148

7.3 Results 148

7.3.1 ldentification of the al AT V¡u¡¡s¡ Variant . 148 7.3.2 Elucidation of the Molecular Basis of the Vmunich Variant..... 150 7.3.3 Confirmation of the lnheritance of the Vmunich Allele 150

7.4 Discussion 156

8. CONCLUDING REMARKS 159

9. BIBLIOGRAPHY 161

10. APPENDICES

A. Publications Arising From This Thesis 212

B. Curriculum Vitae 214

vil ABSTRACT

Analyses of six rare alpha 1-antitrypsin (a1AT) alleles were undertaken to determine their molecular basis, assess the ol AT biosynthesis directed by these alleles and, examine their clinical correlation.

The absence of detectable cr,1AT in the serum, resulting in premature emphysema, due to the Nullsranire 1¿¡¡s allele was demonstrated to be due to absent alAT mRNA. cr,1AT Mmatton (Phesz TTC "" delete TTC), a severely deficient variant, is associated with emphysema and liver disease. Examination of al AT biosynthesis in blood monocytes from an Mr¿¡1on homozygote revealed intracellular aggregation of a1AT, a process which not only gives rise to serum alAT deficiency but is also likely to be critical in the genesis of the hepatic damage due to this cx,1AT variant,

The W6s¡rres¿a alAT allele (Alagse GCf ", Thr ACT), a moderately deficient variant with the potential of risk of emphysema if inherited in a heterozygous fashion with a severely deficient allele, directs the synthesis of a protein which is degraded intracellulady as demonstrated by in vitro experiments in COS I monkey kidney cells transfected with a Wbethesda ul AT cDNA. cr,1AT Pbwe¡ (Aspzse GAT "" Val GTT), gives riseto serum crlAT levels of 24/" predicted and is associated with emphysema. Analysis of alAT biosynthesis directed the crlAT P¡e,¡s¡¡ cDNA inserted into NIH-3T3 cells by retroviral transfer revealed intracellular degradation of cr1AT. ln contrast, another olAT "P qAC ,,,, is a Variant", Psaint atbans (Asps+t Asn Af,C; Asp2s0 GAT "" Asp GAC) normal level allele with normal cx,lAT biosynthesis. Elucidation of the ø1AT biosynthetic defect associated with the P¡syys¡ allele directed successful treatment with tamoxifen to raise deficient crl AT levels.

The unique isoelectric focusing pattern for the Vmunich crlAT variant, a normal level alAT can be explained by its molecular basis (Asp2 GAT "" Ala GCT). The position of the amino acid substitution in the crlAT molecule changes the isoelectric point of only some of the microheterogenous forms of V¡y¡¡"¡ q1AT.

These studies demonstrate the utility of molecular analyses to discover the genetic basis of slAT variation, determine the resultant defects of crlAT biosynthesis, help to explain the clinical correlates associated with variant alleles and aid in a rational approach to therapy.

vill. DECLARATION

I declare that this thesis is a record of original work and that it contains no material which has been accepted for the award of any other Degree or Diploma in any University.

To the best of my knowledge and belief, this thesis contains no material previously published or written by any other person, except where due reference is given in the text of the thesis.

I consent to this thesis being made available for photocopying or loan.

Mark D. Holmes

July, 1992.

tx" ACKNOWLEDG EM ENTS

The work presented in this thesis was carried out between 1987 and 1989 during a three-year tenure as a Fogarty lnternational Visiting Associate in the Pulmonary Branch, National Heart, Lung and Blood lnstitute, National lnstitutes of Health in Bethesda, Maryland, United States of America.

I am indebted to my External Supervisor, Dr. Ronald G. Crystal, not only for providing me with an opportunity to pursue these studies, but also for his inspiration, enthusiasm and encouragement during my time in his laboratory.

I sincerely thank those people who taught me the techniques and procedures necessary to get started in molecular biology : Dr. Robert L Garver Jr., Dr. Mark L. Brantly, and in particular Dr. David T. Curiel who took it upon himself to take me step-by-step through many experiments and who took personal pride and satisfaction in the progress of these studies.

Many other people in the Pulmonary Branch were friends and co-workers and helped willingly in these studies : Ms. A. Chytil, Ms. L. Stier, Dr. H. Okayama and Dr. B.C. Trapnell.

I would also like to acknowledge the following contributions to the work in this thesis : Dr. C. Vogelmeierfordetermining the function of ø1AT M¡¡¿¡sn; Mr. G. Fells for purifying and concentrating the ø1AT of the SW¡erhesda heterozygote individual; Dr. M.L. Brantly and Ms. L. Stier for pedorming isoelectric focusing and measuring cllAT serum levels; Dr. S. Weidinger (Munich, Germany) for providing the Vn¡u¡¡6n câsê, Dr. W.D. Travis for examining the liver biopsies in the M¡¿¡1sn study; and Dr. C. Saltini for performing differential cell counts of bronchoalveolar lavage returns.

I wish to thank my Supervisor in Australia, Professor A.G. Wangel, for his invaluable help in preparing this thesis.

Finally, many thanks to Christine McLennan for her expert typing and editing of this manuscript and Dr. Geoffrey McLennan for encouraging me to pursue scientific research.

x LIST OF ILLUSTRATIONS

Page

FIGURE1.1 : crlATGene 6

FIGURE 3.1 : Pedigree of a Family with the crlAT

Nullgranite talls Allele 72

FIGURE 3.2 : Strategy for Genotypic Analysis of the NUllgranite 1"¡¡" Allele 73

FIGURE 3.3 : Genotype Analysis of Genomic DNA of Family Members for Presence of the Nullsranire fals Allele 76

FIGURE 3.4 Evaluation of q,1AT Synthesizing Cells of an

ol AT Nullsranite l¿¡¡s Homozygote for the Presence of ol AT mRNA Transcripts 78

FIGURE 4.1 : Evaluation of Hepatic Tissue of an Mmafton Homozygote ...... 85

FIGURE 4.2 : Pedigree of a Family Carrying the alAT Mmahon Allele 88

FIGURE 4.3 : IEF Pattern of crlAT Type Mmatton in Serum 89

FIGURE 4.4 : Time-Dependent lnhibition of Neutrophil Elastase by al AT Type Mmafton 90

FIGURE 4.5 : ldentification of the Mmahon Mutation in Exon ll by Sequence Analysis ..... 91

XI Page

FIGURE 4.6 : Strategy to Determine Genotypic lnheritance of the Mmafton Allele 93

FIGURE 4.7 : Demonstration of Genotypic lnheritance of the Mmahon Allele 94

FIGURE 4.8 : Qualitative Analysis of M¡"¡1s¡ alAT mRNA Transcripts ...... 95

FIGURE 4.9 Quantitative Analysis of Mr"¡on cx,1AT mRNA Transcripts ...... -. 96

FIGURE 4.10: Analysis of olAT Biosynthesis Associated with the cll AT M¡¿¡s¡ Allele 98

FIGURE 5.1 Pedigree of a Family with the alAT W5s1¡ss6¿ Allele 107

FIGURE 5.2 : Evaluation of crlAT Wberhesda by IEF 108

FIGURE 5.3 ldentification of the ø1AT W5s1¡s"6" Mutation by DNA Sequence Analysis 110

FIGURE 5.4 : Strategy of Allele Specific Amplification to ldentify the W6s1¡s.6" Allele 111

FIGURE 5.5 lnheritance of crlAT W6s1¡g"6" ldentified by Allele Specific Amplification 112

FIGURE 5.6 : ln Vitro Translation of W6s1¡ss6¿ mRNA Transcripts 113

FIGURE 5.7 Biosynthesis of slAT Wbethesda in COS I Cells 115

xil Page

FIGURE 5.8 : Pattern of the cllAT Secretory Defect Associated with al AT Wberhesda 116

FIGURE 6.1 Pedigree of a Family Carrying the P¡q'/vs¡¡ a1 AT 122

FIGURE 6.2 : Pedigree of a Family Carrying the Ps¿¡¡¡ atbans a1 AT 123

FIGURE 6.3 : N2 Retroviral Vector Constructs 125

FIGURE 6.4 : Characterization of al AT Variants Pbwe¡ and Ps¿¡¡latbans by IEF at pH 4-5 127

FIGURE 6.5 IEF with lmmobilized pH Gradient at pH 4.45 - 4.751o Maximize Separation between the

Pbwe¡ and Ps¿¡¡l atbans Variants 128

FIGURE 6.6 : Gene Sequence Analysis of the P¡s'¡ys¡¡Z and

M3Ps¿¡¡¡ atbans lndex Cases Around Codon 256...... 130

FIGURE 6.7 : Gene Sequence Analysis of the P¡o'¡,s¡¡Z and M3Ps¿¡¡l atbans lndex Cases Around Codon 341 ...... 131

FIGURE 6.8 l Genotypic Analysis by Allele Specific Amplification ...... 134

FIGURE 6.9 : ln VitroTranslation of M1(Val2ts¡, P¡svys¡¡ ârìd Psaint atbans mRNA Transcripts 136

xilt Page

FIGURE 6.10 Analysis of Human alAT mRNA Production by Modified NIH-3T3 Cells 137

FIGURE 6.11 Analysis of Human alAT Biosynthesis by Modified NIH-3T3 Cells 138

FIGURE 6.12: Evaluation of the crl AT mRNA Transcripts of a PlowellZ Heterozygote 140

FIGURE 6.13: Evaluation of Tamoxifen in the P¡e,¡s¡¡Z Heterozygote 141

FIGURE 7.1 Microheterogeneity of the Serum crlAT IEF Pattern 147

FIGURE 7.2 : Pedigree of a Family Carrying the cllAT Vmunich Allele 149

FIGURE 7.3 : ldentification of al AT Vmunich by IEF at pH 4.2 - 4.9 151

FIGURE 7.4 : IEF with lmmunofixation to Characterize al AT Vmunich 152

FIGURE 7.5 : DNA Sequence Analysis of the V¡u¡¡6¡ lndex Case 153

FIGURE 7.6 : Scheme for Demonstration of lnheritance of the clAT V¡u¡¡s¡ Allele by Allele Specific Amplification 154

FIGURE 7.7 : Demonstration of lnheritance of the Vmunich Allele 155

XIV Page

FIGURE 7.8 : Basis of the Unique IEF Pattern of ø1AT Vmunich...... 157

XV LIST OF TABLES

Page

TABLE 1.1 Normal crl AT Alleles 24

TABLE 1.2 o1 AT Deficiency Alleles 26

TABLE 1.3 alAT Null Alleles 27

TABLE 6.1 Allele Specific Amplification Primers 133

XVI AIMS AND GENERAL INTRODUCTION

olAT is the major inhibitor of neutrophil elastase in the serum and the lower respiratory tract. Deficiency of alAT is associated with premature onset of emphysema and in certain cases liver disease. ln 1987, at the time of beginning the studies presented in this thesis, the molecular basis of only the major deficiency variants was known. At the outset, a primary aim of this thesis was to determine the gene sequence of a number of rare crl AT variants. These analyses are detailed in Chapters 4 -7. Knowledge of the gene sequence was important to develop hypotheses regarding the structural implications to the alAT protein of the gene mutations and resultant amino acid changes and for subsequent identification of those variants at the gene level by methods simpler than sequencing. ln addition, the knowledge of the gene sequence allowed generation of the mutant alAT cDNAs lor in vitro analysis of crlAT biosynthesis associated with those variant alleles which, because they were found only in the heterozygous state with another crl AT variant, could not be adequately analyzed in vivo.

A second majoraim was to explore the biosynthesis of ø1AT directed by the variant alleles. Where possible, this was achieved by examining tissues and cells from the individuals with the particular variant. This could be done for the Nullsranire 1¿¡¡s ârìd Mmatron alleles which were found in rare homozygote individuals (Chapters 3 and 4). Where in vivo experiments were not possible, in vifro systems were utilized to examine olAT biosynthesis (Chapters 5 and 6). An aim linked to this was to compare transient versus permanenl in vitro systems for analysis of biosynthesis, and to this end, transient expression of cr,1AT in COS lcells was utilized in the study in Chapter 5 and a permanent cell line expressing human cllAT produced by retroviral transfer into mouse NIH-3T3 fibroblasts used in the study in Chapter 6.

Finally, an aim of this thesis was to apply the findings of molecular analyses of ø1AT to the clinical situation to : firstly, enable simple and rapid determination of the variant allele at the gene level for family studies

XVII. (Chapters 2 - l); secondly, use the knowledge of alAT biosynthesis to predict clinical sequelae of the particular alAT variant (Chapters 3 - O)' and thirdly, where indicated, to use the results of analysis of olAT biosynthesis to direct possible treatment for the deficiency of ol AT (Chapter 6).

xviii. CHAPTER 1

Literature Review

1.1. EARLY HISTORY

The ability of serum to inhibit trypsin was first demonstrated in 1897 by Camus and Gley and independently by Hahn in the same year. The majority (90%) of this trypsin inhibitory capacity of serum was subsequently found to be attributable to a glycoprotein with a sedimentation constant of 3.5 Svedberg units which resided in the alpha 1 zone of serum atler agar-gel electrophoresis (Jacobsson, 1953; Jacobsson, 1955: Schultz et al, 1955; Moll et al, 1958; Bundy & Mehl, 1959; Schultze et al, 1962). The name alpha 1-antitrypsin (a1AT) was proposed by Schultze et al (1962) who purified this glycoprotein. Most of the remaining antitrypsin activity of serum is due to cr2-macroglobulin (Kueppers et al, 1964; Eriksson, 1965).

Major scientific and clinical interest in crl AT was generated by the observations of Laurell and Eriksson in 1963 who, when analysing serum by paper electrophoresis, demonstrated an association between the absence of a particular alpha 1-globulin band and premature chronic obstructive lung disease. Three of 5 individuals studied with the missing globulin band had widespread pulmonary disease as did the sister of one of these. The missing globulin was shown to be alAT and they suggested the term slAT deficiency for this disorder. lt was also noted that in sl AT deficiency there was slightly reduced mobility of ø1AT on immunoelectrophoresis in agar gel without abnormalities of other serum proteins (Laurell & Eriksson, 1963; Eriksson, 1965). ln addition, there was a diminished trypsin inhibitory capacity of serum in these individuals to 12/" of normal, a feature which was used to demonstrate the hereditary nature of the absence of crlAT in members of the families of those with early onset emphysema (Laurell & Eriksson, 1963; Eriksson, 1964; Eriksson, 1965). The meticulous and extensive studies of Eriksson (196a; 1965) added much more strength to the evidence of an

1 association between olAT deficiency and emphysema, as well as highlighting the premature onset of lung disease in these individuals.

lnitially, cr,1AT deficiency was thought to be inherited in an autosomal recessive fashion (Eriksson, 1964; Eriksson, 1965). However, coincident with the studies of Laurell and Eriksson (1963), Fagerhol and Braend (1965) were investigating the genetic polymorphism of the prealbumin region of serum by starch-gel electrophoresis. Using this very sensitive method, they found 5 different prealbumin phenotypes and suggested codominant inheritance of 3 different alleles (i.e. each of the two alleles of an indivídual contribute equally to the electrophoretic phenotype). The prealbumin prote¡n was later shown to be alAT by antigen-antibody crossed electrophoresis utilizing an anti-a1AT antibody (Laurell, 1965; Fagerhol & Laurell, 1967). These studies and others established that ctlAT production was directed by a single autosomal gene locus expressed in a codominant fashion and have formed the foundation for the intensive study of olAT variation that has since followed (Eriksson, 1964; Kueppers et al, 1964; Eriksson, 1965; Fagerhol & Braend, 1965; Axelsson & Laurell, 1965; Fagerhol & Laurell, 1967; Fagerhol, 1969; Fagerhol & Gedde- Dahl, 1969). Early in the study of cr,lAT deficiency it was recognized this disorder was very common, for example, a gene frequency ol 0.024 for the deficiency allele was reported in the Swedish population (Eriksson, 1965). The polymorphic variants of olAT are categorized as belonging to the protease inhibitor or Pi system (Fagerhol & Laurell, 1967).

ln parallel with these studies examining serum cr 1AT, the pathophysiologic consequences of crlAT deficiency in the lung, i.e. emphysema, were becoming more explicable by other experimental observations. Firstly, the presence of proteolytic enzymes in blood leukocytes had been described by Opie in 1905. Then in 1964, Gross et al reported an animal model of emphysema. ln this model, lesions similar to human emphysema, were produced in the lungs of rats by instilling into their tracheas the powerful plant protease, papain. lntuitively, Eriksson (1965) had suggested "proteolytic ferments", possibly derived from a number of sources such as bacteria, leukocytes and macrophages, might be responsible for the lung destruction in cr,1AT deficiency. Subsequently, slAT was shown to inhibit a crude preparation of human neutrophil derived proteases and it was suggested that this was most likely the link between crl AT deficiency and

2 pulmonary emphysema (Kueppers & Bearn, 1966). Further evidence of the potential importance of elastase and its inhibitors in emphysema came in 1968 when Janoff and Scherer purified an elastase from neutrophil granules and demonstrated its ability to hydrolyze elastin. cl AT was observed to be the major inhibitor of neutrophil elastase (Ohlsson, 1971). Later, neutrophil elastase was shown in animal and ín vitro models to be capable of inducing emphysema (Mass el a\,1972; Lieberman ,1972; Janoff et al, 1977; Senior et al, 1977: Snider et al, 1984).

From these observations of cr,lAT deficiency and its association with emphysema, together with induction of experimental emphysema with elastase, was born the now popular protease-antiprotease theory of the pathogenesis of emphysema (Janoff, 1985). According to this theory, when the protective antiprotease screen of serum and tissues is overcome by an excess of protease, there exists a potential for lung damage and emphysema. The excess of protease may be due to an increase in inflammatory cells in the lung or a quantitative or qualitative reduction of antiprotease. This concept has given rise to the development of cr,1AT replacement therapy which is becoming an established treatment of alAT deficiency in the United States of America (Hubbard & Crystal, 1988; Crystal et al, 1989; Buist et al, 1989; Crystal, 1eeo).

Although not initially recognized, the association between ol AT deficiency and liver disease in both children and adults soon became apparent (Sharp et al, 1969; Berg & Eriksson, 1972). Unlike the lung disease however, the pathogenetic link between deficiency of crlAT and disease is not so readily apparent for the liver. The two major theories have been: firstly, that deficiency of al AT allows the liver to be damaged by proteases, particularly those from the gut (Udall et al, 1982); and secondly, thatthe hepatocytes are damaged by accumulated crlAT within the endoplasmic reticulum in olAT deficient patients (Carrell et al, 1982; Carrell, 1986). Only recently, utilizing advanced techniques of molecular biology in transgenic mice, has very convincing evidence for the second hypothesis been obtained (Dycaico et al, 1988; Carlson et al, 1989; Sifers et al, 1989a).

From the early 1980's to the present, many rapid advances in the understanding of ø1AT, its variation, and its association with disease as well

3. as preliminary experiments into the possibility of gene therapy for o1 AT deficiency have been made possible by the application of cell and molecular biology to this field. (Carrell et al, 1982; Carrell, 1986; Brantly et al, 1988a; Crystal et al, 1989; Cox, 1989; Sifers et al, 1989b; Sifers et al, 1989c; Crystal, 1990; Kalshekerand Morgan, 1990; Perlmutter, 1991). These will be highlighted in this literature review.

1.2. THE NORMAL alAT GENE

1.2.1. crlAT Gene Locus

The olAT gene locus is designated P¡ or protease inhibitor (Fagerhol & Laurell, 1967; Cox et al, 1980; Fagerhol & Cox, 1981). The chromosomal localization of the crlAT gene is to the long arm of chromosome 14 al q32.1 (Cox et al, 1990). A number of studies published over fifteen years led to elucidation of this position. Firstly, examination of two informative families showed linkage of the P¡ locus to the gene for the heavy chain of immunoglobulin designated Gm (Gedde-Dahl et al, 1972). Subsequently, the Gm locus was found to localize to chromosome 14, indirect but convincing evidence that the crl AT gene was also to be found on chromosome 14 (Croce et al, 1979). Three groups independently and virtually simultaneously confirmed the position of the P¡ locus on chromosome 14 using somatic cell hybrids between animal and human cells and analyzing co-segregation of human olAT gene expression and the presence of the human chromosome 14 (Darlington et al, 1982; Lai et al, 1983; Pearson et al, 1983). Utilizing two families with abnormalities of the long arm of chromosome 14, Cox et al, (1982a), were able to assign the P¡ locus to between chromosomal bands q24.3 and q32.1. Using the higher resolution ol in sifu hybridization, further refinement of the assignment of the P¡ locus lo 14q24.1-32.1 and later 14q31- 32.3 was possible (Schroeder et al, 1985; Rabin et al, 1986). Finally, localization of the P¡ locus to chromosome 14q32.1 was achieved by analysis of rodent/human somatic cell hybrids carrying translocations of the long arm of chromosome 14 (Purello et al, 1987). Assignment to chromosome 14q32.1 had been previously suggested by Turleau et al (1984) and Yamamoto et al (1986) in their independent studies of children with 14q deletions.

4. Apart from linkage to Gm which is centromeric to P¡ (Keyeux et al, 1989), the olAT gene is linked to a sequence related gene termed the crlAT-like gene (Lai et al, 1983; Schroeder et al, 1985; Kelsey et al, 1987; Kelsey et al, 1988; Hofker et al, 1988; Bao et al, 1988). This gene is approximately 12 kilobases (kb) downstream to the crl AT gene and by virtue of its sequence and absence of demonstrable expression in vivo is thought to be a non-functional pseudogene (Kelsey et al, 1988; Hofker et al, 1988). A number of restriction fragment length polymorphisms have been described for the crlAT-like gene (Hodgson & Kalsheker, 1986; Cox & Coulson, 1987).

Three other members of serine protease inhibitors family (serpins) so named because of the active site amino acid of their target enzyme and/or their closely related gene structure or sequence homology (Hunt & Dayhoff, 1980; Carrell et al, 1982; Carrell et al, 1989) are also found on chromosome 14 with the P¡ locus suggesting they likely arose by gene duplication from a common ancestor. These are alpha 1-antichymotrypsin (o1ACT), protein C inhibitor, and corticosteroid binding globulin (CBG) (Chandra et al, 1983; Rabin et al, 1986; Bao et al, 1987; Billingsley et al, 1989; Underhill & Hammond, 1989; Seralini et al, 1990; Sefton et al, 1990; Meijers & Chung, 1991).

1.2.2. olAT Gene Structure

The alAT gene is 12.2 kb long and is comprised of 7 exons and 6 intervening intron sequences (Figure 1.1). The first clAT gene to be cloned and sequenced was serendipitously found to be the S deficiency crlAT allele, the crlAT product of which was known to differ from the normal crlAT by a single amino acid substitution of valine for glutamine (Owen et al, 1976; Owen & Carrell, 1976; Leicht et al, 1982; Long et al, 1984).

Though the olAT gene was originally reported as having 4 exons and 3 introns (Leicht et al, 1982), and subsequently 5 exons and 4 introns (Long et al, 1984), current evidence suggests there are 7 exons and 6 introns (Perlino et al, 1987). The numbering system of the exons is le-c and ll- V, the 3' portion of exon lc being the original exon I in the report of Long et al (1984) and exons ln and le and the 5'50 nucleotides of exon lc being those discovered by Perlino et al (1987).

5 FIGURE 1.1

sl AT Gene

Schematic of the clAT gene showing 7 exons (ln-c; ll-V) and 6 introns. ln exon ll, the start codon (ATG) and signal peptide sequence area (shading) are indicated. ln exon V resides the stop codon (TAA) and polyadenylation signal (ATTAAA). The underlining indicates the original 5 exons described by Long et al (1984) and the areas sequenced in subsequent experiments outlined in this thesis. tr1-antitrypsin gene lkb

ln le lc tl ilt tvv

5', 3',

ATG signal ATTA/tuA (start) peptide (stop) (polyadenylation signal) When compared with the genes of other members of the serpin family, crlAT is most closely related to cr1ACT (Chandra et al, 1983; Bao et al, 1987; Sefton et al, 1990). The deduced amino-acid homology between these two proteins is 42%. The exon-intron organization and structure of crl ACT is very similar to cr1AT, a characteristic also shared by anglotensinogen (Gaillard et al, 1989), leuserpin 2 (Ragg & Preibisch, 1988), CBG (Underhill & Hammond, 1989), and Kallikrein-binding protein (Chai et al, 1gg1). lnterestingly, at least two other members of the serpin family, chicken ovalbumin and antithrombin lll, though sharing reasonable sequence homology with clAT (a1AT-antithrombin lll - 33%; crlAT-ovalbumin - 24"/"), have very different gene organization (Leicht et al, 1982; Chandra et al, 1983; Prochowniket al, 1985). This has been suggestedto be due to divergence by intron insertion after duplication (Leicht et al, 1983; Chandra et al, 1983; Long et al, 1984). Other members of the related serpin family in man include - Heparin cofactor ll, antiplasmin, plasminogen activator inhibitor, C1-inhibitor and thyroxine binding globulin (Carrell et al, 1989)'

Exons ln - c of the crl AT gene are not protein coding exons whereas exons ll-V code forthe crlAT protein (Long et al, 1984; Perlino et al, 1gB7). The translation start site (ATG) is in exon ll and the translation stop (TAA) and polyadenylation signals (ATTAAA) are in exon V (Long etal, 1984). Together with a TATA box within exon lc, a number of studies have demonstrated that the region 5'to exon lc contains crs-regulatory sequences which bind at least two frans-acting hepatic nuclear factors (HNF-1 and HNF-2) important in transcriptional regulation and cell specific expression of the ol AT gene (Ciliberto et al, 1985; Shen et al, 1987; De Simone et al, 1987; Courtois et al, 1987; Grayson et al, 1988; Hardon et al, 1988; Monaci et al, 1988; Li et al, 1988; Kugler et al, 1988; Costa et al, 1989; Sifers et al, 1989c; Shen et al, 1g8g). These hepatic cell-specific regulatory sequences have been mapped to the area between bases -37 and -137 of the transcription staft site in exon lc (De Simone et al, 1987; Li et al, 1988). 5'to this area are other positive regulatory elements which are not cell-specific (De Simone et al, 1987). Mitchelmore et al (1991) have found that two widely produced zinc finger DNA binding proteins (AT - BP1 and AT - BP2) bind to the area just 5' to exon 1c and although the physiologic role is uncertain, they are likely functioning as transcriptional regulators. lnterestingly, the area just 5' to exon lc which functions as a positive regulatory element in hepatocytes, may function as a

7 negative cr.sacting element in non-livercells (De Simone & Cortese, 1989). ln addition to these cis regulatory regions is a transcriptional enhancer element (Shen et al, 1987) and within exon la two binding sites for a transcriptional factor, the AP-1 c-jun proto-oncogene product, a gene upregulated by surface stimulation such as inflammation (Trapnell etal, 1989; Crystal, 1989). An Sp1 binding site is present 5'to exon ln as well as a potential TATA-like sequence (Perlino et al, 1987).

1.2.3 alAT Gene Expression

The position of the cell-specific transcriptional regulatory elements of the crlAT gene dictates the mRNA transcript species produced by the olAT gene. Three different mRNA species are transcribed from the c¡1AT gene (Perlino et al, 1987). ln the liver, a 1.4 kb mRNA is produced comprised of part of exon lc and all of exons ll - V (Long et al, 1984; Ciliberto et al, 1985; Perlino et al, 1987). ln monocytes, there are 2 major mRNA species, a 1.8 kb transcript containing all of exons lR -c, ll - V and a usually more abundant 1 .6 kb transcript with exon la spliced out (Perlino et al, 1987). Both monocyte transcripts may arise from one of two transcription start sites, 37 base pairs apart, 5'to exon ln (Perlino et al, 1987). Although the two ø1AT promotors are generally cell specific, i.e. the hepatocyte promotor is silent in monocytes and the monocyte promotor silent in hepatocytes (Ciliberto et al, 1985; Perlino et al, 1987), some recent evidence suggeststhatthis is not absolute. Trapnell et al (1989) have shown with an RNA nuclease protection assay that resting macrophages contain low levels of mRNA transcripts of "liver" type which are upregulated by surface activation with phorbol myristate acetate, an observation with interesting implications for crlAT production at sites of inflammation.

The major site of olAT gene expression is in the hepatocyte (Schultz & Heremans, 1966; Fagerhol & Cox, 1981). This has been demonstrated rn vivo after liver transplantion in olAT deficient patients (Sharp, 1971; Alper et al, 1980; Van Furth et al, 1986) and by examining human t¡ssue mRNA with a sensitive nuclease protection assay (Kelsey et al, 1987) as well as by creating transgenic animals which carry the human crlAT gene and studying human olAT gene expression ln these animals (Kelsey et al, 1987; Sifers et al, 1987; Carlson et al, 1988; Koopman et al, 1989; Shen

8. et al, 1989). Furthermore, in vitroanalysis of alAT production by short term hepatocyte cultures (Bhan et al, 1976) isolated perfused liver (Koj et al, 1978), foetal liver fragment cultures (Erikson et al, 1978), human hepatocyte cultures (Gautier et al, 1977) and hepatoma cell lines (Glasgow et al, 1982) are consistent with and complementary to the in vivo studies. Although monocytes have also been generally accepted as crlAT producers, albeit at a much lower level than the hepatocyte (lsaacson et al, 1981; Andersen, 1983; Van Furth et al, 1983; Rogers et al, 1983; Perlmutteret al, 1985a; Mornex et al, 1986) it has only more recently begun to be accepted that other tissues express the al AT gene and produce cr1AT. Convincing evidence has been published for low level crlAT mRNA production in brain (Dziegielewska et al, 1986), lung, kidney, spleen, small intestine, fetal liver, fetal intestine and yolk (Kelsey et al, 1987), pancreas (Carlson et al, 1988), intestinal epithelium (Perlmutter et al, 1989a) and neutrophils (du Bois et al, 1991). The studies of human crlAT gene expression or oncogene expression driven by the olAT promotor in transgenic animals have confirmed that expression occurs in most of these tissues highlighting that the olAT gene has all the elements required for tissue specific expression (Kelsey et al, 1987; Sifers et al, 1987; Carlson et al, 1988; Koopman et al, 1989; Shen et al, 1989; Butel et al, 1990). ln transgenic animals, crlAT was also observed in the adrenal, Sertoli cells of the testis, bronchi and sebaceous glands in the skin (Carlson et al, 1988) as well as salivary glands (Koopman et al, 1989). That olAT gene expression occurs in tissues other than the liver is not unexpected in light of the recent demonstration of expression of the gene coding for the nuclear factor HNF-1 in not only the liver, but also in the kidney, intestine and spleen, indicating that HNF-1 is a more broadly acting transcription factor than previously thought (Courtois et al, 1987; Courtois et al, 1988; Baumhueter et al, 1990). HNF-2 has also been isolated (Rangan et al, 1990) and studies of HNF-2 gene expression are awaited. HNF-2 appears to be an important transcription factor forglAT in light of the observation that mutations in the HNF-2 binding site of the glAT gene promoter reduces ø1AT production dramatically (Tripodi et al, 1991 ; Bulla et al, 1992).

Further detail of the control of clAT gene expression is not extensive, however, a number of recent findings requiring fufther elucidation are of interest. The natural target of cl1AT, neutrophil elastase (NE) (Bieth, 1986) has been shown to upregulate olAT gene expression in monocytes

I (Perlmutter et al, 1988). ln addition, a receptor (serpin - enzyme complex: SEC) which recognizes a comformation specific pentapeptide sequence of alAT in crlAT-NE complexes which is highly conserved among serpins had been demonstrated. This receptor activates a signal transduction pathway lor upregulation of olAT gene expression in human hepatoma cells and human monocytes (Perlmutter et al, 1990a; Perlmutter et al, 1990b; Joslin et al, 1991). lnterest¡ngly, pseudomonas elastase also changes the conformation of al AT to expose the pentapeptide sequence and activate the SEC receptor (Barbey-Morel & Perlmutter, 1991). These observations suggest an active feedback control of alAT gene expression to maintain protease-antiprotease balance in vivo. Finally, Kalsheker and Swanson (1990) have observed that w1h lnterleukin-6 stimulation of the U937 myelomonocytic cell line, only the 1.6 kb crlAT mRNA species is transcribed with exclusion of the 1.8 kb transcript. This suggests the possibility of a novel way of modifying crlAT gene expression during inflammation.

1.3 THE NORMAL cxlAT PROTEIN

1.3.1 crlAT Protein Structure

The cr,1AT mRNA transcript codes for a single chain 418 amino acid polypeptide which includes a 24 amino acid signal peptide subsequently removed by post-translational modification resulting in the mature 394 amino acid slAT protein (Carrell et al, 1982; Bollen et al, 1983; Long et al, 1984). The extracellular c¡1AT has three complex carbohydrate side chains N-linked to asparagine residues 46, 83 and 147 (Mega et al, ,1980a; Mega et al, 1980b; Carrell et al, 1981; Carrell et al, 1982; Long et al, 1984) and these comprise 12/" ol the weight of the 52 kilodalton protein (Laurell & Jeppsson, 1975; Travis & Salvesen, 1983). The carbohydrate side chains may be biantennary or triantennary depending on the number of antennae ending in N-acetylneuramic acid arising from the terminal mannose (Chan & Rees, 1976; Mega et al, 1980b). Because of the minor charge difference between these carbohydrates the pattern of distribution of the carbohydrate side chains linked to the three asparagine residues accounts for some of the microheterogeneity of olAT revealed by electrophoresis of serum

10 (Fagerhol & Braend, 1965; Fagerhol & Laurell, 1967; Laurell & Persson, 1973; Laurell & Jeppsson, 1975; Vaughan et al, 1982). The rest of the microheterogeneity is due to limited proteolytic cleavage of the al AT molecule (Hercz, 1985; Jeppsson et al, 1985). There are seven electrophoretic species of cr,1AT, five of these account for 9O/" of the al AT (Laurell & Jeppsson, 1975). The 1ve main bands are further subcategorized as two major (referred to as 4 and 6) and three minor bands (2,7 and 8) (See Chapter 7, Figure 7.1). The major 4 band has two biantennary carbohydrate side chains and one triantennary side chain whereas the major 6 band has three biantennary side chains. The more anodal, minor 2 band has two triantennary side chains and one biantennary side chain. The 7 and 8 bands are identical to the 4 and 6 bands, respectively, except they do not contain the first five amino acids of the mature olAT protein, thought to be caused by proteolytic cleavage by unknown mechanisms (Hercz, 1985; Jeppsson, 1985). These N-terminal five amino acids include the sequence GlulAsp2ProsGlnaGlys, two negatively charged and three neutral amino acids, thus explaining the cathodal position of these minor qlAT bands relative to the major 4 and 6 bands (the ø1AT molecules missing these aminoacids are more positively charged than the full- length alAT).

The three dimensional or tertiary structure of crlAT has been elucidated utilizing crystallography and an olAT cleaved at the methionine 3$8-serine 359 bond by chymotrypsin (Loebermann et al, 1982; Loebermann et al, 1984). lt is avery well ordered globularstructure with 80% of the amino acids in eight cr helices (A-H) and three p-pleated sheets (A-C) (Loebermann et al, 1984; Huber & Carrell, 1989). The three carbohydrate attachment sites are on turns of the molecule and protrude from its surface (Loebermann et al, 1984; Smith et al, 1990). lmportantly, in considering alAT variants (see later) there are salt bridges between Glu264 and Lys387' [5p256 and His231, and between Glu342 and Lys2eo (Huber & Carrell, 1989). The active site of alAT centred around Met358 (Carrell et al, 1982: Boswell et al, 1gB3) is on an external bend of the molecule and is under tension. This stressed loop is believed to fit perfectly into the active site pocket of neutrophil elastase. Neutrophil elastase , a 220 amino acid glycoprotein with a molecular weight ol 29 kilodaltons encoded by the 4 kb neutrophil elastase gene on chromosome 11, has the characteristic catalytic triad of serine proteases conveying the proteolytic capacity (Sinha et al, 1987; Takahashi et al, 1988a;

11 Crystal, 1989). The catalytic triad is found in the reactive site specificity pocket of the enzyme and is comprised of Hisa1, Asp88 and Se117s lBode et al, 1986). Cleavage of the Met 358-Ser 359 bond by Ss¡17s of elastase causes conformational change in alAT to a more stable unstressed form (Loebermann etal, 1984; Carrell &Owen, 1985; Carrell, 1986). lt is possible that the stressed uncleaved crlAT has fewer hydrogen bonds in its secondary structure and that cleavage allows development of a fully hydrogen bonded stable glAT (Harris et al, 1990). The extent to which this proteolysis of olAT by NE occurs in vivo is uncertain though modified or cleaved crlAT can be found in the bronchoalveolar lavage of human smokers in abundance (Stockley & Afford, 1984). Recently it has been found that the exposed loop of ovalbumin is in fact an a-helix and it has been suggested that this is also likely to be the case for crl AT (Stein et al, 1990).

1.3.2 crlAT Biosynthesis

As indicated previously, the major site of serum o1 AT production is in the liver. Contribution to the ø1AT serum levels from other sources is negligible (Krivit et al, 1988). Synthesis of crlAT in the liver is believed to follow the pathways for glycoproteins initially elucidated studying viral glycoproteins (Hercz & Harpaz, 1980; Kelly, 1985). Thus, assembly of the polypeptide back bone of cr,1AT occurs in the ribosomes attached to the rough endoplasmic reticulum (RER). The peptide then has the 24 amino acid signal peptide removed as it moves to the cisternae of the RER where high mannose carbohydrates are added and the alAT folds into its three dimensional form. lt is then transported to the Golgi apparatus in RER - to Golgi transport vesicles where the carbohydrate side chains are modified to form complex carbohydrate prior to protein secretion (Lodish & Kong, 1984; Hirschberg & Snider,1987; Elbein, 1987; Pfeffer & Rothman, 1987; Lodish, lgBB). Two studies of human crlAT biosynthesis, one of normal crlAT in transgenic mice and one of mutant olAT in COS I monkey kidney cells, have raised the possibility of receptor mediated transport from the RER which if proved true will change present concepts of intracellular protein biosynthetic pathways (Lodish, 1988; Sifers et al, 1989a; Sifers et al, 1989b; Mcoracken et al, 19Bg). Brodbeck & Brown (1992) studying truncated crlAT in COS I cells have concluded that at least 391 of the 394 amino acids of cx,1AT are required for efficient exist from the RER to Golgi and that residue 391 is especially important for this process.

12. Normal serum levels of olAT are 150-350 mg/dl using a commercial standard and 20-53 micromolar (pM) using a true laboratory standard (Wewers et al, 1987a; Brantly et al, 1991). The commonly used commercial standards overestimate al AT levels by 40% (Jeppsson et al, 19ZBa; Carrell & Owen, 1979). The small alAT is able to diffuse through most tissues though at lower levels than in the serum (Fagerhol & Cox, 1981; Gadek & Crystal, 1982). ln the epithelial lining fluid of the lung sampled by bronchoalveolar lavage, presumably a critical site for ø1AT action, al AT is present at a level of 2 to 5 pM (Olsen et al, 1975; Gadek et al, 1981a; Wewers et al, 1987a; Wewers et al, 1987b).

Approximately 34 mg/kg body weight of olAT is produced by the liver each day and released into the circulation (Jones et al, 1978). The plasma half-life of crlAT is 4-6 days (Kueppers & Fallat, 1969; Makino & Reed, 1970; Laurell et al, 1977). The half-life is reduced significantly with loss of carbohydrate side chains (Glaser et al, 1977; Yu & Gan, 1977; Satoh et al, 19Bg) as is protein stability (Guzdek et al, 1990). The catabolism of crlAT appears to be predominately in the liver and be largely dictated by desialylation of crlAT (Laurell et al, 1977; Glaser et al, 1977; Jones et al, 1978; Jeppsson et al, 1978b; Brantly et al, 1988a)'

cr 1AT is an acute phase reactant and therefore its biosynthesis is affected by inflammation and other insults causing an acute phase response (Fagerhol & Cox, 1981). This response is mediated at least in part by various cytokines (Takemura et al, 1986; Mier et al, 1987; Perlmutter et al, 1989b; Mackiewicz el al,199O). lt is also likely that upregulation of ol AT biosynthesis is mediated by receptor recognition of olAT-NE complexes by hepatocytes (Travis et al, 1988; Perlmutter et al, 1990a). ln addition Barbey- Morel et al (1987) have shown that the bacterial lipopolysaccharide is able to increase crlAT production by increasing efficiency of translation. crlAT also increases during pregnancy and with oestrogen therapy (Ganrot & Bjerre, 1967; Laurell et al, 1967).

1.3.3 alAT Function

al AT has a broad spectrum of protease inhibition being able to inactivate most serine proteases (Carrell et al, 1982; Travis &

13. Salvesen, 1983). Measuring the kinetics of association of ø1AT with serine proteases, Beatty et al (1980) demonstrated most avid interact¡on with NE followed in order by chymotrypsin, cathepsin G, anionic trypsin, plasmin and thrombin. The association rate constant of ü1AT with NE (6.5 + 4.0 x 102 M-1S-1) was more than ten times greaterthan its association with any other serine protease. These observations together with the studies described previously (1.1) provide further evidence that the major function of alAT is to inhibit neutrophil elastase and its proteolytic attack of the connective tissue matrix, particularly in the lung. NE is capable of degrading most components of the connective tissue matrix, including elastin, collagens type l, lll, and lV, proteoglycans and laminin (Bieth, 1986). Furthermore, it has been demonstrated that crl AT provides greater than 85% of the anti-elastase protection of the lower respiratory tract (Gadek et al, 1981a; Wewers et al, 1gg7b). A constant presence of olAT is required in the lower respiratory tract to protect the alveoli from the chronic low burden of neutrophil elastase released by the neutrophils present in the lung even under normal circumstances (Gadek et al, 1981a; Hunninghake & Crystal, 1983; Janoff, lgBS) Some studies have strongly suggested that other antiproteases found in sputum and bronchoalveolar lavage, particularly low molecular weight bronchial proteinase inhibitor or secretory leukoprotease inhibitor (SLPI), play a significant role in the antiprotease screen of the lower respiratory tract (Stockley et al, 1984; Morrison, 1987). The physiologic significance of the ç¿lAT produced in tissues other than the liver is unknown and speculative though it is likely that it plays an important anti-elastase role in the local environment into which it is secreted (Crystal et al, 1989).

The structural interaction between crlAT and NE is a one- to-one molar ratio competitive reaction and is virtually irreversible under physiologic conditions (Johnson & Travis, 1976; Beatty et al, 1980; Travis & Salvesen, 1983). The dissociation constant (Ki) is 10-14 for this reaction (Beatty et al, 1984). Cleavage of slAT at the Metsss-$s¡Sse bond releases the 36 amino acid segment from $s¡35e to Lys3e4 (Carrell et al, 1981) and inactivates crlAT (Travis & Salvesen, 1983). ø1AT has only one reactive centre based around Met358 (Johnson & Travis, 1978; Carrell et al, 1981). The presence of methionine at the reactive site has important implications for the function of alAT (Johnson & Travis, 1978; Johnson & Travis, 1979; Beatty et al, 1980). The reactive site methionine is susceptible to oxidation to

14. methionine sulfoxide with resultant dramatic reduction in the association rate constant of ol AT with NE by 2OOO, a reduction in the Ki from 10-14 to 10-10 and a loss of the ability of crlAT to protect elastin from degradation by NE (Beatty et al, 1980; Beatty et al, 1984). The physiologic importance of this has been studied in smokers and a number of these studies have demonstrated that the oxidants in cigarette smoke can inactivate the alAT in the lower respiratory tract causing a functional crlAT deficiency (Janoff et al, 1979; Gadek et al, 1g7g; Carp et al, 1982). Although these data have been challenged (Stone et al, 1979; Boudier et al, 1983; Wyss et al, 1984; Abboud et al, 1985) the most recent data confirms that cigarette smoking is associated with functional deficiency of cr,lAT in the lower respiratory tract (Ogushi et al, 1991). Similarly, data suggesting the detectable presence of oxidized olAT in the serum of smokers has also been challenged (Beatty et al, 1982; Cox & Billingsley, 1gB4). The study of Ogushi et al (1991) was also unable to detect non- functional crlAT in the serum above that of normal. lnterestingly, erythrocytes which possess anti-oxidant properties have been shown to prevent cr1 AT inactivation by cigarette smoke (Mangione et al, 1991). The physiologic significance of this is uncertain. Another recent observation which requires verification is that of Nowak & Ruta (1990) who have found that nicotine inhibits inactivation of crl AT by oxidants suggest¡ng that low nicotine cigarettes may be potentially more deleterious to lung function than those with a higher nicotine content. Neutrophils, which are increased in the lungs of smokers, are also a potent source of oxidants and oxidant producing enzymes which can inactivate glAT (Matheson et al, 1979; Carp & Janoff, 1979; Carp & Janoff, 1980; Matheson et al, 1981 ; Clark et al, 1981; Hunninghake & Crystal, 1983). Alveolar macrophages in smokers are also able to release oxidants and inactivate alAT (Hubbard et al, 1987). Why a methionine atthe reactive site has evolved rather than an oxidation resistant amino acid such as valine is unknown. The need for exclusion of active crlAT at sites of tissue remodelling after injury and from areas undergoing liquefaction prior to phagocytosis such as in an abscess have been forwarded as possible advantages in having an oxidation sensitive reactive site (Carrell et al, 1982; Travis & Salvesen, 1983; Ossanna et al, 1986). The potential for oxidative inactivation of native alAT has prompted investigators to synthesize oxidation resistant mutant forms of alAT by recombinant techniques for possible therapeutic use (Rosenberg et al, 1984; George et al, 1984; Travis et al, 1985; Matheson et al, 1986; Jallat et al, 1986; Janoff et al, 1986). lnterestingly, oxidized alAT remains a potent

15. and rapid inhibitor of human pancreatic elastase which may be physiologically important in pancreatitis (Padrines et al, 1992)'

GlAT also appears to play a role in neutrophil chemotaxis. (Goetzl et al, 1975; Hakansson & Venge, 1983; Breit et al, 1983; Baran et al, 1989; Stockley et al, 1990). The native crlAT inhibits neutroph¡l migration whereas crlAT complexed with NE, the olAT cleavage product from crlAT NE interactions, the crlAT cleavage product from macrophage elastase-cr1AT interactions and, oxidized c1AT, all appear to increase neutrophil chemotaxis which has important implications at the site of inflammation (Banda et al, 1988a; Banda et al, 1988b; Stockley et al, 1990). Other physiologic effects of o¿lAT are less clear. There is however some evidence that slAT may play a role in inflammatory and immune function either directly or in its effects on NE (Fagerhol & Cox, 1981; Breit et al, 1985; Cox, 1989). Proteases, and therefore by inference cr1AT, may have an effect on immunity and inflammation directly or by their effects on activation of complement, macrophages and B- lymphocytes as well as by their effects on coagulation, fibrinolysis and kinin activation (Breit et al, 1985). olAT also has been reported to have inhibitory effects on T-cell cytotoxicity and may enhance T-helper cell function all of which may help to explain some of the reported associations of al AT deficiency with diseases other than those of the lung and liver (Breit et al, 1s85).

1.4 cxl AT VARIATION

The polymorphic nature of the cllAT (Pi¡ gene locus, and the fact that each crlAT allele contr¡butes equally to the alAT phenotype in a codominant fashion, was apparent from the early studies of crlAT (Laurell & Eriksson, 1963; Eriksson, 1964; Kueppers et al, 1964; Eriksson, 1965; Fagerhol & Braend, 1965; Axelsson & Laurell, 1965; Fagerhol & Laurell, 1967; Fagerhol, 1969; Fagerhol & Gedde-Dahl, 1969). The extent of the alAT gene polymorphism, is evidenced by the large number of crlAT variants reported in the literature. ln a 1988 review, Brantly et al (1988a) counted at least 75 known variants of cx,1AT, and since that time more than 15 new alleles have been described (Sifers et al, 1988; Weber & Weidinger, 1988; Curiel et al,

16 1989b; Graham et al, 1989; Whitehouse et al, 1989; Fraizer et al, 1989; Fraizer et al, 1990; Graham et al, 1990a; Takahashi & Crystal, 1990; Matsunaga et al, 1990; Curiel et al, 1990; Holmes et al, 1990a; Holmes et al, 1990b; Holmesetal, 1990c; Seyamaetal, 1991; Polleretal, 1991)- Atthe gene level, the majority of ol AT variants are due to nucleotide substitutions, deletions or insertions causing amino acid substitutions, deletions or a premature translation stop codon (Brantly et al, 1988a; Crystal, 1990; Kalsheker & Morgan, 1990). ln two cases (Null¡sq¡¿ di procida and Nullr¡edenbers) there is a deletion of a major portion of the ø1AT gene (Takahashi & Crystal, 1990; Poller et al, 1991). The heterogeneity of the ø1AT gene and protein necessitates sensitive and convenient methods for differentiating one variant from another and also a reliable classification system.

1.4.1 Classificat¡on of q,l AT Variants

The genetic variation of alAT may be manifested physiologically at the level of cr,1AT protein quantity, quality or function and methods for detecting these changes as well as for detecting the basic nucleotide mutation are used to characterize a given new variant (Cox et al, 1980; Brantly et al, 1988a; Cox, 1989; Crystal, 1990). The allelic variants of crlAT can be classified, according to the level of crlAT protein in the serum attributable to that allele, into "Normal" and "Deficient" (including "Null") crl AT variants (Cox et al, 1980; Fagerhol & Cox, 1981). Within these categories the variants can be further subcategorized according to the position of the crl AT variant relative to other standard crlAT variants after gel electrophoresis (Cox et al, 1980; Fagerhol & Cox, 1981). Although historically the subcategorization was reliant on acid-starch gel electrophoresis, it is now dependent on the relative mobility of the cllAT variant by isoelectric focusing (lEF) in polyacrylamide gels (Cox et al, 1980; Fagerhol & Cox, 1981). The variants are given a letter designation from Alo Z with the most anodal var¡ant being A and the most cathodal being Z. Additionally, if there is more than one variant at or near a given isoelectric point, the rare variants (allele frequency < O.O1)are named according to the birthplace of the oldest living family member with that allele (e.g. Mr"non) and the more common variant alAT proteins give a number designation (e.g. M1 , M2) (Cox et al, 1980). The alleles that do not direct synthesis of any detectable crlAT give rise to a "Null" phenotype. An individual's phenotype is therefore determined by both of their ol AT alleles

17. and is written as'M1M2, BM1, SZ, SNull (S-), etc. The corresponding alleles are written as P¡M1, PiM2, and so forth, except for the Null alleles which are designated PIQOHong Kong or PiQOludwigshafen etc, the QO electrophoretic position being arbitrarily assigned to the Null variants (Cox et al, 1980). Otten, for convenience, alleles will be referred to as "the M2 allele or the Null¡¿¡¿'¡s allele (Brantly et al, 1988a; Crystal, 1990). Another subcategory of crlAT based on crl AT function is the "Dysfunctional" variants which have altered clAT function (Brantly, 1988a; Crystal et al, 1989; Crystal, 1990). ln addition to determining ø1AT protein levels and electrophoretic properties against known standard, it is recommended that for a new crlAT variant the crlAT protein be confirmed as crlAT by immunofixation techniques and that a family study be performed to ensure that the variant is inherited (Cox et al, 1980; Fagerhol & Cox, 1981). Analysis of the genetic basis of an crlAT variant provides the most definitive and unambiguous means for classifying crl AT variation (Carrell & Owen, 1979). This approach does not however presently lend itself easily to the routine clinical situation (Kidd et al, 1983; Nukiwa et al, 1986a).

1.4.2 Detection of olAT Variation

The measurement of crlAT serum levels is the first and most clinically important step in the detection of crl AT variation (Brantly et al, 1gg1). This is usually followed by electrophoretic analysis in the case of deficiency of crl AT or for a suspected new variant (Cox et al, 1980; Fagerhol & Cox, 1gB1). An antibody specific for the most common deficiency olAT variant, Z, is available and although proposed as useful for screening studies, it has not been utilized extensively (Wallmark et al, 1984). For characterizing a new variant in the research setting or in special situations (e.9. pre-natal diagnosis; investigating the possibility of a Null allele in a patient with crl AT deficiency), assessment of the olAT gene mutation is more otten undertaken (Brantly et al, 1988a; Cox, 1989; Crystal et al, 1989; Crystal, 1990)'

1.4.2.1 al AT Serum Levels

There are two basic approaches to measuring alAT Serum levels, firstly, by measuring the biological function of crlAT or secondly, by determining immunological levels of crlAT (Laurell & Jeppsson, 1975; Kueppers , 1978; Fagerhol & Cox, 1981 ; Arnaud & Chapuis-Cellier, 1988; Cox, 1989). 18. Functional assays are based on the ability of the serum to inhibit either trypsin, pancreatic elastase or more recently, neutrophil elastase from digesting either their natural substrates or synthetic substrates (Eriksson, 1965; Senior et al, 1971; Billingsley & Cox, 1980; Ogushi et al, 1987; Arnaud & Chapuis-Cellier, 1988; Cox, 1989; Lloyd & Travis, 1989). Although functional measurements of al AT correlate well with immunochemical determination of alAT levels, the latter are recommended for routine measurements and are most often utilized (Talamo et al, 1972; Billingsley & Cox, 1980; Cox et al, 1980, Brantly et al, 1991).

lmmunochemical methods which are reliant on an antigen- antibody react¡on between alAT and a specific anti-a1AT antibody include immunoassay, and nephelometry (Mancini et al, 1965; Laurell, 1966; Ritchie, 1967; Rennard et al, 1980).

The problems with overestimation of alAT levels using commercial standards has been discussed previously (1.3.2) and are overcome by using a healthy donor pooled sample or a highly purified laboratory standard (Jeppsson et al, 1978b; Wewers et al, 1987a; Brantly et al, 1991). Using a highly purified laboratory standard and nephelometry, Brantly et al (1991) have recently reported ranges forthe common normal and deficient a1-antitrypsin phenotypes which are now being used for the crlAT Deficiency Registry in the United States of America.

1.4.2.2 Electrophoretic Variation of al AT

Although the original description of alAT variation was by Laurell & Eriksson (1963) using paper electrophoresis, it was the acid starch electrophoresis method which was used most extensively to determine crl AT variation (Fagerhol & Braend, 1965; Fagerhol, 1967). Acid-starch electrophoresis and another technique, agarose gel electrophoresis, have in the most part been replaced by the more sensitive technique of IEF of serum in polyacrylamide gels, though the former two methods remain useful for distinguishing some variants not easily resolved by lEF alone (Arnaud et al, 1974; Allen et al, 1974: Cox et al, 1980; Fagerhol & Cox, 1981; Frants et al, 1gB1). A number of satisfactory methods for IEF have been published (Constans et al, 1980; Jeppsson & Franzén, 1982; Weidinger et al, 1985;

19. Arnaud & Chapuis-Cellìer, 19SS). All techniques rely upon the different isoelectric points, i.e. charge differences, of the ol AT protein variants to distinguish between the variants (Allen et al, 1974; Charlionet et al, 1979). A resolving power of at least O.OO5 pH units has been reported for the basic IEF technique (Charlionet et al, 1979). lncreased resolution has been obtained by additional amphoteric substances in the lEF, and importantly, by utilizing immobilized pH gradients (Frants et al, 1978; Klasen & Rigutti , 1982; Bjellqvist et al, 1982; Gorg et al, 1983; Weidinger & Cleve, 1984; Gorg et al, 1gg5; Righetti et al, 1988). These lattertechniques are particularly useful for differentiating variants which focus close together on convential IEF (Gorg et al, 1983; Weidinger & Cleve, 1984; Gorg et al, 1985). Confirmation of the detectable protein bands in electrophoretic gels is undertaken by immunofixation (Johnson, 1976; Arnaud et al, 1977; Cox et al, 1980)' The sensitivity of IEF in polyacrylamide gels has allowed prenatal diagnosis of crl AT deficiency variants from foetal blood obtained at fetoscopy (Jeppsson et al, 1979; Jeppsson et al, 1981). ln addition, it is possible to use dried blood spots for IEF which may be useful in neonatal screening (Carracedo & Concheiro,l982i Jeppsson & Sveger, 1984). Although IEF is a very sensitive technique for determining alAT variation, it is important to recognize that the IEF pattern may be changed in certain inflammatory states, in neonates, during pregnancy, and with some drugs (Hug et al, 1982a; Vaughan et al, 1982: Jeppsson et al, 1985; Attenburrow, 1985; Whitehouse et al, 1989).

1.4.2.3 Analysis of the al AT Gene Mutations

To date all functionally significant alAT gene mutations have been confined to the protein coding exons facilitating analysis of these nucleotide variations (Brantly et al, 1988a; Crystal, 1990; Kalsheker & Morgan, 1g9O). The "gold standard" for genetic analysis remains sequencing the glAT gene (Long et al, 1984). This approach is being increasingly utilized to fully characterize o-lAT variants and has been made much more simple and rapid with the application of polymerase chain reaction (PCR) techniques (Saiki et al, 1985; Gyllensten & Erlich, 1988; Brantly et al, 1988a; Crystal et al, 1989; Crystal, 1990).

With knowledge of the gene sequence differences of various alleles it is possible to synthesize short radiolabelled oligonucleotides with the

20 known mutation at the centre which, under the correct conditions, will differentially bind to the DNA to which they have complete complementarity (Kidd et al, 1983; Nukiwa et al, 1986a; Klasen et al, 1987). The binding of the labelled oligonucleotide to membrane immobilized DNA can then be visualized by autoradiography. This method has been recently refined using PCR to amplify the segment of the olAT gene in which the mutation in question resides, allowing more rapid analysis (Peterson et al, 1988; Dermer & Johnson, 1988). The latter approach is also much more sensitive, and allows the use of biotin-labelled oligonucleotides rather than using radiolabel (Gregersen et al, 1989). Another simple, non-radioactive method to detect point mutations is also reliant on PCR. PCR primers can be designed with the mutational difference at their 3' end and these primers will amplify only genomic DNA to which they are exactly complementary at the 3' end (okayama et al, 1989a; Newton et al, 1989; Ehlen & Dubeau, 1989).

A number of other methods for genetic analysis which are not reliant upon knowledge of the gene sequence have been applied to the study of crlAT variants. These include restriction fragment length polymorphisms (RFLP's), ribonuclease cleavage and denaturing gradient gel electrophoresis (DGGE) (Cox, 1987; Abe et al, 1989; Dubel et al, 1989; Johnson et al, 1991). None of these methods are able to exactly define the gene mutation but they are useful for screening purposes.

RFLP's will identify an olAT mutation if that mutation changes a restriction enzyme recognition site, thus changing the length of DNA segments created by that particular enzyme, which can then be detected by agarose gel electrophoresis, Southern transfer and probing with a specific radiolabelled alAT gene probe (Southern,1975: Cox, 1987; Crystal et al, 1989). Only a small number of cr,1AT mutations change restriction enzyme recognition sites and these include (see also 1.4.4 and 1.4.5) - a BstEll and an Maelll polymorphism in exon tll distinguishing M1(Valzts) from M1(Alazts¡' an Rsal polymorphism in exon ll distinguishing M2 from M1; a Pvull polymorphism in exon ll for the Mprocida allele and an Ava ll polymorphism in exon ll associated with the Mminerat sprinss allele (Cox & Billingsley, 1986; Nukiwa et al, 1987a; Nukiwa et al, 1988; Takahashi et al, 1988b; Curiel et al, 1990). Although the relatively small number of mutation specific RFLP's together with the cumbersome nature of Southern analysis, makes this approach of limited

21. clinical use, combining newer PCR technologies with RFLP's may increase theirutility(Mullis&Faloona,1987; Abbottetal, 1988; Schwartzetal, 1989). To increase the range of ø1AT mutations detectable by RFLP's it is possible to create a mutation specific restriction site by using modified PCR primers and then analyse the digestion products simply on an agarose gel (Mullis & Faloona, 1987; Dty, 1991). After restriction enzyme digestion of DNA, Southern analysis and hybridization with an cl AT gene probe, ceftain RFLP's, though not caused by the specific olAT mutation in question, are linked to that mutation and have been useful in the past in diagnosis of olAT variants (Hodgson & Kalsheker, 1986; Cox, 1987; Cox, 1989), but with the newer techniques described, their utility is much less. lnterestingly, polymorphisms with the enzyme Taql may be linked to an increased propensity to chronic airways disease without overt association with crlAT deficiency (Kalsheker et al, 1987; Poller et al, 1990a).

Ribonuclease cleavage and DGGE are particularly suited for screening DNAs for known mutation or for detecting the general gene localization of new mutations prior to sequence analysis (Caskey, 1987: Gibbs & Gaskey, 1987; Abe et al, 1989; Dubel et al, 1989). DGGE, RFLP's and oligonucleotide probe analyses have all been utilized for prenatal diagnosis of cllAT deficiency variants (Kidd et al, 1984; Cox & Billingsley, 1986; Hejtmanciket al, 1986; Cox & Mansifled,1987; Abbott et al, 1988; Dubel et al, 1989; Dubel et al, 1991). The powerof newer PCR basedtechnologies is illustrated by the ability to type the al AT gene in an oocyte prior to in vitro fertilization and implantation (Verlinsky et al, 1990).

1.4.3 Normal crlAT Variants

The normal Gl AT variants are characterized by normal crlAT serum levels and, in those variants tested, normal crlAT function as an inhibitor of elastase (Billingsley & Cox, 1980; Billingsley & Cox, 1982; Brantly et al, 1988a; Cox, 1989). A large number of population studies have shown that, although there are minor differences between populations, the most common subtype is M1 , followed by M2, then M3 and that together, these variants account for more than 85% of alAT variants in all populations and over 95% in many (Fagerhol & Cox, 1981). The most recent studies from Sweden, South East Asia and in the Jewish population in lsrael are

22. comparable to previous studies (Hjalmarsson, 1988; Saha, 1990; Nevo & Cleve, 1991). An Australian study of alAT subtypes in umbilical cord bloods showed the common normal allelicfrequencies as:M1 (0.73); M2 (0.14) and M3 (O.Og) (Mulley, 1982). Clark (1982) found similar allelic frequencies in white Australian blood donors although the M3 frequency (0.03) was lower. ln full-blood Australian aborigines, the M2 allelic frequency is quite high (0.55 - 0.68) and M1 proportionately lower (0.28 - 0.38) (Glark, 1982). ln United States Caucasians, the allelic frequencies of the common normal alleles are : M1 (0.640 - 0.724); M2 (0.137 - 0.190); M3 (0.095 - 0.110) (Kueppers & Christopherson, 1978; Dykes et al, 1984). ln the English population these allelic frequencies have been reported as : M1 (0.746): M2 (0.118); M3 (0.061) (Arnaud et al, 1979).

Another relatively frequent normal M subtype is cll AT M4 which has an allelic frequency in Europeans of 0.015 - 0.048 (Constans et al, 1980; Weidinger et al, 1982; Klasen et al, 1982). The only other normal olAT variant reported relatively frequently, particularly in some populations, is the F variant (Fagerhol & Laurell, 1967; Kellermann & Walter, 1970)'

The gene sequence of all of the above normal crl AT variants is known (see Table 1.1). lnterestingly, molecular analysis of the M1 gene showed that the M1 phenotype can be coded by two variant olAT genes (Nukiwa et al, 1987a). The more common allele M1(Valzt3) accounting for 68% of M1 alleles codes for Valine at Codon 213, whereas the M1(Alazts¡ allele accounting for 32/" ol the M1 alleles codes for Alanine at Codon 213 (Nukiwa et al, 1987a). Comparison of the human crlAT genes with primates suggests that the M1(Al¿zts) is the oldest in evolutionary terms and that a single nucleotide substitution GQ.G ))))> GIG in Codon 213of the crlAT gene created the M1(Valzts¡ allele (Nukiwa et al, 1987a).

The many other reported normal al AT variants are tare (allelic frequencies < 0.001) and the molecular basis of only a few (including those reported in this thesis) have been elucidated (see Table 1.1) (Brantly et al, 1988a; Crystal et al, 1989; Cox, 1989). Those rare normal variants which are generally believed to be separate variants or which have been sequenced are :B (Martin et al, 1973); Bathambra (Yoshida et al, 1979); Bsaskatoon (Horne et al, 1982); C and D (Robinet-Levy & Reunier,1972); E (Fagerhol,1972);

23 TABLE 1.1

Normal cxl AT Allelest

ALLELE ALLELIC MUTATION REFERENCE BACKGROUND2

Nukiwa el al, 1987a. M1 1Rta213¡

al, 1987a. M1 ffat213¡ M1 (eta213¡ ¡¡¿2136ç6, VatGIG Nukiwa et

1989a; M3 M1 ryat213¡ 6¡1376ç44 n Asp GAg Curiel et al, Graham et a|,1990b.

M2 M3 Arg1o1 CGT ' His CAT Nukiwa et al, 1988. M4 Ml (Vat213¡ Arg101 CGT ' His CAT okayama et al, 1989b 223çet, al, 1991. F M1 1Vat213¡ ¡rg cyslGT Okayama et

Holmes et al, 1990c. Psaint albans M1 1Vat213¡ Asp341 GAC " Asn 4lC and Asp256 G{f " Asp GAQ

et al, 1990a. Vmunich tvtt 1Val213¡ tsp2 oAt , Ala GQT Holmes

& Laurell, X M1 1vat213¡ 6¡u204 6¡6 , Lys AAG3 Jeppsson 1 988.

Xchristchurc¡ Unknown 6¡u363 gAc , Lys AAG3 Brennan & Canell, 1986.

Those alleles with identified molecular basis.

2. The normal crlAT allele on which the mutation occurred.

3. Deduced nucleotide substitution f rom amino acid sequencing and known sequence of c¡1AT (Long et al, 1984)' Ecincinnati (Hug et al, 1980b)l Erranktin (Cox et al, 1982b)l Elembers (Cox, 1981); Emarsue (Yuasa et al, 198a); Etokyo (Miyaka et al, 1979); F (Eriksson & Laurell, 1963; Fagerhol & Laurell, 1967); G (Fagerhol, 1972), Gcler (Plazonnet et al, 19S0); Jhory"o (Ying et al, 1984); L (Vandeville et al, 1974): Lu"ij¡ng ffing et al, 1984); M5 (Gorg et al, 1985; Weidinger et al, 1985); Mchapel 1980); N n¡ll (Johnson et al, 1976; Gox et al, 1980); Msa¡a (Frants & Eriksson, (Cox & Celhoffer, 1974); Nadetaide (Mulley et al, 1983); Nsrossoeuvr" (Sesboüé et al, 1984), Nhampton (Arnaud et al, 1978); Ntetrait (Charlionet et al, 1981); Nn"s"ro (Yuasa et al, 1984)l Nyervi¡e (Sesboüé et al, 1984); Pbudapest (Cox, 1981); Pcastoria (Coxet al, 1982b), Pctifton (Hug etal, 1981); Pkyoto (Miyake et al, 1979; Cox et al, 1980); Ps¡¡ (Yuasa & Okada, 1984); Psaintalbans (Holmes et R al, 1990c); Psaint ¡ou¡s (Pierce & Eradio, 1981); Pweishi ffing et al, 1985a); (cox, 1981), scotosne (weber & weidinger, 1988); T (Kühnl & spielmann, 1979); V (Fagerhol, 1967); Vmunich (Holmes et al, 1990a); Wrinne¡own (Hug et al, 1982b); Wsazac (Cox et al, 1980; Cox et al, 1982b)i Wsaterno (Cox, 1975); X (Axelsson & Laurell, 1965); Xatban (Cox et al, 1982b)l )bnr¡stcnu¡.¡ (Brennan & Carrell, 1986)t Xenchenn ffing et al, 1985a)l Yurishton (Cook, 1975)l Yhasi (Yuasa et al, 198a); Yroronro (Cox, 1981); Zprar (Hug et al, 1980a).

ln addition, a number of other normal crlAT variants have been described but not fully substantiated : including Ez, h"".rp, Lvibeur (Charlionet et al, 1981); Mlanaheim (Taylor et al, 1980); Msambia (Welch et al, 1980); Mh"itin, Mhuairou (Ying et al, 1985a); Ponomich, Pyasusi (Yuasa et al, 198a); Sberber (Chaabani et al, 1984); Wcotumbus (Hug et al, 1982b).

1.4.4 olAT DeficiencY Variants

As with the normal slAT var¡ants, the majority of crlAT deficiency states are caused by a small number of more common variants with a large numberof rare alleles (allelicfrequency < 0.001) causing deficiency in more isolated instances (Cox et al, 1980; Cox et al, 1981 ; Brantly, 1988a; Crystal et al, 1989; Cox, 1989; Crystal, 1990; Perlmutter, 1991). A larger number of the deficiency alleles (including Null alleles) have been investigated at the molecular level (summarized in Tables 1.2 and 1 .3 and including those in this thesis).

25 TABLE 1.2

o(1AT Deficiency Alleles

ALLELE ALLELIC MUTATION REFERENCE BACKGROUNDl

Yoshida et al, 1976; z M1 1Rta213¡ 6¡1342 GAG ', Lys f,AG Jeppsson 1976; Kidd et al, 1983; Nukiwa et al, 1986b; Jeppsson & Laurell, 1988.

Zaugsburg2 M2 6¡u342 GAG " Lys IAG Faber et al, 1990.

Zwrexham M1 (Ala213) 6¡u342 GAG , Lys f,,AG Graham et al, 1990a. and ser19 tçe n Leu ¡.G

& Canell, 1976; s M1 ryat213¡ ç¡u264 Gf,f, " Lys GfA Owen Owen et al, 1976; Long et al, 1984.

Si¡yama M1 (Val213¡ seFg rec " Phe tlc Seyama et al, 1991.

goc Cys Graham et al, 1989. I M1 1Vat213¡ Rrg39 ' IGC p¡s369 et al, 1989. Mheerlen Mt (Ala213¡ CQC " Leu CIC Hofker

Mmafton M2 Phe52 TTc o delete[Ç. Curiel et al, 1989c; Fraizer et al, 1989a; Graham et al, 1989.

Curiel et al, 1990. Mmineral springs Mt 1Rla213¡ cly67 ece ' Gtu GAG Matsunaga et al, 1990 Mnichinan M1 (Val213¡ Phe52 ttc " delete[Q. and Gy148GGG, ATgAGG

Mprocida M1 (Vat213¡ Leu41 cIe , Pro CQG Takahashi et al, 1988b.

1990c. Plowell3 M1 ryat213¡ Asp256 Gf,T, Gf- Holmes et al,

wbethesda M1 (Ala213) Ata336GCT, Thrf,CT Holmes et al, 1990b.

1 The normal cl1 AT allele onlo which the mutation occurred' 2 Probably identical to Z1u¡þ¡iiqs r¡ys¡¡s (Whitehouse et al, 1989)' 3 Also called Nullç¿¡6¡11 (GrahaÉr et al, 1989) and likely identical to the original P (Faber et al, 1989). TABLE 1.3

cxlAT Null Alleles

ALLELE ALLELIC MUTATION REFERENCE BACKGROUNDl

Null6e¡¡¡¡g¡"t M1 (Vat213¡ Lys217, Stop217 Garver et al, 1986; satoh et al, 1988 AAG ]AG

o et al, 1987b. Nullgranite falls M1 1eta213¡ Ty¡160 Stop160 Nukiwa TA8 TAG | 5'shift delete

Null¡6¡9 ¡e¡g M2 ¡"11318 Stop$e Sifers et al, 1988. CIQ o TAA | 5'shift delete

gene Takahashi & Crystal, Null¡se¡¿ di procida c2 17 kb deletion includingcrlAT 1 990. exons ll - V pro362 Null6s¡s¡ M1 1Vat213¡ Stop3TS Fraizer et al, 1989b. CCC , TAA | 5'shift delete

Null¡¿11¿r¡¿ M1 (Vat213¡ ¡.1353 Stop3TO Curiel et al, 1989b TTA u TAG | 3'shift delete T insertion

Null¡u6w¡gs¡¿1s¡ M2 lle92 , Asn Fraizer et al, 1990. A,IC MC

Null¡¡e6g¡6uyg¡ 72 Gene deletion including Poller et al, 1991 crlATexons 1c - V.

al, 1990. Null¡gwpert M1 1Rta213¡ 6¡u342,, ¡ys Graham et GAG AAG Gly115 ,' t"t GGC Aæ

1 The normal c¡1AT allele on which the Null mutations occurred. 2 Because all of the coding exons are deleled, the allelic background is known. The most common crl AT deficiency variant is the S variant (Fagerhol & Cox, 1981; Lieberman, 1983; Cox, 1989). ln the United States the frequency of the S allele is 0.023 to 0.042 (Pierce et al, 1975; Kueppers and Christopherson, 1978; Dykes et al, 1984). Although in Northern Europe the allelic frequency of the S allele is similar to the United States, in Southern Europe, particularly Spain and Portugal, it is much higher with levels of 0.1 12 to 0.141 respectively (Fagerhol & Tenfjord, 1968). The S allele is absent in most Asian populations (Lee et al, 1981; Yuasa et al, 1984; Ying et al, 1985a; Ying et al, 1985b). The amount of S olAT produced and secreted by liver each day is 65% of normal (Fagerhol & Hauge, 1969). ln an SS homozygote, the serum levels of c¡1AT using a true laboratory standard are 20-48 pM or using a commercial standard 1OO-2OO mgidL (Wewers et al, 1987a; Brantly et al, 1991). Evaluation of the function of the S alAT as an inhibitor of neutrophil elastase reveals that although it functions slightly less well than M crlAT (K association 7.1 x 106 M-1 s-1 versus 9.6 x 106 M-1 s-1 respectively), this is unlikely to be physiologically significant (Ogushi et al, 1988).

The other common deficiency variant, lhe Z variant, which was first described by Laurell & Eriksson (1963), and which is the most common cause of physiologically important crl AT deficiency has, in most Caucasian populations, an allelic frequency of 0.01 to 0'02 (Fagerhol & Cox, 1981 ; Lieberman, 1983; Cox, 1989). ln the United States the allelic frequency has been reported to be between 0.0123 and 0.014 whereas in Denmark it is as high as 0.023 (Pierce et al, 1975; Kueppers & Christopherson, 1978; Dykes et al, 1984; Thymann, 1986). The highest frequency (0.035) is in New Zealand Maoris (Janus et al, 1975). TheZ allele is absent or extremely low in the Japanese and Chinese populations (Lee et al, 1981; Yuasa et al, 1984; Ying et al, 1985a; Ying et al, 1985b). The amount of ZulAT synthesized and reaching the serum is approximately 15/o ol normal (Fagerhol & Hauge, 1969). ln aZZ homozygote, serum levels of olAT are of 3.4 to 7.0 pM using a true laboratory standard or 15-50 mg/dl using commercial standards (Wewers et al, 1987a; Brantly et al, 1991). ln addition to being at lower levels in the serum, lhe Z ol AT also functions significantly less well as an inhibitor of neutrophil elastase as evidenced by the K association ol Z otlAT for neutrophil elastase which is 4.5 x 106 M-1 s-1 whereas M alAT is 9.7 x 106 M-1 s-1 (Ogushi et al, 1987). Although earlier studies with RFLP's had suggested a single evolutionary origin of the Z allele

28 (Cox et al, 1985), more recently a number of other variants with the Z mutation, Glu342 GAG ", Lys342 AAG), have been described including : Null¡"*oon (Z mutation and Gly115 GGC ,,,, $s¡115 AGC; Graham et al, 1990a); Zwr.*ham (Z mutation and Ser'1e TCO ,,,,, Leu-1e TIG; Graham et al, 1990a); Z^ussaurg(Z mutation and M2 background; Faber et al, 1990). Ztunbridse wells, â var¡ant characterized by RFLP's and oligonucleotide probes, is likely identical to Zaussburs (Whitehouse et al, 1989a).

Of the rare deficiency alleles, those which have not been investigated at the molecular level and which are not summarized in Tables 1.2or 1.3 are - Mdr",r" (Lieberman et al, 1976); Ml¡te (Kueppers et al, 1977); M,or"n (Martin et al, 1975).

1.4.5 Dysfunctional crlAT Variants

Perhaps the most interesting and well known dysfunctional variant is the Pittsburgh variant (Lewis et al, 1978; Owen et al, 1983). This variant has a mutation in the codon forthe active site methionine, [¡tlsl358 ¡¡¡ Arg 3sa (Owen et al, 1983). This gives rise to an crl AT which functions poorly as an inhibitor of neutrophil elastase, but which is an excellent inhibitor of thrombin, FactorXla, Kallikrein and FactorXll (Lewis et al, 1978; Scott et al, 1986; Travis et al, 1986). The patient described with this ø1AT variant died with a haemorrhagic complication (Lewis et al, 1978).

As well as the decreased ability of the ZaIAT and, to a lesser extent, the S glAT to function as inhibitors of neutrophil elastase, the Mmineratsprinss ø1AT also has a decreased K association for neutrophil elastase compared with M crlAT (5.8 x 106 M-1 s-1 versus 8.9 x 106 M-1 s-1, respectively; Cu¡el et al, lggo). No other dysfunctional olAT variants have been described although the F variant may prove to be dysfunctional if reported associations with emphysema are correct (Beckman et al, 1984a; Kelly et al, 1989).

29 1.5 MOLECULAR AND CELLULAR BASIS OF olAT DEFICIENCY

Three basic mechanisms for alAT deficiency (relative or absolute) in serum have been identified. The first and most easily conceptualized mechanism is the absence of cr,1AT due to either absence of the ol AT gene as seen with the Null¡"ota di procida and Null¡¡s6s¡6ursh âlleles Or absenCe of crl AT mRNA transcripts as shown for the Null6"¡insham and Nullnranite fatts alleles (Garver et al, 1986; Holmes et al, 1989a; Takahashi & Crystal, 1990; Poller et al, 1991). Secondly, the variant crlAT protein may accumulate within the olAT producing cells, a pattern demonstrated in 6¡lAT Z Mduarte, Mmahon, S¡¡y¿¡¿ ând al, Null¡onn kons vâriânts (Sharp et al, 1971; Lieberman et al, 1976; Sifers et 1988; Curiel et al, 1989c; Seyama et al, 1991). Thirdly, there may be increased degradation of the newly synthesized olAT prior to its secretion from the cell as has been observed for the common S deficiency variant, Mmineralsprings, Ptowe¡, Wbethesda and Nullrailawa (Curiel et al, 1989b; Curiel et al, 1989d; Curiel et al, 1990; Holmes et al, 1990b; Holmes et al, 1990c). The mechanism of ol AT deficiency for ol AT deficiency variants other than those indicated above is unknown.

A large number of studies and approaches have been used to dissect the mechanism of the olAT deficiency directed by the crlAT Zallele. lt has been shown that the cr,1 AT Z allele is transcribed equally as well as the normal M gl AT allele and that the resultant cr,1AT mRNA transcripts are translated with equal efficiency (Errington et al, 1982; Bathurst et al, 1983; Perlmutter et al, 1985b; Errington et al, 1985; Mornex et al, 1986; Verbanac & Heath, 1986; Schwarzenberg et al, 1986). These observations, together with the knowledge that the plasma half-life of the ZaIAT in serum is similarto the M o1AT, suggest that the deficiency ol Z alAT is due to abnormal processing in the intracellular protein secretory pathways (Laurell et al, 1977; Glaser et al, 1977). lt is generally accepted that the deficiency of Z ulAT is due to intracellular accumulation of this variant protein which is consistent with the observation in hepatocytes of Z homozygotes of accumulated crlAT (Sharp, 1971; Gordon et al, 1972; De Lellis et al, 1972; Palmer et al, 1974)- ln addition, in vitro studies of translation utilizing Z ulAT mRNA also demonstrate intracellular accumulation of crlAT (Errington et al, 1982; Errington et al, 1985; perlmutter et al, 1985). Furthermore, both in vivo and in vitro analyses have

30 demonstrated that this accumulation of ol AT is in the RER prior to transport to the Golgi, which explains the earlier studies indicating abnormal glycosylation of the intracellular Z alAT (Feldman et al, 1974i Eriksson & Larsson, 1975i Jeppsson et al, 1975; Hercz et al, 1978; Hercz & Harpaz, 1980; Bathurst et al, 1984; Callea et al, 1984; Foreman et al, 1984). The mechanism of the accumulation of the Z ø1AT is still not completely understood (Sifers et al, 1989a; Sifers et al, 1989b; Perlmutter, 1991). lt has been suggested that loss of the salt-br¡dge between Qlu3a2 and Lys2eo in the Zc^IAT variant might slow folding of the protein and lead to aggregation (Loebermann et al, 1984). This is consistent with the observations that the Z a1A-l has an increased tendency to aggregate compared with M cl AT and the findings of insoluble aggregates of Z 6-lAT in the hepatocytes of Z homozygotes (Eriksson & Larsson, 1975; Jeppsson et al, 1975; Foreman et al, 1984; Cox et al, 1986). To test the "salt- bridge" hypothesis, a number of investigators have recently utilized in vitro mutagenesis to create various mutant olAT cDNAs as follows : Normal, Glu342 - Lys2so (normal salt-bridge), Z, Lys342 - Lyszeo (disrupt salt-bridge); Lyss+z - Glu2eo (salt-bridge possible); Glu342 - Glu2eo (disrupt salt-bridge); Asp3az - Lys2eo (salt-bridge possible); Argstz - Lys2so (disrupt salt-bridge); Gln342 - Lys2so (disrupt salt bridge); Ala342 - Lys2so (disrupt salt-bridge) (Foreman,1987; Brantly et al, 1988b; Sifers et al, 1989d; Mccracken et al, 1g89; Wu & Foreman, 1990). One study suggests that the loss of the salt- bridge is the major reason for cl AT deficiency associated with the Z a1 Af allele (Brantly et al, 1988b). However, the other studies, though not ¡n exact agreement, indicate that although the salt-bridge disruption is a significant factor in crlAT deficiency associated with the Z allele, the critical factor is the amino acid substitution at position 342 (Sifers et al, 1989d; McOracken et al, 1989; Wu & Foreman, 1990). Even more detailed study by McCracken et al (1991) with ten different amino acid substitutions at position 342 ol alAT and assessment of crl AT product in transfect COS I monkey kidney cells have revealed the cr,1AT secretion is most efficient when position 342 is occupied by a negatively charged amino acid, least efficient with a positively charged amino acid and intermediate with a neutral amino acid. ln keeping with these observations, analysis of the amino acid sequence of crlAT aligned with other serpin members shows that Glu3a2 is a highly conserved amino acid (Huber & Carrell, 1989). ln addition, these studies, and at least one other, have confirmed that the valine or alanine substitution at position2l3 of crlAT does not have any effect on cx,1AT secretion (Sifers et al, 1989d; McOracken et al,

31. 1989; Wu & Foreman, 1 990; Davis et al, 1 990). Why the Z al AT accumulates in the RER is uncertain. The possibility that lhe Z mutation disturbs a receptor mediated exit of al AT from the RER has been raised but not yet demonstrated (Sifers et al, 1989d; McOracken, 1989). Perlmutter et al (1989b) have demonstrated an increased synthesis of some stress proteins in the Z al AT synthesizing cells and it is possible that the malfolded ZoIAT interacts with these causing its retention in the RER. However, three groups have been unable to demonstrate an increased synthesis of the likely stress protein candidate for this interaction, the grp 71lBiP resident endoplasmic reticulum protein, in Z alAT producing cells leading them to question whether or not the ZulAT is in fact misfolded (McQracken et al, 1989; Graham et al, 1990; Cresteil et al, 1990). lnterestingly, the study of Sifers et al (1989d) found the human Z alAT accumulated in a soluble form in transfected mouse hepatoma cells, suggesting aggregation is secondary to retention in the RER rather than its cause. ln addition to retention within the RER, lheZ crlAT is also degraded within and following its exit from the endoplasmic reticulum (Le et al, 1990; Le et al, 1992). Although the cr,1AT degradation demonstrated by Le et al (1990) did not appear to involve lysosomal enzymes, previous studies by Bathurst et al (198S) have suggested that the accumulaledZ slAT in xenopus oocytes stimulated lysosomal activity suggesting a possible role of lysosomal enzymes in the degradation of Z a1AT. lnvestigation of two other GlAT deficiency variants which accumulate within hepatocytes, Mmatton and Null¡ons kons indicate they too accumulate in the RER (Sifers et al, 1988; Curiel et al,

1 989c).

Like the Z ç-lAT variant, the common S olAT deficiency variant amino acid substitution (Gluzo¿ ,,,,, Val) is thought to disrupt a salt bridge (Gluzo¿ - Lysssz¡ and an internal hydrogen bond (Gluz0¿ - Tyrsa¡ (Loebermann et al, 1984; Engh et al, 1989). However, unlike Z a1AT, S Û1AT has not been observed to accumulate within cells (Carrell & Owen, 1979). Although the S alAT has a slightly higher catabolic rate and possibly increased heat lability compared to M cr1AT, this is not enough to account for the serum deficiency of crlAT associated with this allete (Lieberman, 1973; Jeppsson et al, 1978b). The S gl AT gene mutation generates a false splice site and this has been suggested as a potential cause of reduced S crlAT mRNA and protein (Carrell, 19g6; Sifers et al, 1989a). Elegant and detailed in vivoand in vitro studies by Curiel et al (1989d) have shown that the deficiency of S cl AT is not due to

32 aberrant splicing but is instead due to intracellular degradation of S ol AT prior to secretion. Whether this degradation was by lysosomal or non-lysosomal pathways was not able to be determined though it was at least in part inhibited by leupeptin (Curiel et al, 1989d). Like Glu3a2,Glu264 is a conserved amino acid among serpins (Huber & Carrell, 1989).

1.6 alAT VARIATION AND DISEASE

f .6.1 Lung Disease

The observation of an association between al AT deficiency and premature obstructive lung disease was a major stimulus to current interest in cr1AT. The early studies forming the basis of the protease- antiprotease theory for the pathogenesis of emphysema have been reviewed (1.1)

The general belief that there is an increased risk of emphysema when serum levels of crlAT fall below 35% of normal (or 11 pM), is based on the observation that ZZhomozygotes (15% of normal serum levels of cr1AT,3.4 - 7 pM) and SZ heterozygotes (35% of normal serum levels of cr1AT, 10 - 23 pM) appear to be at an increased risk to develop obstructive lung disease (Fagerhol & Hauge, 1969; Larsson et al, 1976; Brantly et al, 1gg1). The SZ Heterozygotes are however at much less risk of disease than ZZhomozygotes, to the extent that it has been questioned whether they are in fact at risk of emphysema at all (Larsson et al, 1976: Gishen et al, 1982; Hutchison et al, 1983; Liebermann et al, 1986; Brantly et al, 1988c). This is consistent with the alAT serum levels of SZ heterozygotes being around this "threshold" level below which there is increased pulmonary disease association.

The ZZ homozygote ¡s by far the most common crl AT deficient phenotype associated with emphysema (Eriksson, 1965; Fagerhol & Hauge, 1969; Larsson et al, 1976; Larsson, 1978; Tobin et al, 1983; Lieberman et al, 1986; Brantly et al, 1988c). crlAT deficiency has been estimated to account for approximately 2"/" of the cases of emphysema in Caucasians (Lieberman et al, 1986; Wewers et al, 1987a).

33 There appears to be a high degree of variability in the development of lung disease among ZZhomozygotes (Eriksson, 1965; Morse et al, 1977; Black & Kueppers, 1978; Sveger, 1978; Larsson, 1978; Gishen et al, 1982; Tobin et al, 1983; Janus et al, 1985; Wu & Eriksson, 1988; Brantly et al, 1988c; Silverman et al, 1989; Hutchison, 1990; Poller et al, 1990b; Silverman et al, 1990a). Although earlier studies suggested a high proportion of ø1AT deficient persons developed lung disease, recent estimates based on known allelic frequency of the Z allele and the observed frequency of deficient phenotypes in populations with emphysema have suggested that up to 90% of ZZ homozygotes may not present with lung disease (Tobin et al, 1983; Silverman et al, 1989; Hutchison, 1990). The reason for this variability is largely unknown. Certainly, cigarette smoking plays a significant role in the accelerated and progressive lung disease associated with 6¡lAT deficiency (Eriksson, 1965; Black & Kueppers, 1978; Tobin et al, 1983; Carrell, 1984; Janus et al, 1985; Wu & Eriksson, 1988; Brantly et al, 1988c; Silverman et al, t989; Evald et al, 1990; Silverman et al, 1ggoa). Evald et al (1990) found that lung disease appears later in non- smokers though once present progresses at a similar rate to smokers. ln contrast, Janus et al (1985) found that although the decl¡ne in FEVI in non- smoking crlAT deficient patients was high (80 ml/year), it was much higher in smokers (317 ml/yeaQ. lnterestingly, both Evald et al (1990) and Janus et al (1gBS) found that ex-smokers with crl AT deficiency tended to have a slower rate of decline of FEVI compared to non-smokers. The reason for this, if significant, is unclear. The ability of cigarette smoke to add a funtional deficiency of crlAT by oxidizing the active site Met358 and therefore amplify the overall deficiency of the antiprotease screen of the lower respiratory tract in glAT deficiency has been discussed (1.3.3). Other factors which may be important in the variability of lung disease in alAT deficient persons include : male sex (although this seems to be due to increased smoking in males) (Eriksson, 1965; Tobin et al, 1983; Brantly et al, 1988c); atopy (Silverman et al, 1990) and; familial factors (Silverman et al, 1990a; Silverman et al,

1 eeob).

Detailed clinical features of the lung disease associated with alAT deficiency have been described in a numberof studies and include : early onset of symptoms (usually less than 40 years of age); males greater than females; smokers much more prevalent than non-smokers; emphysema

34 of the panacinar type, predominently affecting the lung bases; lung function testing typical of emphysema particularly with reduced FEVI and diffusing capacity; ventilation scans showing retention of gas in the bases of the lungs; perfusion scans showing reduced perfusion in the lower and mid zones of the lungs; chest x-rays typical of emphysema; accelerated loss of lung function; and reduced survival (Eriksson, 1965; Kueppers & Black, 1974; Gishen et al, 1982; Tobin et al, 1983; Janus et al, 1985; Brantly et al, 1988a). ln addition to the common Z deficiency allele, severe crlAT deficiency and lung disease is associated with the rare Null alleles, Mprocida, Mmineratsprings, Mmafton, Mheerlen, Mdr"rt", Mnichinan, Plowell, Siiy"r", Zaugsburg, and Zwrêx¡",¡ when gach allele is inherited in the homozygous state or in heterozygous combination with another severely deficient crlAT allele (Cox, 1989; Crystal et al, 1989; Kalsheker & Morgan, 1990; Crystal, 1990; Matsunaga et al, 1990; Seyama et al, 1gg1). lt has been suggested that the F allele is associated with emphysema although this is not conclusive (Beckman et al, 1984a; Kelly et al, 1e8e).

lnvestigations of a possible association between intermediate al AT deficiency phenotypes MZ (a1AT levels 15 - 42 pM) and MS (cr1AT levels 18 - 52 pM) have been an area of intense activity. MS heterozygotes and SS homozygotes do not appear to be at increased risk for developing emphysema (Cox et al, 1976; Lieberman et al, 1986). ln the case ol MZ heterzygotes, a large number of studies have suggested a significant increase in the number of MZ heterozygotes in populations with emphysema (Kueppers et al, 1969; Lieberman, 1969; Kueppers & Donhardl,1974; Cox et al, 1976; Larsson et al, 1977; Bartmann et al, 1985; Lieberman et al, 1986; Janus, 1986). A number of other studies however, have not found an association between MZ heterozygosity and emphysema (Eriksson, 1965; Shigeoka et al, 1976; Morse et al, 1977; Bruce et al, 1984). The issue is not resolved, however, it may be that for the MZ phenotype to have an influence on the development of lung disease, there needs to be another significant risk factor such as smoking or occupational dusts (Larsson et al, 1977; Eriksson et al, 1985; Janus, 1985; Liebermann et al, 1986; Horne et al, 1986; Klasen et al, 1986).

There may be an association between ø1AT phenotype and other lung diseases. Eriksson's original study in 1965 included examples

35 of bronchiectasis among ZZ homozygotes and other studies have also suggested this association (Longstretch et al, 1975; Varpela et al, 1978; Jones et al, 1985). The S allele has been implicated in asthma, particularly in MS heterozygotes (Townley et al, 1990; Colp et al, 1990; Mackay et al, 1990).

1.6.2. Liver Disease

The association between ZZ ulAT deficiency and liver disease in the form of neonatal hepatitis was first reported by Sharp et al (1969). This observation has been confirmed on numerous occasions (Sharp, 1971: Porter et al, 1972; Aagenaes et al, 1972; McPhie et al, 1976; Burke et al, 1976; Odievre et al, 1976; Moroz et al, 1976a; Sveger, 1976). The clinical picture of the liver disease in neonates includes prolonged obstructive jaundice (10% - 2O%) and/or abnormal liver function tests (up to 50%) (Aagenaes et al, 1972; Sveger, 1976). Upto 30%- 4O/"ol cases of neonatal hepatitis are due to alAT deficiency (Aagenaes et al, 1972; Moroz et al, 1976a). The liver abnormalities usually resolve within the neonatal period (Sveger, 1976). A proportion (up to 50% in some studies) of infants with prolonged obstructive jaundice go on to develop cirrhosis and there may be persisting liver enzyme abnormalities into childhood but the prognosis is not as bad as initiallythought (Sharp et al, 1969; Sveger & Thelin, 1981; Nebbia et al, 1983; Sveger 1984; Sveger, 1988). A study in 1985 suggested that breast fed infants may be protected from severe liver disease in ol AT deficiency, however recent studies did not confirm this (Udall et al, 1985; lbarguen et al, 1990).

The histology in the neonatal liver disease includes : periodic acid-Schiff (PAS) positive alAT globules (by immunohistochemistry) in periportal hepatocytes which are shown by electron microscopy to be in the endoplasmic reticulum; portal inflammation and fibrosis; nodular regeneration and ductular proliferation (Sharp, 1971; Moroz et al, 1976a: Odievre et al, 1976). alAT deposits have been found even in foetal livers in ctlAT deficiency (Malone et al, 1989).

Adults with crlAT deficiency may also develop liver disease (Berg et al, 1972; Delellis et al, 1972; Fisher et al, 1976; Cox & Sm¡h, 1983). This disease is predominantly in the form of cirrhosis and

36 occurs in 12 to 15% of adult ZZ homozygotes, particularly males (Larsson, 1978; Cox & Smyth, 1983; Eriksson, 1985). Malignant hepatoma appears also to be associated with crlAT deficiency (Eriksson & Hagerstrand, 1974; Schleissner & Cohen, 1975; Reintoft & Hagerstrand, 1979; Eriksson, 1985; Eriksson et al, 1986; Parham et al, 1989) although this is controversial (Govindarajan et al, 1981). The Mmalls¡ ârìd Mduarte alAT variants are also associated with liver disease and slAT globules in hepatocytes (Crowley et al, 1987; Reid et al, 1987).

As with pulmonary disease, an association between liver disease and intermediate crlAT deficiency (SZ, MZ and MS) phenotypes is controversial. The balance of evidence suggests there is a small risk of cirrhosis in SZ and MZ patients but not MS heterozygotes (Hodges et al, 1981 ; Carlson & Eriksson, 1985; Pilacik et al, 1990; Bell et al, 1990).

The pathogenesis of the liver disease associated with alAT deficiency has been obscure until recently. Udall et al, (1982) suggested that deficiency of olAT exposed the liver to uninhibited intestinal derived proteases resulting in damage. Littleton et al (1991) have some evidence that deficient NE inhibition causes complement activation and subsequent liver damage. Carrell & Owen (1979) suggested that liver disease was more likely due to damage caused by the accumulated crlAT in hepatocytes. This was supported by the association between al AT deficiency and liver disease being only in those variants with liver inclusions (Feldmann et al, 1975; Carrell & Owen, 1979; Crowley et al, 1987; Reid et al, 1987). Dilatation of the endoplasmic reticulum was shown to be associated with subsequent cell death (Hultcrantz & Mengarelli, 1984). Definitive evidence that ¡t is the accumulation of crlAT in hepatocytes rather than deficiency of ol AT which causes liver disease has come from studies in transgenic mice carrying the human Z alAT (Dycaico et al, 1988; Carlson et al, 1989)' Despite normal serum crlAT (mouse crlAT), animals developed neonatal hepatitis in conjunction with hepatocyte accumulation of human Z o1AI.

1.6.3 Other Disease Associations

crl AT variation has been reported to be associated with a wide range of disorders some of which are only single case reporls. A severe

37. panniculitis and olAT deficiency has been reported often (Rubinstein et al, 1977; Brandrup & Ostergaard, 1978; Pottage et al, 1983; Bleumink & Klokke, 1984; smith et al, 1987; smith et al, 1989). A controversial association with Rheumatoid Arthritis has been described for the Z allele (Cox & Huber, 1980; Beckman et al, 19S4b). There may be an increased incidence of glomerulonephritis in alAT deficiency possibly due to the effect on the immune system (Miller & Kuschner, 1969; Moroz et al, 1976b; Lewis et al, t985; Noble-Jamieson et al, 1990). A necrotizing vasculitis has been found in association with olAT deficiency in three cases (Fortin et al, 1991). Blood pressure may be lower in Z alAT deficient individuals (Boomsma et al, 1991). Uveitis has increased incidence in crlAT deficiency (Brewefton et al, 1977: Brown et al, 1979; Wakefield et al, 1982; Fearnley et al, 1988). For none of these disorders or the many others reported is the link between al AT variation and disease fully substantiated.

1.7 TREATMENT OF crlAT DEFICIENCY

The approaches to specific treatment of the consequences of crl AT deficiency can be conceptualized in the context of the protease-antiprotease theory of emphysema (Gadeket al, 1980a). Attempts may be made to raise crlAT levels above the threshold for lung disease by replacement of crlAT or stimulation of cr,1AT production or alternatively, the goal of treatment can be reduction of the neutrophil elastase burden.

Currently, in the United States of America, accepted treatment of severe olAT deficiency with evidence of associated destructive lung disease is weekly intravenous infusions of alAT purified from the Cohn lV fraction of plasma (Coan, 1988; Buist et al, 1989; Crystal et al, 1989). The aim of replacement treatment is to get the olAT serum levels above the threshold level for emphysema (11 pM) to restore the antiprotease screen of the blood and tissues in particular the lung (Hubbard & Crystal, 1988). Biochemical efficacy in the blood and epithelial lining fluid of the lung, as well as safety of this treatment has been demonstrated by a number of studies (Gadek et al, 1981 b; Wewers et al, 1987a; Wewers et al, 1987b). Because of the difficulties associated with obtaining an assessable clinical endpoint to test the efficacy of

38 olAT replacement therapy, a biochemical endpoint has been accepted by the Federal Drug Administration in America (Buist et al, 1989). Trials of this therapy are just beginning in Australia (Janus & Burdon, 1990). Hubbard et al, (1988) have shown that monthly infusions of fourtimes the weekly dose (60 mg/Kg versus 250 mg/Kg) are able to maintain the olAT level above the threshold protective level without adverse side-effects and it seems likely this more convenient approach will become accepted. Attempts to prolong the time between administration even further by giving very large doses three monthly during plasmapheresis did not achieve more than monthly therapy (Guriel & Crystal, personal communication). The supply of crl AT for treatment derived from human serum is limited and, in an attempt to overcome this, transgenic animals have been proposed as a potential source of human crl AT (Archibald et al, 1990; Massoud et al, 1990). More recently, it has been demonstrated that aerosolized recombinant crlAT is able to restore the antiprotease screen of the lower respiratory tract, suggesting that this safe and convenient delivery method may become the approach of choice (Hubbard et al, 1989a; Hubbard et al, 1989b; Hubbard et al, 1989c). The use of recombinant crlAT also raises the possibility of using oxidation resistant forms of cr,1AT for treatment (Crystal, 1990). Secretory leukoprotease inhibitor, another significant antiprotease found in the respiratory tract has recently been assessed for aerosol treatment in sheep with encouraging results (Vogelmeier et al, 1990). This latterantiprotease may also be useful in alAT deficiency to help restore the antiprotease screen of the lower respiratory tract.

A number of synthetic inhibitors of neutrophil elastase are being evaluated and these hold the potential for oral treatment, oxidant resistant molecules, ease of manufacture and unlimited supply (Powers et al, 1977; Janoff & Dearing, 1980; Kleinerman et al, 1980; Powers, 1983; Roberts & Surgenor, 1986; Doherty et al, 1986). Capitalizing on the known response of slAT production to hormones in pregnancy, attempts to increase levels of serum clAT by stimulating increased production of crlAT by the liver have been made using the synthetic androgen, Danazol, and the anti-oestrogen, Tamoxifen (Laurell et al, 1967; Gadek et al, 1980b; Eriksson, 1983; Wewers et al, 1986; Wewers et al, 1987c). These studies have been only modestly successful and more so with upregulating the S allele than the Z allele presumably due to the intracellular accumulation of alAT associated with the Zallele (Eriksson, 1983; Wewers et al, 1987c).

39 The potential for gene therapy for ø1AT deficiency has been explored by inserting a human crl AT gene into non-human cells, for example mouse fibroblasts or hepatocytes, and then assessing the production of human crl AT in vitro (Ledley & Woo, 1986; Ledley et al, 1987; Garver et al, 1987a). Garver et al (1987a) were able to show that NIH-3T3 mouse fibroblasts, modified to contain a human al AT cDNA by retroviral gene transfer, stably expressed the human clAT gene in vivo when transplanted back into a mouse. Curiel et al (1989e) have used a novel approach by inserting the human olAT gene into antigen-specific T-lymphocytes which have the potential to be targeted to particular sites such as the lung by localized antigen exposure. Modified adenoviruses are also showing promise as vehicles to deliver a normal ø1AT gene to the airways of crl AT deficient individuals (Rosenfeld et al, 1991).

Gene therapy for cll AT deficiency utilizing autologous liver cells modified by retroviral gene transferto contain a human crlAT cDNA is being explored in dogs (Kay et al, 1992). The modified cells are then transplanted back to the liver of the dog by the portal vasculature and are able to produce significant quantities of human q,1AT for at least one month.

A definitive treatment of alAT deficiency is liver transplantation however, this is reserved for end stage liver disease in olAT patients rather than for crl AT deficiency per se (Hood et al, 1980).

Specific approaches to reduce elastase load in the lung are limited. Colchicine is known to reduce the release of NE from neutrophils and has recently been used in a clinical trial in patients with chronic obstructive pulmonary disease (Cohen et al, 1990). Unfortunately, no variables related to elastase burden of the lung were altered significantly by colchicine therapy.

Lung or heart-lung transplantation is being increasingly utilized in alAT deficient subjects with end-stage lung disease (Khaghani et al, 1991). Results are very promising with overall survival of up to 91"/" al one year post- transplantation.

40 CHAPTER 2

MATERIALS AND METHODS

2.1 INTRODUCTION

Rapid advances in the approaches of Molecular Biology during the course of experiments comprising this thesis, in particular relating to the applications of the PCR are reflected in the changes in methodologies utilized. Laborious and time consuming gene sequence analyses requiring the creation of genomic libraries in cosmid vectors and subsequent isolation of glAT clones were replaced by simple and rapid PCR techniques (Saiki et al, 1985; Saiki et al 1986; Mullis and Faloona, 1987; Saiki et al, 1988). PCR also greatly simplified the analysis of point mutations and the production of recombinant plasmids carrying various mutant crlAT genes.

Although materials and methods are described in full in a general manner in this section, because of changing techniques, each chapter describing experimental results includes a brief "Methods" section outlining the techniques used in that particular study and specific details of gene probes and oligonucleotide PCR primers.

Those techniques which are routine (e.9. plasmid DNA preparation, Southern blot analysis) or which I undertook closely following a manufacturer's recommendation are described enough to identify the method used but details are to be found in listed references or manufacturer's inserts.

The methodologies used which are peculiar to the studies in this thesis or, which have undergone significant modifications either prior to introduction to, or within, the Pulmonary Branch, NHLBI are fully described. The procedures and techniques which I did not directly perform myself are described in brief and appropriate references stated. These include examination of liver biopsies, measurement of crlAT serum levels, IEF of

41. serum forglAT, examination of ø1AT function and, differential cell counts of bronchoalveolar lavage returns (see Acknowledgements).

2.2 REAGENTS, RADIOLABELS AND DNA RESTRICTION E NZYM ES

Chemical reagents for common buffers and solutions were obtained from Sigma Chemical Company, St. Louis, Mo.

Tissue culture media phosphate buffered saline (PBS), sterile water and bacterial broths/agar were purchased from Biofluids, Rockville, MD.

Agarose (Ultrapure) was obtained f rom Bethesda Research Laboratories (BRL), Gaithersburg, MD.

Radiolabels were purchased from Amersham Corporation, Arlington Heights, lL.

DNA restriction enzymes were bought from BRL or New England Biolabs, Beverly, MA.

The source of other reagents, enzymes and apparatus is indicated in the text.

2.3. STUDY POPULATION

2.3.1 Source

All patients studied in this thesis (with the exception of the lndex case and family carrying the Vru^¡ch crlAT allele, see below) were referred to the Pulmonary Branch, NHLBI, NlH. Some referrals were either because earlier ctlAT screening had suggested a deficiency and/or a rare crlAT allele, or for assessment for possible commencement of experimental

42 therapies. Alternatively, in the case of the lndex cases, carrying the Ps¿¡¡l albans ârìd Wuethesda alleles, initial contact was for assessment of industrial lung disease. ln the latter situation, the crlAT variants were discovered serendipitously. lnformed consent for all studies was obtained from each patient. Family studies were obtained by liaison with each family member to obtain consent, followed by mailing insulated blood sampling packs to each. These were subsequently taken to a local doctor to draw blood and then returned, on ice, to the Pulmonary Branch. Each family member studied supplied a clotted sample (10 ml) for serum to study crlAT protein phenotype as well as a heparinized sample (40 ml) from which white blood cells could be isolated for extraction of high molecular weight DNA. ln the case of the family carrying the V,nu¡¡.rr ol AT allele, clinical details and samples were obtained by Dr. S. Weidingerwho first notedthis newolAT phenotype in a Munich family. Blood samples were sent from Germany in a similar fashion to that described above.

2.3.2 Glinical Characteristics

The clinical characteristics of lndex cases were evaluated utilizing the facilities of the Clinical Center of the National lnstitutes of Health. Apart from obtaining clinical history and examination, this included - chest x- ray; lung function testing (spirometry, lung volumes measured by body plethysmography, carbon monoxide diffusion capacity, arterial blood gases) 133f,s¡6¡ ventilation lung scans; esmlss¡¡etium-macroaggregated albumin perfusion lung scans; multiple biochemical analysis; and complete blood picture. Other tests, e.g. abdominal ultrasound, were obtained if examination or biochemistry suggested a liver abnormality.

2.3.3 Bronchoalveolar Lavage

Bronchoalveolar lavage was performed to obtain alveolar macrophages for the study ol in vivo ø1AT expression in these patients. The procedure was performed under local anaesthetic with an Olympus lT10 fibreoptic bronchoscope (Olympus Corporation of America, New Hyde Park, Ny). The bronchoscope was passed trans-nasally with the patient in the supine position. One hundred ml of sterile normal saline, 37'C, in 20 ml aliquots was infused into each of 3 segmental bronchi into which the tip of the

43 bronchoscope had been wedged. This was then aspirated and collected for cell counting and isolation of alveolar macrophages (Hunninghake et al, 197e).

The differential cell count of lavage specimens was assessed by removing an aliquot and counting total cells and then removing an appropriate aliquot to make two cytocentrifuge preparations (100,000 cells) and two filter preparations (200,000 cells). The latter was performed to ensure accurate quanitification of lymphocyte populations (Saltini et al, 1984). The remaining cells were pelleted by centrifugation in preparation for subsequent analyses.

2.3.4 MonocytaPheresis

Selected patients underwent monocytapheresis in the Blood Transfusion Service of the Clinical Center to obtain blood monocytes for subsequent study of ol AT biosynthesis.

The cell harvest from monocytapheresis was further purified with modifications of a previously described method (Mornex et al, 1986). The essential modification was to subject the blood cells to two adherence steps to achieve a more pure monocyte population. First, the blood cells were diluted up to 350 ml total volume with Dulbecco's modified Eagle's medium (DMEM), 10% heat inactivated fetal calf serum (FCS), 17o glutamine, 100 p/ml penicillin G and 1OO pg/ml streptomycin (P/S). This mixture was then aliquoted into 50 ml conical tubes (Falcon 2070, Falcon Plastics, Oxnard, Ca.) and fractionated by density-gradient centrifugation in lymphocyte separation medium (LSM, Organon Teknika, Durham, N.C.). The mononuclear cell layer was then washed twice in ice cold PBS. The cell pellet was then resuspended in DMEM , 10"/o FCS, 1% glutamine, P/S medium and aliquoted onto 20 cm tissue culture plates (Falcon 3025) and incubated t hr,37"C in 10% COz The supernatant was then aspirated and saved for future DNA extraction whilst adherent cells (predominantly monocytes and some B-lymphocytes) were washed with ice cold PBS before being removed with a cell scraper and then being replated overnight as above. Adherent cells were then recovered by scraping, washed and counted utilizing a Burker's hemacytometer. Cells were generally greater than 90% monocytes as assessed by morphology.

44. 2.3.5 Liver BioPsY

ln the case of the M¡¿¡¡on lndex case (see Chapter 4) a percutaneous liver biopsy was obtained. Tissue sections were stained with hematoxylin and eosin, PAS with and without diastase digestion, and examined by immunohistochemistry for slAT utilizing an avidin-biotin- conjugated peroxidase labelled anti-cr1AT antibody (Hsu et al, 1980).

2.3.6 Drug TheraPY

To evaluate whether or not the crlAT synthesizing cells of the alAT deficient P¡s,,vs¡¡Z index case could be stimulated in vivo to result in qlAT levels rising above the ll pM threshold that defines the risk for premature emphysema (Wewers et â1, 1987a), tamoxifen (trans lzl isomer of triphenylethylene), 1O mg twice daily, was administered orally. This approach was based on the knowledge that tamoxifen therapy will elevate qlAT levels associated with the M or s ol AT alleles, but not those of the z allele (Eriksson, 19g3; Wewers et al, 1987c). Serial alAT levels, pulmonary function tests, serum electrolytes, hematologic parameters, and liver function tests were followed during treatment.

2.4 ANALYSIS OF G1AT PHENOTYPE

2.4.1 General

al AT phenotype was characterized by a combination of alAT serum levels, IEF and family studies. When necessary immunofixation techniques and IEF with immobilized pH gradients were utilized to confirm IEF bands as al AT or for better resolution of crl AT protein bands respectively (Cox et al, 1980; Fagerhol and Gox, 1981).

2.4.2 al AT Levels

Blood was obtained from each patient or relative in a container without anticoagulant. The clotted sample was then centrifuged at 3,000 RPM and the serum aspirated, aliquoted and stored at -70"C until subsequent analYSis.

45 Three different methods were used to determine crl AT serum levels in these studies: radial immunodiffusion plates (Calbiochem- Behring Corporation, La Jolla, California); enzyme-linked immunoassay ; and nephelometry (Behring Nephelometer Analyser, Calbiochem-Behring Corporation, La Jolla, California) (Mancini et al 1965; Rennard et al, 1980).

An important consideration in analyzing the serum crl AT levels was that commercial standards for ø1AT (e.9. Calbiochem-Behring) overestimate glAT levels by up lo 4O/" (Jeppsson et al, 1978a; Travis and Salvesen, 1983). To overcome this, a known purified true laboratory standard crlAT was used to determine olAT levels in these studies (Wewers et al, 1987a). Results are expressed in micromolar quantities and levels can be converted to more commonly published commercial values (in mg/dl) by multiplying by 7.32 (Wewers et al, 1987a). Using the true laboratory standard, ø1AT levels (pM) corresponding to lEF phenotype are :M1M1 (20-53); M1S (21-43); Mlz(17-33); SS (18-33); Sz(s-22); zz (2.5-8); Null (0) (Brantlvet al, 1991).

2.4.3 crlAT Function

The lndex case for the M¡¿¡on studies is homozygous for this allele (see Chapter 4). Therefore, all crlAT in the serum of this individual is of the Mmahon type, permitting analysis of function of this variant cr,lAT protein as an inhibitor of its natural substrate, NE. This functional assessment of crl AT was carried out by first purifying thê Mmatton protêin from serum by positive selection affinity chromotography, molecular sieving followed by negative selection affinity chromotography and finally pressure filtration (Ogushi et al, 1gB7). An association rate constant of the purified ø1AT for human neutrophil elastase was determined as described by Beatty et al (1980) with the modifications of Straus et al (1985). Purified crlAT preparations from two normal M1(Valzts) homozygotes were used as controls.

2.4.4 Isoelectric Focusing of Serum

IEF in 0.5 mm thick, flat bed polyacrylamide gels (Allen et a;,1974; Frants et al, 1978), was carried outwith a separatortechnique at pH 4-5 according to recommended methods (Constans et al, 1980; Jeppsson et

46 al, 1982) us¡ng ampholytes obtained from Pharmacia, LKB Biotechnology, Piscataway, N.J., and Serva Park Fine Biochemicals lnc., Garden City, N.Y. ln the case of the Vmunich family study (see Chapter 7), IEF was undertaken in the pH range 4.2 - 4.9 using the method of Weidinger et al, (1985).

IEF with an immobilized pH gradient in the range 4.45 - 4.75 was undertaken as described by Gorg et al, (1985) and Weidinger and Cleve, (1986). 0.5 mm x 2OO mm x 260 mm polyacrylamide gels with an immobilized pH gradient from 4.45 - 4J5 were prepared with immobilin (lmmobiline, LKB, Bromma, Sweden) of pK 4.6 and 9.3. After polymerization, washing in glycerol and drying to its original weight, the gel was placed in an LKB2117 multiphor ll (LKB, Bromma, Sweden) at 10"C. 20 ¡tl serum samples containing 1O% dithiothreitol were loaded on the gel and run for t hr,300 V, 15 mA and then 18 hr, SOOO V, 15 mA. This technique was utilized for the Mmatron variant and P variant studies (see Chapters 4 and 6 respectively).

lmmunofixation prints, to confirm the identity of protein bands seen on IEF as cr1AT, were performed by pressure blotting onto nitrocellulose paper (Biorad, Richmond, Ca) using the technique described by Boutin et al, (1985). A rabbit anit-human alAT antibody (Accurate Chemical and Scientific, Westbury, NY), followed by binding with gold conjugate according to the manufacturer's recommendations (lmmuno-gold, Biorad) was used to visualize cr1AT.

Concentration of the olAT from the serum of the Wbethesda lndex case to highlight in IEF WbethesdaclAT bands was undertaken by first pu¡fying the ø1AT by positive selection affinity chromotography followed by molecular sieving and then pressure filtration exactly as described in detail by Ogushi et al, 1987.

2.5 slAT GENE SEQUENCE ANALYSIS

2.5.1 General

Nucleotide sequence analysis of the various crlAT alleles described in this thesis was made possible by using the published nucleotide

47 sequence and restriction fragment profile of the normal ol AT cDNA and genomic DNA (Kurachi et al, 1981; Leicht et al, 1982; Long et al, 1984).

As mentioned earlier, initial studies utilized cosmid cloning to isolate the crlAT gene of interest for subsequent Sequence analysis (Mmalton, Wbethesda) whereas later studies (P¡s,¡vsll, Psaint albans, Vmunich) used the asymmetric primer concentration PCR method to generate single stranded DNA for sequence analysis (Gyllensten and Erlich, 1988).

The technique of cosmid cloning was introduced into our laboratory by M. Hofker, State University of Leiden, The Netherlands. lt was subsequently modified and therefore it is described in detail, but it remains basically the method of Van Ommen et al, (1983) and Van Ommen, personal communication.

2.5.2 Extraction of High Molecular Weight DNA

High molecular weight (HMW) DNA was obtained from whole heparinized blood by modification of a method used for tissue culture cells (Blin and Stafford, 1976; Shimada and Nienhuis, 1985)' Firstly white blood cells (WBC) were isolated from 30-40 ml of blood by mixing with 20-30 mls of polygeline (Behringwerke, Marburg, FRG), in a 60 ml syringe and this was then stood upright for 2 hours. During this time the red blood cells settled to the bottom allowing isolation of the WBC enriched top fraction. WBC's were then pelleted by centrifugation (2,OOO RPM, 5 min) in 50 ml conicals (Falcon 2O7O). The pellet was washed 2 times in ice cold PBS. The final pellet was then resuspended in 2 ml of TE (10 mM Tris Cl pH7.4,1 mM EDTA, pH 8.0). To this was added dropwise,2 ml of 2 times lysis buffer (1% sodium dodecylsulfate [SDS],0.6 M NaCl, 1O mM EDTA,20 mM Tris pH 7.5). One fiftieth of this volume (80 ¡rL) of proteinase K (10 mg/ml, BRL) was added and incubated overnight at 37"C. This mixture was subsequently extracted with phenol, followed by phenol/chloroform, and then chloroform. The HMW DNA was then precipitated in 95% ethanol and spooled out with a glass rod and resuspended in 2 ml of TE overnight. Concentration and purity of the DNA was measured by spectrophotometry (Beckman Du7 Spectrophotometer, Beckman lnstruments lnc. Palo Alto, CA) and integrity of the DNA was analyzed by 1"/" agarose gel electrophoresis of undigested DNA as described by Maniatis et al, (1982).

48 2.5.3 lsolation of the Human olAT Gene by Cosmid Cloning

HMW DNA was subject to a partial Mbol restriction enzyme digest to generate fragments sized between 25 and 50 kilobases (kb) with modifications of the method of Maniatis et al, (1982). 150 pg of DNA was made upto 100 pg/ml with TE. To thiswas added Mbol reaction buffer (170 pl), Ribonuclease A (RNase A, United States Biochemical, Cleveland, OH) 10 mg/ml (1 pl) and sterile bovine serum albumin (100 pg/ml). This was mixed at 4"C for 24lo 72 hours. Aliquots (20 pl x 1 and 10 pl x 9) were put into 1'5 ml microcentrifuge tubes (Eppendorf, Hamburg, FRG) on ice. 2 units of Mbol was added to the 20 ¡tl aliquot and mixed and then a 10 pl aliquot from this tube was mixed with the next tube containing 10 ¡rl of the original solution and so on until the tenth serial dilution. These tubes were then incubated t hr, 37"C. The resulting digests were then run overnight al 20 V on a 0.6"/" agarose gel stained with ethidium bromide. Marker lanes contain¡ng uncut lambda phage and lambda phage digested with the restriction enzyme Hindlll were included. The gel was then photographed under ultraviolet light to calculate the correct amount of enzyme to digest the HMW DNA to 25-50 Kb fragments. The remaining DNA mix was then aliquoted into microcentrifuge tubes (400 pl x 1, A 200 ¡rl x 6) and the calculated amount of enzyme added to the 400 pl tube. serial dilution was then done, and the mixtures incubated, 37'C, t hr. The tubes were then floated in a 65"C water bath to inactivate the enzyme. 10 pl aliquots were analyzed on a 0.6"/" agarose gel as previously' Digests containing 25-50 kb fragments were extracted with phenol/chloroform, precipitated with ethanol and resuspended in TE.

The 25-50 Kb fragments were further purified by a sucrose density gradient centrifugation from 5-25% sucrose in TE/10 mM NaCl. The gradient was made by underlaying 1.5 ml of increasing concentrations of sucrose solution (5, 10, 15, 20, 25%) in a Beckman polyallomer centrifuge tube and finally carefully overlaying the 200 ¡rl sample. This was centrifuged in a Beckman ultra-centrifuge with an SW 41 Ti rotor, 17 hr, 22,000 RPM, 20"C without brake. The centrifuge tube was then pierced in the bottom and 20 microfuge tubes each capturing 3 drops (- 250 pl) of the centrifuged solution were obtained. Aliquots of each (10 pl) were anal'yzed by 0.6% agarose gel electrophoresis and the DNA in those samples containing the 25-50 kb

49 fragments was precipitated in alcohol/sodium acetate and resuspended in TE. Concentration of DNA was checked against a serial dilution of uncut lambda DNA.

The purified fragments were then ligated to the two arms each containing a cos site of the cosmid vector C2RB (Maniatis et al, 1982; Bates and Swift, 1983) previously prepared by a Smal and BamHl DNA restriction enzyme digest. The ligation was of 500 ng of both cosmid arms and inseft DNA using T4 ligase (United States Biochemical). The ligation mix was incubated at 14"C overnight and then packed into lambda phage extract (Gigapack Gold, Stratagene, La Jolla, CA.) according to the manufacturer's recommendations to create infectious phage/cosmid particles.

A small scale infection of E. Coli 1046 cells, which had been grown up the night before and made competent with 10 mM MgS04 (Maniatis et al, 1982) was undertaken to titre the phage/cosmid solution. 5 pl of competent E. Coli 1046 cells was mixed with 5 ¡rl 10mM MgS04 and incubated 97"C,20 min. 200 pl of room temperature LB broth was added and the incubation continued for 30 minutes. 20 ¡tl aliquots of this solution were then addedto 2OO pl aliquots of LB broth, plated on 10 cm LB-agar, ampicillin (Sigma) plates, and then incubated 37"C overnight. Bacterial colonies were counted from which colonies per ¡rl of phage/cosmid solution could be calculated. Then, a "large scale" infection at the same ratios as above was done to make 20 bacterial plates each with 40,OOO colonies per plate (done in plates). 10 tubes [Falcon 2059] each with enough phage/cosmid to make 2

After incubation,37"C,30 min the tubes were centrifuged, 3,000 RPM, 4 min and the pellet resuspended gentty in 720 ¡tl of the supernatant. 350 pl aliquots were spread onto numbered nitrocellulose discs (BA 85/2l,Schleicher and Schuell, Keene, NH) which had been placed on 10 cm LB-agar, ampicillin plates. The discs had been numbered M-1 to 20 (Master -1 to 20) and pierced with 5 random holes through both the filter and the LB-agar for later re-alignment. The plates were incubated,37"C, overnight (from late afternoon). Duplicate filters were then made by compressing the grown filters to a new, numbered (but without M) filter with a soft rubber ball. ldentical holes in the new filter were made through the holes of the original Master filter. The two filters were carefully peeled apart and put onto fresh LB-

50 agar ampicillin plates and incubated,37"C, until "recovered" (colonies become rounded again which usually took t hour for the master plate and 1 to 6 hours for the replica). The Master plates were then refrigerated at 4"C. The replica plates were transferred to fresh LB-agar ampicillin, chloramphenicol plates and incubated, 37'C, overnight to amplify cosmid growth.

The filters were then prepared for pre-hybridization by laying them sequentially on filter papers (G8004, Schleicher and Schuell) soaked in 0.2 M Na0H/1% SDS, 2 - 4 min (to lyse bacteria and fix DNA to the nitrocellulose), 0.5 M Na0H/1.5 M NaCl, 2 - 4 min (to denature DNA), 0.5% Tris/1.SM NaCl, 4 min, (to neutralize) followed by washing in 2 x SSC (Standard Saline Citrate)/O.1% SDS and rubbing gently with a gloved finger to remove bacterial proteins and finally rinsed in 2 x SSC. Filters were then air dried for 30 min and vacuum baked, 80"C, 90 min between filter paper. Filters were then pre-hybridized in sealed plastic bags with 50 ml of pre-hybridization fluid (40% formamide, 0.9 M NaCl, 50 mM NaH2P04 pH 7.4,5 mM EDTA, 0.1% SDS, 10 x Denhaft's solution, 100 pg/ml salmon sperm DNA) at 42oC in a shaking waterbath overnight. Five ml of 50% Dextran was then mixed into the pre-hybridization mixture and also 1 pg of denatured ø1AT cDNA (pPB01, Courtney et al, 1984, Garver et al, 1987b), labelled with [as2p]-CTP by Nick Translation (Nick Translation Kit, Amersham Corporation) to greater than 108 cpm/pg as measured by liquid scintillation. The bag was sealed and left overnight , 42'C, in a shaking water bath. The nitrocellulose filter discs were then washed serially in 2 x SSC1}.1% SDS at room temperature, 2 x SSC/O.1% SDS at 65"C 15 min (twice), 1 x SSC/O.1% SDS at 65"C (twice),0'3 x SSC/0.1% SDS at 65"C (twice) and finally 2 x SSC at room temperature. Discs were dried and fixed to a clear piece of autoradiographic film which was marked in the corners with radioactive dye for later alignment. The discs were then autoradiographed (Kodak XAR 5, Eastman Kodak, Rochester, NY) -7O'C, overnight.

The exposed film was aligned with the filters by the radioactive markers and dots were made on the film corresponding to the holes in the filter. Positive signals on the autoradiogram were pierced with a hol22 gauge needle and numbered (1 to X). The master plates were aligned onto the film using the five random dots. Through the pierced holes, a mark was made on the master plates corresponding to the positive signal. Positive

51. areas on the master filters were then aspirated with 2 ¡tl ol 10 mM MgSOa by setting plates on a light box and looking directly from above to avoid parallax. This material was put into correspondingly numbered microcentrifuge tubes containing 1O pl 1OmM MgS04 and mixed gently. 2 ¡tl oÍ this solution was transferred to microcentrifuge tubes labelled D1 to DX containing 1 ml 10 mM MgSO4 and mixed. Tubes labelled 1 to X were stored at 4oC. 20 ¡tl and 100 pl aliquots of each sample D1 to DX were plated onto 10 cm LB-agar, ampicillin plates and incubated, 37"C, overnight.

For each resultant agar plate which was not too confluent with colonies (:200 colonies/plate) a nitrocellulose disc was used for a colony lift by placing it carefully on the plate, marking and piercing 5 random holes and then lifting off quickly. The filter discs were fixed as previously and the plate incubated, 37"C, until colony areas became cloudier (usually 1 to 2 hrs). The filters were pre-hybridized, hybridized to a labelled crlAT CDNA and autoradiographed as previously. Positive colonies were picked with a sterile loop after alignment as described. These were grown overnight in 5 ml LB- ampicillin broth, 37'C, in a shaking incubator. Cosmid DNA "miniprep" extractions were then done exactly following the method of lsh-Horowicz and Burke (1981). Aliquots of the cosmid DNA (2 pl) were then digested with EcoRl and run on 1"/o agarose gels stained with ethidium bromide. Known full length crlAT gene conta¡n¡ng cosmids digested with EcoR lwere run as controls (curiel et al, 1989b). cosmids contain¡ng the correct restriction profile suggesting full length clones based on the control and the known EcoRl restriction profile of cx,1AT (Leicht et al, 1982) were saved for further analysis.

ln the case of the M¡¿¡sn lndex case who appeared, by IEF and family studies, to be homozygous for the M¡n¿¡1on allele (or possibly an Mmatron Null heterozygote), no further characterization of full length cosmid clones was undertaken prior to subcloning into the pUC19 plasmid for nucleotide Sequence analysis. ln the case of the W6s1¡""6" allele, the lndex case was an SWoetn""6" heterozygote, so that cosmids containing a full length otlAT gene COUId be either "S" Or "Wberhesda" COSmidS. TO avoid SubClOning and sequencing the cllAT S allele, those full length cosmids containing the S allele were identified by dot-blot (see below) of "miniprep" cosmid DNA onto nitrocellulose and probing with synthetic oligonucleotides corresponding to the normal or mutant sequence at the site of the S mutation (Nukiwa et al, 1986a).

52 gzP]-ATP Oligonucleotides were '5' end-labelled with [cr using T4 polynucleotide kinase (BRL).

Prior to use in subcloning, pUC19 was cut with Pstl and treated with calf intestinal alkaline phosphatase to prevent self-ligation according to manufacturer's recommendations (United States Biochemical). "Miniprep" DNA of selected cosmids was then digested with Pstl, extracted with phenol/chloroform, ethanol precipitated, resuspended in TE buffer and quantitated by agarose gel electrophoresis. A ligation reaction favouring single inserts into pUC19 (molar ratio of vector greater than insert) was undertaken and after an overnight incubation, 14'C, the ligation mix was transformed into competent E. Coli JM109 cells (Stratagene) strictly following manufacturer's recommendations. The transformation was plated onto LB- agar, ampicillin plates and incubated, 37'C, overnight. A colony lift was then done as previously, probed with an [cr32P]-CTP labelled crlAT probe and positive colonies harvested and a "miniprep" DNA extraction performed. The resultant plasmid DNA was then digested with Pstl and electrophoresed on 1o/o agarose gels with a Pstl digested cosmid containing a full length crl AT gene used as a control. This gel was then transferred by the method of Southern (1975) onto a nitrocellulose filter (Schleicher and Schuell) with modifications as suggested by Maniatis et al, (1982). The resultant filter was probed with an szP]-CTP [çr labelled crl AT cDNA probe and autoradiographed.

Four different subclones were sought and identified by comparison to the cosmid control: a clone containing a 0.5 kb insert encompassing exon lc of crlAT; a clone with a 1.6 Kb insert encompassing exon ll; a clone with a 2.4 kb insert encompassing exons lll and lV; and a clone with a 1.1 kb insert encompassing exon V (Nukiwa et al, 1986b; Curiel et al, 1989b). One of each type of subclone was chosen in each case and a large scale plasmid preparation undertaken by the alkaline lysis method (Birnboim & Doly, 1979; Maniatis et al, 1982). This method yielded plasmid DNA which was suitable for subsequent nucleotide sequencing'

2.5.4 Single Stranded DNA Generation by the Polymerase Chain Reaction

To achieve sequencing quality DNA by PCR, the method of Gyllensten & Erlich (1988) was used. This method relies on unequal

53 concentrations of PCR primers such that after limited cycles, one primer is exhausted leaving only the other primer to generate single stranded DNA from the template by the DNA polymerase in subsequent cycles. lnitial attempts to use this technique to generate single stranded DNA from genomic DNA as a template were unsuccessful due to unacceptable background on autoradiographs when sequencing this material. However, by first amplifying the region of interest in the normal manner as recommended by Perkin Elmer Cetus Corporation, Norwalk, CT, purifying this double stranded amplified DNA by agarose gel electrophoresis, isolating the DNA by electroelution (Maniatis, 1982) and ethanol precipitation, and then using a small aliquot (1ng) of this as a template for subsequent single stranded DNA generation using unequal primer concentrations, an excellent sequencing reaction was possible. This approach was used to sequence the olAT Ptowe¡, Psaint atbans and Vmu¡¡s¡ alleles (see Chapters 6 and 7 ).

Exons lc, ll-V, were amplified with oligonucleotide primers based on the AAT gene sequence published by Long et al (1984)'

For the following primer pairs (1 - 5), the first number indicates the exon amplified, P or S denotes Antisense or Sense strand primers respectively, the numbering in brackets refers to the position of the primer in the cx,1AT sequence corresponding to the numbering of Long et al (1984) and finally, the expected size of the amplification product for a given pair of primers is given.

1 rs(1586-1608) 5'- GTGACTCAGTAAATGGTAGATCT - 3'

1P(2136-2155) 5' . TAGGAGCTCAGCTGCAGCCT - 3'

exon lc amplification product = 569 bases

2. 2s(7234-7257) 5' - ACGTGGTGTCAATCCCTGATCACT . 3'

54. 2P (8017 - 8043) 5' - GGAGAGTTCAAGAACTGATGGTTTGAG . 3'

exon ll amplification product = 809 bases .

3. 3S(e333-e360) 5' - GACTCATGGTTTCTTATTCTGCTACACT - 3'

3 P (9787 - 981 1) 5' - AGAGTAGCAGTGACCCAGGGATGTG - 3'

exon lll amplification product = 478 bases

4. 4S(10850-10877) 5' - TCAGAAAAGAAAACACTTGCACTGTGGT - 3'

4 P (11135 - 111s7) 5' . CAGTGCTGCAGCCCCCACACATTC - 3'

exon lV amplification product = 307 bases

5. ss(11851-11874) 5' - AGCCTTGCTCGAGGCCTGGGATCA . 3'

5P(12197-12222) 5' - CAGAGAAAACATCCCAGGGATTTACA - 3'

exon V amplification product = 371 bases

All oligonucleotide primers were synthesized on an Applied Biosystems DNA synthesizer (Applied Biosystems, Foster City, CA). After synthesis and "deblocking" in ammonium hydroxide by overnight incubation at 55"C, the oligonucleotides were purified according to the method of Jayaraman (1987). This involved simply passing the oligonucleotide / ammonium hydroxide mixture through a PD-10 Sephadex G-25 column (Pharmacia) and eluting with 10 ml of water. One ml fractions passing through the column were collected. Fractions 2 and 3 contain oligonucleotide. The

55 concentration of oligonucleotide was checked by spectrophotometry and a 20 pM solution made for subsequent PGR reactions.

PCR was performed using a Gene Amp* DNA amplification Reagent Kit (Perkin Elmer Cetus) in a Thermal Cycler (Perkin Elmer Cetus).

To generate double stranded DNA, 1 pg of genomic DNA template was used in a 100 pl reaction according to the manufacturer's recommendations and overlaid with mineral oil. An initial 3 min denaturation was followed by 25 cycles of : 94"C, 1 min; 54"C, 2 min; 72"C,3 min and then a final cycle ol72'C,7 min. After purification as described previously, 1 ng of this amplified DNA was used as a template for single stranded DNA generation in 100 pl reactions using the same primers as for the double stranded DNA generation but at unequal concentrations (2'5 pl of a 20 pM primer and 2.5 ¡rl of 0.2 pM other primer). Single stranded DNA was made in both directions (in separate reactions). This reaction mix underwent 40 cycles of PCR as above. The amplification product was then collected carefully without aspiration of the mineral oil overlay and purified 3 times in Centricon 30 Ultrafiltration tubes (Amicon, Danvers, MA) to remove unused primers' The final retentate (usually around aO pl) was divided into 2 microcentrifuge tubes and dried in a vacuum centrifuge. Usually a small (<1 mm) glistening DNA pellet could be seen. This material was then ready for the sequencing reaction.

2.5.5 DNA Sequence Analysis of crlAT Variants

To sequence each of the exons of the variant olAT genes, either20 pg of cloned double stranded crlAT in pUC19, orthe product of the single stranded DNA generation as above were used in the di-deoxy chain termination reaction (Sanger et al, 1977), using Tt DNA polymerase (SequenaserM DNA Sequencing Kit, United States Biochemical) and [3sS]- dATP (Amersham) as a label. The sequencing reaction followed the manufacturers recommendations. Sequencing primers were synthesized based on the published olAT sequence (Long et al, 1984) to cover each of exons lc, and ll-V as well as exon/intron boundaries and 100 bases 5'to exon lc (Nukiwa et al, 1986b). All mutations, ambiguities, or variations from the

56 published sequence of crlAT were confirmed or resolved by sequencing in both the sense and anti-sense directions.

Sequencing primers were : (numbering is corresponding to the published sequence).

Exon lc Sense Primer 1 (1758 -1775) 5' . TCGGTAAGTTGCAGTGGAA - 3'

Antisense Primer 1 (2015 - 2029) 5'- AACCCCTCGCAGTGA - 3'

Exon ll Sense Primer 1 (7234 - 7257) 5'- ACGTGGTGTCAATCCCTGATCACT - 3'

Sense Primer 2 (7415 - 7429) 5'- AAGACAGATACATCC - 3'

Antisense Primer 1 (8000 - 8013) 5'-TATTTTTGCTTGTT-3'

Antisense Primer 2 (7760-7774) 5' - CTTATCCACTAGCTT. 3'

Exon lll Sense Primer 1 (9372 - 9386) 5'- TCACTCACCCCTGGT - 3'

Sense Primer 2 (9510 - 9525) 5'- GGCATGTTTAACATC - 3'

Antisense Primer 1 (9475 - 9489) 5'- CCTTCACGGTGGTCA - 3'

57 Exon lV Sense Primer 1 (10880 - 10894) 5' - GTCCCAGAAGAACAA . 3'

Antisense Primer 1 (11115 - 11129) 5' . CTACAGATACCATGG . 3'

Exon V Sense Primer 1 (1 1875 - 11889) 5'- GCCTTACAACGTGTC . 3'

Sense Primer 2 (12035 - 12049) 5'- AATGATTGAACfufuM - 3I

Antisense Primer 1 (12071 - 12085) 5' - TTCACCACTTTTCCC - 3'

2.6 IDENTIFICATION OF crlAT POINT GENE M UTATIONS

2.6.1 General

Throughout the studies described in this thesis, there was a need to be able to rapidly and conveniently identify point mutations in the slAT gene. This was necessary for:screening clones during isolation of the crlAT gene; identification of new variant slAT cDNAs formed by mutagenesis of normal crl AT cDNAs; screening of family members for the presence or absence of a given gene mutation. lnitial studies used radiolabelled allele specific oligonucleotide probes (l(dd et al, 1983), however because of the inconvenience of working with radioactive materials, a simple non-radioactive method was sought. To this end the method al allele specific amplification (ASA) with PCR was conceptualized by Dr. H. Okayama in our laboratory and developed by our group (Okayama et al, 1989a).

58 2.6.2 Allele Specific Oligonucleotide Analysis

For allele specific oligonucleotide analysis, either 50 ng of purified cloned plasmid crlAT DNA or, 50 ng of cr,lAT DNA amplified from genomic DNA by PCR, was denatured and immobilized onto a nylon membrane (Zetaprobe, Biorad) with a dot blot apparatus (Minifold, Schleicher & schuell) according to the method of Peterson et al (1988). PCR of genomic DNA used the specific exon amplifying primers described previously to amplify the exon containing the mutation being evaluated at the centre. Allele specific 19 or 21 base oligonucleotide probes designed with either the specific mutation being evaluated or the normal sequence at the site of the mutation (Kidd et al, 1983; Nukiwa et al, 1986a) were synthesized and labelled by 5' end-labelling as previously described. Hybridization and washing conditions of the probes and filters were ascertained by computation of theoretical melting temperatures (Maniatis et al, 1982) hybridizing at 4"C below this and then washing with 2"C incremental increases in the temperature of the final wash until a clear unambiguous signal was obtained.

2.6.3 Allele Specific Amplification with the Polymerase Chain Reaction

ASA a rapid, simple, non-radioactive technique is based on the fact that priming of DNA synthesis in PCR by DNA polymerases is much less efficient if there is a mis-match at the 3'end of one of the PCR primers (the allele specific amplification or ASA primer). Using this concept, primers were designed such that the 3' base of the ASA primer was either complementary to the normal sequence or the mutated nucleotide at the site of a given mutation (Okayama et al, 1989a). Amplification in combination with a common distal primer (conserved sequence in each of the alleles being evaluated) occurred only if the 3' base of the ASA primer is complementary to the target DNA. Using 2 reactions for each target (one "normal", one "mutant" ASA primer) the alAT genotype of a given DNA (cloned or genomic) could be asceilained by analysing the product of PCR for the presence or absence of PCR product by 1o/o agarose gel electrophoresis, staining with ethidium bromide and ultraviolet photography. Other groups simultaneously developed this very convenient method (Newton et al, 1989; Ehlen and Dubeau, 1989).

59 When analysing genomic DNA, 500 ng of template DNA in a SO pl reaction was used whereas for cloned DNA, 0.5 ng or less was used. For each mutation a number of ASA primers were often evaluated before achieving the best result. As with all PCR reactions, negative controls (no DNA) and, where possible, DNA controls with known genotype at the site of the mutation being evaluated, were included in the examinations. Details of ASA and common distal primers used for the various mutat¡ons are given in the chapters describing individual variants.

2.7 ANALYSIS OF IN VIVO crlAT BIOSYNTHESIS

2.7.1 olAT mRNA AnalYsis

Where appropriate and where possible, al AT biosynthesis at the level of al AT mRNA production was evaluated in readily accessible cells normally expressing the crl AT gene (alveolar macrophages or blood monocytes, Perlmutter et al 1985a; Mornex et al, 1986). Cells obtained from normal volunteerswith an olAT M1M1 genotype were used as controls. Total cellular RNA was extracted from cells by routine methods using a combination of guanidine hydrochloride extraction and cesium chloride ultracentrifugation (Cox, 1968; Ghsin et al, 1974).

Qualitative analysis of crlAT mRNA was undertaken by denaturing formaldehyde agarose gel electrophoresis of 10 pg of total cellular RNA per lane and Northern blotting to nitrocellulose filters as described in Maniatis et al, (1982). Filters were probed with an lcrszPl-CTP labelled y-actin cDNA probe pHFyA-1; courtesyof P. Gunning, L. Kedes (Gunning et al 1983) followed by autoradiography, -70'C.

Quantitive olAT mRNA analysis was by cytoplasmic dot- btot hybridization (White & Bancroft, 1983) with the only modification being the addition of 1O pl of ribonuclease inhibitor (RNasin, Promega Corporation, Madison, Wl) to each 1 ml of 5% Nonidet P-40 (Sigma)' 5 x 106 cells (monocytes or macrophages) for lndex cells or controls were extracted, and RNA preparations serially diluted onto a nitrocellulose membrane (8485,

60 Schleicher and Schuell) by a dot-blotter (Minifold, Schleicher and Schuell). After hybridization with an [øszP¡-CTP labelled crl AT cDNA probe (pBO1) and autoradiography, resultant autoradiograms were quantitated by laser densitometry (LKB). Experiments were performed in triplicate and results analysed statistically by the two-tailed Students t-test.

2.7.2 alAT Protein BiosYnthesis

ln the case of the studies of the crlAT Mmalton allele, the Mmatron homozygote lndex case enabled evaluation of the production of the Mmatron crlAT protein in cells obtained from the patient as this was the only cr,1AT protein produced by this individual'

Blood monocytes were obtained by monocytapheresis as described previously. These cells were then examined by pulse-chase protein protein a labelling with 1sss¡-methionine, immunoprecipitation of labelled with rabbit anti-human crlAT antibody (Dako Corporation, Santa Barbara, CA), and SDS-polyacrylamine gel electrophoresis followed by fluorography using described methods with minor modifications as described below (Perlmutter et al, 1984; Perlmutter et al, 1985b; Mornex et al, 1986).

After overnight adherence, purified monocytes were resuspended in room temperature "pulSe media" (45 ml Methionine Free DMEM; 5 ml dialyzed foetal cell serum; 0.5 ml glutamine) at 15 x 106 cells per ml. To 2 ml of this suspension was added 1 mOi/ml of lsss]-methionine in a 15 ml polypropylene tube (Falcon 2059), and then an incubation 37"C, 1O"/" COz was undertaken for t hour (i.e. pulse labelling). ln this media which contained only labelled methionine, any protein synthesized during the labelling period was labelled with [35S]-methionine. After t hour this suspension was centrifuged, 1,500 RPM, 3 min. The supernatant was aspirated and the cell pellet resuspended in PBS to wash away excess methionine. Another centrifugatin followed and the cell pellet was resuspended in 2 ml of "chase media" (45 ml DMEM; 5 ml dialyzed foetal calf serum; 0.5 ml glutamine; 0.5 ml 10OmM Methionine). A 1 ml aliquot in a Falcon 2059 tube was incubated g7"C, 10% COz. 2 hrs (Chase period). The other 1 ml aliquot was again centrifuged 3,OOO RPM, 4 min atter which the supernatant was aspirated and discarded and the cell pellet resuspended in 800 pl PBS, centrifuged again,

61. 3,OOO RPM, 4 min and this repeated. Finally, the cell pellet was resuspended in 8OO pl of fresh "lysis medium" [99 ml PBS; 1 ml 1% Triton X 100; 500 mg sodium deoxycholate; 336 mg EDTA; 34 mg PMSF (phenylmethlysulfonyl fluoride, BRL); Smg Leupeptin (Boehringer Mannheim, lndianapolis, lN); 10 mg Aprotinin (Boehringer Mannheim)l and put on ice for 30 minutes followed by freezing at -70"C. This preparation freed intracellular labelled protein whilst protecting it from intracellular proteases during disruption of the cell'

Atter 2 hours, the 1 ml "chase" sample was centrifuged at 1,500 RPM, 3 min and the supernatant aspirated and stored at -70'C for subsequent analysis of labelled crlAT secreted into the extracellular partition. The cell pellet was treated as above.

Total labelled protein in each sample was then quantitated by trichloracetic acid (TCA) precipitation (Roberts & Paterson, 1973; Perlmutter et al, 1984). 10 pl of fluid from each sample was mixed with 500 pl of 10 mg/ml bovine serum albumin (BSA, BRL) and then mixed with 500 pl ZO% TCA and kept on ice, 15 min. The protein was then deposited on a glass microfibre filter (Whatman) using a baffle, suction apparatus (Millipore, Bedford, MA). The filters were washed on the baffle with 5% TCA and 95% ethanol (5 ml of each, 3 times) under suction. After drying, the filters were put in scintillation fluid and radioactivity counted in a Beckman Scintillation Counter. Then, 106 dpm for each intracellular compartment (immediately after the pulse and after the chase) and 102 dpm for the extracellular supernatant was immunoprecipitated with the rabbit anti-human ø1AT antibody coupled to cyanogen bromide-activated sepharose 48 beads (Pharmacia)' The calculated amount of cell lysate or supernatant was added to 50 pl of 10 x methionine immunoprecipitation buffer (1.5 M NaCl; 0.5 M Tris, pH 7.5; 50 mM EDTA; 0.5% Nonidet P-40; 0.1"/o SDS; 0.5% sodium deoxycholate; 0.2% sodium azide; 10 mg/ml BSA; 10% 100 x aminoacids mixture; 30 mg/ml methionine), and 100 pl of antibody-bead solution. This was mixed al4"C lor 2 hours and then washed 5 times in 1x immunoprecipitation buffer (10 x methionine immunoprecipitation buffer without methionine or aminoacids and diluted to one tenth).

To confirm identity of the immunoprecipitated prote¡n as cr1AT, a "blocking experiment" using excess unlabelled al AT added to and

62. incubated with the anti-human alAT antibody sepharose 48 beads prior to addition of lysates or supernatants was undertaken.

To the final pellet after centrifugation (14,000 RPM, 5 min) was added 50 pl of sample buffer (1.25 ml 0.5% Tris pH 6.8; 2.0 ml 10% SDS; 500 pl B-mercaptoethanol; 1 ml glycerol; 100 pl bromophenol blue; 5.7 ml distilledwater). Thiswas heated 95"C,5 min, centrifuged 14,000 RPM, 1 min. S0 pl of the solution was then aspirated with a fine syringe (to avoid aspirating beads) mixed with 5 ¡rl of 0.1% dithioehreitol (Sigma) and loaded into awell of a7.5/o SDS polyacryamide gel in a glycine running buffer and run at 50 volts overnight (Laemmli, 1970). Radiolabelled size markers (Rainbow markers, Amersham) were run on each side of the gel for accurate sizing of protein bands. The gel was treated with a fluorographic agent according to manufacturers directions (Amplify, Amersham), transferred to filter paper, dried under vacuum and then put in contact with Kodak XAR 5 x-ray film at -70"C overnight, or longer if necessary, to visualize protein bands.

2.8 ANALYSIS OF CTlAT BIOSYNTHESIS USING IN VITRO SYSTEMS

2.8.1 General

ln the case of the crlAT variants for which a homozygote case was not available for analysis, studies ol in vivo biosynthesis of the variant slAT protein in isolation were not generally possible due to olAT production directed by the other olAT allele. To overcome this, in vítro systems were utilized.

2.8.2 Production of Recombinant Plasmids Containing Variant alAT Genes

To create variant cx,1AT cDNAs with the relevant point mutations, the technique of in vitro mutagenesis with PCR of normal crl AT cDNAs was used (Mullis & Faloona, 1987). The normal template crlAT cDNA used for mutagenesis to Create the P¡6,¡s¡¡ or P."¡nl atbans ol AT cDNA was the

63 M1(Vat)zt3 alAT CDNA in the N2-FAT construct (Garver et al, 1987b). This construct is of the early SV40 gene promotor and the crlAT cDNA (pPBO1) in the N2 retroviral vector (Keller et al, 1985; Eglitis et al, 1985). The N2 vector conta¡ns the neomycin resistance gene for selection. The ø1AT cDNA template for the production of the Wbethesda crlAT cDNA by PCR mutagenesis was the M1(Ala)zt3 crlAT cDNA in the pSVL expression vector (Pharmacia) as described by Brantly et â1, (1988b). For mutagenesis, primers were synthesized which corresponded to published sequences on either side of the cloning site of these vectors (N5' and N3' for N2-FAT and P5' and P3' for the pSVL construct - generally called vector 5' and vector 3' or V5' and V3'). As well, complementary primers in the sense (5') or antisense (3') direction with the nucleotide mutation being studied at their centre were synthesized (Puwelr- 5', Plowell-3', Psaint albans-S', Psaint albans-3', Wbethesda-S', Wb"th"sda-3', - generally referred to as mutation 5' and mutation 3' or M5' and M3'). M5' and M3' primers were constructed to be exactly complementary Then, the following PCR reactions were undertaken using 10 ng of appropriate plasmid template in a 1OO pl reaction according to manufacturers recommendations with the following primer pairs -

PCR-1 V5' + M3' PCR-2 M5'+ V3'

The products of these reactions, are each a partial crl AT cDNA which together overlap at the sites of the M5' and M3' primers to form a full length cl AT cDNA. These were then agarose gel purified to remove unused primers and template. The final PCR reaction to anneal the products of PCR-1 and PCR-2 to form a full length mutant alAT cDNA was then undertaken. To do this, equimolar amounts of each of the products of PCR-1 and PCR-2 were used as the template in a reaction using V5'and V3'as primers, i.e.

PCR-3 V5' + V3'

During the first denaturation and annealing of this reaction, a portion of the denatured strands from PCR-1 would anneal to the complementary portion of the single strands of PCR-2 reaction (the site of the M3' and M5' primers). This would allow extension by Taq polymerase to form a full length ol AT cDNA which would be the template for subsequent rounds of amplification with the

64 V5' and V3' primers. The resultant amplification product was then checked for correct length by agarose gel electrophoresis, purified by phenol-chloroform extraction and ethanol precipitation and then digested by appropriate restriction enzymes (Xhol for those based on N2-FAT and BamHl and Pstl for those based on the pSVL construct). These were then ligated into their respective, appropriately digested vectors (Xhol for N2 and BamHl and Pstl for pSVL) by standard means, transformed into competent E. Coli. JM109 bacteria. Positive clones were characterized by restriction digest analysis for overall structure and orientation, ASA with PCR to ensure the desired mutation was present and finally, complete sequence analysis of the amplified fragment to confirm the presence of the mutation and to make sure there were no other mutations caused by misincorporation of nucleotides by the Taq DNA polymerase (Saiki et al, 1988).

Specific sequences of these PCR primers were (underlining = site of mutation) -

N5' 5' - TGCAGGCATCTCATGAGTG - 3'

N3' 5'- GACCTTGCAGCATCGCCG - 3'

P5' - TGCTCCTCAGTGGATGTTG . 3'

P3'- CAAATAAAGCAATAGCATC - 3'

Pbwett 5' 5'- CCTGCCTGITGAGGGGAA - 3'

Pbwetl3' 5'- TTCCCCTCAACAGGCAGG - 3'

Psaint albans 5' 5'- CTGACCATCAACGAGAAAG - 3'

65 Psaint albans 3' s',- CTTTCTCGTIGATGGTCAG - 3'

Wbethesda 5' 5'- CGTGCATAAGACTGTGCTGAC - 3'

Wb"thesda 3' 5'- GTCAGCACAGfCTTATGCACG - 3'.

The cllAT cDNA constructs in the N2 retroviral vector were used for studying in vitrocrlAT biosynthesis in NIH-3T3 mouse fibroblasts and the pSVL constructs for examination of alAT biosynthesis in COS I monkey kidney cells (Garver et al, 1987a; Garver et al, 1987b; Brantly et al, 1988b).

2.8.3 crlAT mRNA Cell Free Translation ln Vitro

To analyze ín vitro expression of the variant alAT genes at the translational level, the variant crlAT cDNAs as well as a normal M1 (Valzts¡ cDNA were cut out of the N2 or pSVL constructs described above and ligated into the pSP64(polyA) vector (Promega). These recombinant plasmids were then used to generate cr,1AT mRNA of the corresponding type using the riboprobe SPG polymerase in vitro transcription system (Promega) (Melton et al, 1984; Kreig and Melton, 1984) exactly according to the manufacturers "Capping Protocol 2" (see package insert for pS64(polyA) vector). Equal amounts (1 pg) of the normal and variant mRNAs were then used in a cell free rabbit reticulocyte lysate in vitro translation system (Promega) according to the manufacturers recommendations using l3sS]-metnionine as a label (Pelham & Jackson, 1976). The resultant labelled translation products were analyzed by SDS polyacrylamide gel electrophoresis, fluorography and quant¡tated by laser densitometry, as described above. Five translation experiments were run for each ø1AT mRNA to compare translational efficiency and results analysed by the two-tailed Students t-test.

2.8.4 Transient Expression of Human ø1AT cDNAs in COS I Monkey Kidney Cells

Biosynthesis of human olAT directed by the crl AT zts¡ Wbethesda allele was compared to that directed by the normal cl AT M1 (Val

66 and deficient g1 AT Z allele utilizing COS I monkey kidneys cells (Gluzman 1981) transfected with the pSVL expression vector constructs containing the respective ol AT cDNAs (Brantly et al, 1988b). Transient expression of human crlAT in these transfected cells was analyzed by pulse-chase protein labelling with ¡esS¡-methionine. COS I cells were grown to confluence in 10 cm tissue culture dishes (Falcon) in DMEM supplemented with 10% foetal calf serum, penicillin/streptomycin and glutamine. These cells were split 1:4 and grown overnight in 1O cm tissue culture dishes. The next morning the COS lcells were transfected with the pSVL constructs by calcium phosphate co- precipitation (Graham & Van der Eb, 1973; Parker & Stark, 1979). 20 pg aliquots of lyophilized plasmid DNA were resuspended in 1 ml 250 mM CaOlz, added dropwise to 1 ml ol 2 x HEPES buffered saline (12.5% of 1M HEPES pH 7.0 (Sigma); 4.1 gm NaCl; 0.1005 gm sodium phosphate dibasic heptahydrate; 237.5 ml HzO) in a Falcon 2059 tube being vortexed at the time, stood at room temperature for 30 minutes and then mixed with 18 ml of DMEM based media as above. 10 ml was laid on top of each plate of COS I cells (after aspirating overnight media) and incubated 37"C , 10o/" COz, 4 hr. After 4 hours, the DNA-media mix was aspirated and 10 ml of glycerol "shock" solution (5 mls 2 x HEPES Buffered Saline; 3 ml 50% glycerol; 2 ml HzO) was carefully added tor 2 minutes, then aspirated. The cells were washed with media and then fresh media was added and the cells incubated,37"C, 10"/" COz,72 hr. One plate of each type of transfectants (M1 [Vat zts¡' Wuetnes¿a, Z) was analysed by cytoblot mRNA analysis as described previously to ensure equal quantities of crlAT mRNA expression in each population The cells were then pulse-chase labelled with [35S]-methionine and analysed by SDS polyacrylamide gel electrophoresis, fluorography and laser densitometry as described previously for blood monocytes except for the following minor modifications. Firstly, a change in the lysis mix as suggested by D.T. Curiel, personal communication, (50 ml PBS; 500 pl Nonidet P4O; 0.25 gm deoxycholic acid; 0.17 gm EDTA; 5 mg Leupeptin; 50 pl 0.1 M PMSF) was made to enhance cell lysis. Secondly, the rabbit anti-human ø1AT antibody was from Boehringer Mannheim. Thirdly, to decrease non-specific protein bands in the SDS-polyacrylamide gel analysis of cell lysates, the CNBr- Sepharose4B-anti'cr1AT beads used to immunoprecipitate the olAT from cell lysates were first incubated with non-transfected COS I cell lysate to block non- specific binding to the beads of non olAT radiolabelled material (D.T. Curiel

67 and M.L. Brantly, personal communication). Experiments were performed on five separate occasions for quantitative analyses of fluorograms by laser densitometry. Statistical comparisons of M1(Valzts¡ transfectants and Wb"th".d" transfectants were done using the two-tailed Students t-test. To ensure that any differences in extracellular crlAT product by these cells was not due to differential turnover or uptake of the cll AT protein in the extracellular supernatant, [ssS1-rethionine labelled crlAT produced by transfected COS I cells was placed on confluent plates of unmodified COS I cells in triplicate for 24 hours and then immunoprecipitated and analyzed by SDS-polyacrylamide gel electrophoresis, fluorography and laser densitometry and compared to an aliquot of equal volume of the same material which had not been exposed to unmodified cells. Statistical comparison was with the two-tailed Students t-test (McOracken et al, 1989).

2.8.5 Permanent Production of Human cr,1AT in Murine NIH-3T3 Fibroblasts by Insertion of a Human alAT cDNA by Retroviral Gene Transfer

Normal or mutant crlAT oDNA N2 constructs were transfected into the g2 mouse fibroblast cells for packaging of the transcripts from the integrated constructs into ecotropic retroviral particles (Miller et al, 1985). This was done as for COS I cell described previously except that the base of the medium was lscove's minimal essential medium (IMEM) rather than DMEM, incubation was in 5% COz and atler 72 hours, the media was changed to one containing 500 mg of G418 (Geneticin, G|BCO, Gaithersburg, MD) a neomycin analogue (Davies & Jimenez, 1980). Colonies resistant to neomycin were pooled after7 to 10 days to harvest infectious retrovirus (Miller et al,, 1985). Transfected92 cells were grown to confluence and changed to non-selective media to eliminate any trace G418. Aller 24 hours, the supernatant was collected and filtered (0.45 pM filter, Nalgene, Rochester, NY). One ml of this material was added to 10 ¡rl polybrene (5 mg/ml, Sigma) and then 9 mls of non-selective media. Serial dilutions were then made (1:10, 1:100, 1:1000) and used to infect a confluent plate of NIH-3T3 cells grown in the same media as q2 cells. After 24 hours cells were changed to selective media. After 7 to 10 days, plates with greaterthan 100 colonies were pooled to create a polyclone for subsequent analysis of al AT production (Garver et al, 1987b; Curiel et al, 1989b). NIH-3T3 polyclones of each variant to be

68 compared were shown to express equal amounts of ø1AT mRNA by cytoblot mRNA analysis of equal numbers of cells, hybridization to a [32P]-CTP labelled alAT çDNA probe, autoradiography and quantition by laser densitometry as described previously. olAT protein expression was analyzed as described for transfected COS I cells previously except for the pulse labelling period being only 30 minutes.

69 CHAPTER 3

ALPHA 1'ANTITRYPSIN Nullqranite farrs : THE INTRACELLULAR MECHANISM CAUSING THE ABSENCE OF DETECTABLE ALPHA l.ANTITRYPSIN

3.1 INTRODUCTION

crlAT Null alleles have, by definition, absence of detectable alAT in the serum attributable to that allele (Talamo et al, 1973; Cox et al ,1980; Fagerhol & Cox, 1981). Studies of individuals carrying crlAT Null alleles have revealed that the Null phenotype can be caused by diverse mutations of the alAT gene (Crystal, 1990). The known Null mutations include substitution, deletion, or insertion of one or more nucleotides in the four crlAT protein coding exons or deletion of all of the protein coding exons of the crlAT gene (Table 1.3). The intracellular mechanisms by which those mutations cause the absence of crlAT production can be grouped into two categories : those in which crlAT mRNA transcripts are present in crlAT synthesizing cells and those in which crl AT mRNA transcripts are not detectable.

The study in this Chapter capitalizes on the rare occurrence of a patient 160 homozygous for the Nullsranite 1¿¡¡s rnutâtion (Tyr TAq, C deletion,5' frameshift ,, Stop 160 TAG [see Results]). The presence of a homozygote for this variant glAT gene permitted analysis of the intracellular mechanism resulting in absence of detectable al AT attributable to this allele.

3.2 M ETHODS

The lndex case, a 38 year old Caucasian woman, was referred to the Pulmonary Branch, NHLBI, when screening investigations performed by a

70 local medical practitioner after she had presented with a two-year history of dry cough and dyspnoea on exertion, suggested absence of detectable slAT in her serum. There was only a 1 pack-year history of smoking with discontinuance one year previously. Routine investigations as previously outlined were undertaken and the patient underwent fibreoptic bronchoscopy and bronchoalveolar lavage to harvest alveolar macrophages for subsequent analysis of crlAT biosynthesis. Four generations of family members were available for phenotypic and genotypic analysis (Figure 3.1)'

al AT phenotype analysis was undertaken by a combination of al AT serum levels (radial immunodiffusion), IEF at pH 4 - 5, IEF with immunofixation and family analysis.

crlAT genotype analysis was undertaken by ASA with PCR. Since glAT phenotype analysis of this family demonstrated the presence of the Null, M1 and M2 alleles (see "Results"), and with the knowledge that there is only one known crlAT allele that migrates in the M2 position on lEF, the identification of the M2 allele in the family is unequivocable (Fagerhol and Cox, 1981 ; Nukiwa et al, 1988). However, there are nine known mutations causing the Null state (see Table 1.3), and two defined mutations yielding alleles migrating at the M1 position on IEF (Nukiwa et al, 1987a) and therefore for convenience initial screening genetic analyses of this Null allele utilized ASA primers for only 3 of the Null alleles: the Nullsranire faus, Null¡1¿¡¡¿,¡., and Null66¡¡¡¡s¡", mutations. The primers and a schematic for the Nullsranire fals mutation analysis by ASA with PCR are shown in Figure 3.2.

The other Null mutation ASA primers are as follows (Okayama et al, 1989a) -

a. Null¡¿¡¿'¡y¿

Null¡¿x¿yy¿ normal sequence ASA primer (sense primer) 5' - TGCTGGGGCCATGTTTTTA . 3'

Null¡¿n¿yy¿ mutation ASA primer (sense primer) 5'- TGCTGGGGCCATGTTTTTT - 3'

71. FIGURE 3.1

Pedigree of a FamilY with the cxl AT Nullsranite ralls Allele

The four generations evaluated are depicted on the left (l-lV). Below each family member is a number assigned to the individual within the generation. Below that is the slAT genotype determined by ASA with the PCR in levels combination with phenotypic analysis by IEF of serum and serum alAT (see "Methods" and Figure 3.3). Below the genotype is the serum ol AT level by (pM). The lndex case Nullnranite ratts homozygote (individual lll+) is indicated an arrow. Shading denotes the presence of the Nullsranite falls lNullsr) allele' I

1 2 M1(Ala213)M1(Val213) M1(Ala213)Nullel 33 21 II

1 2 M2Nullnr M1(Val213)Nullsf 21 20

III

1 234 5678 M1(Ata213)M2 Nulln¡Nullnr M1(val2t3)M2 NullnlNulln¡ Null¡Nullnr M1(val2ß)Nulls, Nulln¡Nulln¡ M2Noullor 39 0 34 0 / IV

2 3 M1(Ala213)Nullsl M1(Ala2r3)Nullof M1(Ala213)Nullqf 19 2't 20 FIGURE 3.2

Strategy for Genotyp¡c _4lqlysis of the NUllsranite talls Allele

Shown is the alAT gene with its Seven exons (ln - c, ll-V), six introns, and 5' and 3' flanking regions. Above exon ll are the nucleotide and amino acid Sequences for the normal M1(Al¿zts) allele and the Nullsranite 1¿¡¡s allele at the site of the Nullsranire ra¡s mutations. Below exon ll of the crlAT gene is an enlarged view of exon ll depicting the site of the Nullsranite falls mutation. ASA primers were constructed such that the 3' end of each primer (underlined) corresponded to either the normal sequence at the Nullsranite tatr nìutâtion site (primer "MGF") or to the Nullgranire rals mutation (primer "GF"). Below the enlarged exon ll are the sequences of the normal (Mer) and Nullsranite talls (GF) ASA primers, and the common 3' distal extension primer (identical for amplification of both the normal and Nullsranire rals segments)' lf the sequence of the genomic DNA is identical to the 5' primer, the amplified segment of 0.49 kb will be observed; if this sequence does not match at the 3' terminus of the 5'primer, there will be no amplification. Normat M1(Ata213) AAA CAG ATC AAC GAT TAC Lysttt Gln1s6 ¡¡"157 ¡tn158Asp15eTyr160

NUllgranite fails AAA CAG ATC AAC GAT TAG Lystuu Glnl56 ¡¡"tsz AtntssAsptsestopl60

ln le lc tt rr IVV 5r 3

of Nu

5', il* 3'

kb 5'primer 3'primer W+ <Ø I I Normal 5' primer (M6r) I I I I 5' AGGCCAAGAAACAGATCAACGATTAC 3' *l Common 3'primer l+ 3' TGAGATCACAGGGTCTTAGTCAT 5' Nullsranne r",,. 5' Primer (GF) i 5' AGGCCAAGAAACAGATCAACGATTAG 3'+1 Null¡¿¡¡sì,r common distal primer (anti-sense primer) 5' - CAGAGAAAACATGGGAGGGATTTACA - 3'

Generating a 265 base pair amplified fragment.

b. Null6s¡¡¡¡s¡.t

Null6s¡¡¡¡s¡am rìotÍìâl sequence ASA primer (ant¡-sense primer) 5'- GCTTCATCATAGGCACCTT - 3'

Null6s¡¡¡¡s¡am mutant sequence ASA primer (anti-sense primer) 5' - GCTTCATCATAGGCACCTA - 3'

Nullsg¡¡¡¡s¡am corlìrrìon distal primer (sense primer) 5' - AAATATGCCTGATGCTCCAAC . 3'

Generating a 444 base pair amplified fragment.

ln addition, each family member carrying an M1 allele on IEF was genotyped at the codon for amino acid213 of crlAT to subtype the M1 allele into M1(Alazls, GCG) or Mr(Val2lg' GTG) the two known variants of M1 (Nukiwa et al, 1987a). Primers used were -

M1(V¿lzts) ASA primer (anti-sense primer) 5' - GCACCTTCACGGTGGTCA . 3'

M1(Al¿zts) ASA primer (anti-sense primer) 5' - GCACCTTCACGGTGGTCG -3'

M1 common distal primer (sense primer) 5' - AAATATGCCTGATGCTCCAAC - 3'

Generating a 330 base pair fragement

Total cellular RNA was extracted from alveolar macrophages obtained at bronchoscopy of the lndex case or an M1(Valzts¡ homozygote control and was examined by Northern analysis for the presence of crl AT or y-actin mRNA transcripts.

74. 3.3 RESULTS

3.3.r Patient Characteristics

Relevant findings on investigation of the lndex case were a 133Xe ventilation scan demonstrating abnormal retention of gas in the bases ¿¡fl esmlc-macroaggregated albumin perfusion scan showing decreased perfusion in the same areas, findings consistent with emphysema associated with crl AT deficiency. Lung function tests demonstrated a vital capacity (VC) of 112% predicted, forced expiratory volume in 1 second (FEVI) o179"/" predicted and FEVr/forced VC ratio ol72o/". A single breath diffusing capacity (corrected for lung volume and haemoglobin) was 98% predicted, and arterial blood gases were normal. No clinical details were available for other family members.

3.2.2 crlAT PhenotyPe AnalYsis

olAT serum levels were undetectable in the lndex case as well as three of her siblings (Figure 3.1) and IEF failed to reveal any ø1AT prote¡n bands in these cases despite enhancement by immunofixation. Otherwise, both the M1 and M2 alAT protein was found in the family and gl AT levels were normal or just below normal (family member lVt , Figure 3.1).

3.3.3 Analysis of crlAT Genotype

Preliminary screening using ASA with PCR revealed that the lndex case (Family member lll¿, Figure 3.3) was a homozygote for the

Nullsranite 1"¡¡s allele and that this allele was present in other family members in eitherthe homozygous state (lllz, llls, lllz) or heterozygous state (lz, llt, llz, lllo, llle, lVr, lVz, lVs). Family members ll, llll, and llls did not carry the Nullsranire rals allele. The Null.ailawa and Null6"¡insham alleles were not detected in this family. Both of the M1 alleles, M1(Valzts) and M1(Al¿zts) were detected by ASA with PCR. The pattern of inheritance of the Nullsranite 1"¡¡s allele in this family was consistent with the known autosomal co-dominant inheritance of this gene (Figure 3.1).

75 FIGURE 3.3

Genotype Analysis of Genomic DNA of Family Membérs for Presence of the Nullsranite ralls Allele

Genomic DNA from each family member was amplified with a 5' ASA primer specific for the normal sequence at the Nullsranite 1¿¡¡e ffiutâtion site (Mor) in conjunction with a common 3' distal extension primer or with a 5' ASA primer 3' specific for the Nullsranire 1¿¡¡s nìutâtion (GF) in conjunction with a common distal extension primer (see Figure 3.2 for details). Samples (10 ¡tl) of each reaction were electrophoresed on a 17o agarose gel with ethidium bromide and photographed under ultraviolet light. Shown is the analysis of each family member corresponding to the generation and individual numbers denoted in Figure 0.1. The expected position of the 0.49 kb amplification product is shown on the left. Those family members without the Nullsranitefatls allele (lt, lllr, llls) show the 0.49 kb amplification product only with normal primer (Mer); those homozygous for the Nullnranire rals allele (lllz, llla, llls, lllz) have an (GF); amplification product only with the Nullsranite ¡¿¡¡s rnutâtion specific primer and heterozygotes fora normal allele andthe Nullsranireraltsallele (lz, llt,lllo, llle, lVr, lVz, lVs) have an amplified fragment with both the Mer and the GF primers. Similar analyses were carried out using allele specific primers for the M1(Al¿zrs) and M1(V¿lzt3) mutation site, permitting correct identification of these two normal M1 alleles. Family Member

It 12 IIr Ílz primer Mar GF Mc, GF Mot GF Mcr GF

kb

0.49>

IfIl IIIZ III3 III¿ Primer Mcr GF Mcr GF Mc, GF Mc, GF kb 0.49> ! IIIs III6 IIIz IUs Primer Mor GF Mo, GF Mot GF Mor GF kb

0.49>

IVr IVz IV3 Primer M6¡ GF Mcr GF M.r GF kb

0.49> 3.3.4 Evaluation of a Nullsranite rals Homozygote for crlAT mRNA Transcripts

Northern analysis of total cellular RNA from the alveolar macrophages of the lndex case or an M1(Valzts¡ homozygote volunteer control utilizing either a 32P-labelled olAT cDNA probe or a 32P-labelled T-actin cDNA probe showed that, whereas the normal control alveolar macrophages had crlAT mRNA transcripts of the expected 1.8 kb size, there were no detectable al AT mRNA transcripts in the macrophages of the lndex case (Figure 3.44). In contrast, the alveolar macrophages of both the normal control and the lndex case contained comparable amounts of the 2.0 kb T-actin mRNA transcripts (Figure 3.48).

3.4 DTSCUSSTON

alAT deficiency can be categorized at the protein level into "deficient" variants in which cllAT is detectable in the serum but in reduced amounts and "Null" variants in which no detectable crlAT protein is present in serum, even when using sensitive methods such as enzyme linked immunoassay and IEF with immunofixation (Cox et al, 1980; Brantly et al, 1988a; Crystal et al, 1989; Cox, 1989; Crystal, 1990). Like the "deficiency" alleles forwhich a variety of mutations result in a similar phenotype at the serum level, it is now recognized that a variety of mutations can cause the Null phenotype. At the gene level, the Null mutations include substitutions of bases in a coding exon causing amino acid substitution (Null¡u6wigsharen, Null¡s,,vps¡t) or a premature stop codon (Null6s¡¡¡¡srram), dêletions of one or two bases causing premature stop codons (Nullsranite ta¡s, Nullus¡1s¡, Null¡ons kons), deletion of all 4 crlAT protein coding exons (Null¡ss¡¿ di procida, Null¡g¡6s¡6urs) and an inSertion mutation resulting in a premature stop codon (Null¡1"11"*") (see Table 1.3 for an overall summary).

Since the al AT Null alleles all give rise to an identical phenotype as assessed by serum alAT level and lEF, it is necessary to genotype an individual to asceftain which crlAT Null variant is present (Cox et al, 1980). ASA with PCR provided a convenient and reliable means to do this, avoiding radioactivity and providing a rapid answer compared with either radiolabelled oligonucleotide probe analysis or gene sequencing (Okayama et al, 1989a).

77 FIGURE 3.4

Evaluation of cxl AT Synthesizing Gells of an crl AT Nullsranire ratrs Homozygote for the Presence of crl AT mRNA Transcripts

Total cellular RNA (10 pg) from alveolar macrophages from a normal M1(V¿l2ts) homozygote and the Nullsranire rals homozygote lndex case (family member lll¿, see Figure 1) was analyzed by Northern blot analysis and hybridization to : Panel A - lanes 1,2, a 32P-labelled full length olAT cDNA probe; Panel B - lanes 3,4, a32P-labelled full length y-actin cDNA probe'

Lane 1 M1(Valzts) homozygote, crlAT cDNA probe

Lane 2 Nullsranite t",," (Nullst) homozygote, al AT cDNA probe

Lane 3 M1(Val 213) homozygote, yactin cDNA probe

Lane 4 Nullsranire t",," (Nullsr) homozygote, 1-actin cDNA probe

The positions of the normal ø1AT mRNAtranscripts (1.8 kb) and the y-actin mRNA transcripts (2.0 kb) are indicated' A. B. a1-antitryPsin l-actin

M1(val213) Nullsr M1(Val213) Nullsr Mt (Val213) Nullsr Mt (Val213) Nullsr

kb kb

t.g) 2.O>

1 2 3 4 As the number of known Null alleles increases, this method would become somewhat more tedious, however initial screening, for example, for the M1(Al¿zts) versus M(Valzte¡ or M2 variant as the base allele, would limit the number of Null mutations possible and therefore make the genotype analysis more manageable (see Table 1.3).

The ol AT Null alleles can be further categorized at the mRNA level depending on the consequences of these mutations on gene transcription. ln this regard, like the mutations in the Null variants of the low density lipoprotein receptor gene in familial hypercholesterolemia (Russell et al, 1987), the hypoxanthine-guanine phosphoribosyl transferase gene mutations in the Lesch-Nyhan Syndrome and some forms of gout (Wilson et al, 1986), and the ß-globin gene mutations in ß"-thalassemia (Chang & Kan, 1979; Maquat et al, 1981; Humphries et al, 1984; Takeshita et al, 1984), the Null olAT mutations result in the presence or absence of detectable crlAT mRNA transcripts in olAT synthesizing cells. The present study demonstrates that the premature stop codon in exon ll of the alAT gene in the Nullsranirel¿¡¡s allele is associated with no detectable alAT mRNA in clAT producing cells despite normal amounts of other mRNA species being produced. Thus, of the 9 known

mutations causing the Null state, three (Nullgranite 1¿¡¡s, Null6elingham, Null¡s6¡¿ 6¡ procida) are associated with no detectable alAT mRNA transcripts, two (Null¡onn kons, Nullmattawa) have crlAT mRNA transcripts, and for four (Null6e¡1e¡, Null¡u6*¡tshafen, Null¡¡s6s¡burg, Nullnewporr), the status of the al AT mRNA transcripts is not known (Table 1.3). The Nulluo¡on allele is likely to give rise to an mRNA transcript which is translated into a truncated protein lacking structurally important amino acids (Fraizer et al, 1989b). lt has been suggested that it is likely that the Null¡y6,¡y¡se¡¿1s¡ mutation results in an mRNA transcript but that this is translated into an unstable protein due to the substitution of a polar amino acid for a non-polar amino acid in an alpha helix (Fraizer et al, 1990). Like Null¡"otadiprocida, the very nature of the glAT gene in Nullr¡"¿snuurg doês not allow alAT mRNA transcripts unless they arise from the crlAT-like gene (Poller et al, 1991).

Although it is clear that the Nullsranite 1¿¡¡s Íìutâtion is associated with no detectable ø1AT mRNA transcripts, it is not known why a premature translation stop codon in an alAT coding exon should result in absent ø1AT mRNA. For the ß'-thalassemias, a premature translation stop codon causing

79 the absence of identifiable mRNA has been suggested to be the result of either changed mRNA structure to create or destroy regulator binding sites, intranuclear degradation, abnormal nuclear-cytoplasmic transport of the transcripts ora combination of these (Maquat et al, 1981). lt is possible that due to lack of protection by polyribosomes of a large portion of the 3' end of the mRNA transcripts, the mRNA is subjected to nucleolytic degradation within the cytoplasm (Chang & Kan, 1979; Maquat et al, 1981; Urlaub et al, 1989). ln prokaryotes, the effect of premature translation stop codons or non-sense mutations on mRNA levels depends on the site of the mutation, the 5' mutations reducing mRNA more than 3'mutations (Adhya et al, 1978). ln keeping with this in eukaryotes, Urlaub et al, (1989) have found that non-sense mutations in the dihydrofolate reductase gene have much more effect on mRNA levels if in the 5' exons versus the 3' exons. ln addition, these stud¡es pointed to the premature stop codons having a major effect on nuclear to cytoplasmic mRNA transport to result in low mRNA levels. lnterestingly, Daar & Maquat, (1988) found similar polarity effects of non-sense mutations in the triosephosphate isomerase gene, but concluded that the mutations causing low levels of mRNA did so by altering mRNA stability by affecting the length of translated mRNA, or in the case of a single mutant, by altering the mRNA structure. Therefore, it seems likely that a number of mechanisms may cause low mRNA levels associated with non-sense mutations. Finally, it is possible that with more sensitive techniques such as PCR of reverse transcribed mRNA, that those ø1AT Null alleles which are presently categorized as being associated with absence of crlAT mRNA may later be found to have some ol AT mRNA transcripts albeit at much lower levels than normal.

The Nullnranite rals homozygote lndex case had evidence of lung disease with a reduced FEVr and ventilation and perfusion lung scan showing gas trapping and diminished vascularity in the lung bases. That the lung disease was only mild despite total lack of c1AT, and therefore absence of the major antiprotease of the lower respiratory tract, illustrates the variability of the lung disease associated with olAT deficiency and highlights the importance of exogenous factors such as significant smoking in developing lung disease in ctlAT deficiency (Larsson, 1978, Cox & Levison, 1988). lnterestingly, a study of two sisters with the Null¡¿x¿ry¿, Null6s¡¡¡.gham 9êrìotype revealed that one, a 39 year old, had a 1O year history of severe lung disease, compared with her sister a 31 year old with lung disease of only similar severity to the lndex case

80 in this present study (Curiel et al, 1989b). The presence of measurable lung disease in the lndex case would be adequate grounds for cr,1AT replacement therapy as an attempt to halt excessive lung destruction and future debilitation (Wewers et al, 1987b; Buist et al, 1989)

81. CHAPTER 4

THE MOLECULAR BASIS OF THE LUNG AND LIVER DISEASE ASSOCIATED WITH THE ALPHA l.ANTITRYPSIN DEFICIENCY ALLELE Mmatton

4.1 INTRODUCTION

olAT deficiency is associated with an increased incidence of both emphysema and, to a lesser extent, liver disease (Laurell and Eriksson 1963; Sharp et al. 1969). lmportantly, the specific alAT phenotype dictates whether the al AT deficient individual is at risk for emphysema only or emphysema and liver disease (Carrell & Owen , 1979; Carrell et al. 1982; Cox et al. 1981 ; Crystal et al. 1989; Crystal, 1990). Whereas emphysema may occur with any combination of olAT deficiency or Null alleles associated with serum levels of alAT <11 pM, the liver disease has only been associated with a small subset of crlAT deficiency alleles and only those alleles associated with crlAT accumulation in hepatocytes (Cox et al. 1981; Reid et al 1987; Crowley et al. 1987; Cox 1989; Crystal 1990).

The study described in this Chapter characterizes the Mmatton 6¡l AT allele, an allele with a mutation which, like the common ZaIAT mutation, is associated with both ol AT deficiency and liver disease (Cox 1975; Roberts et al. 1984; Allen et al. 1986; Reid et al. 1987). ln addition, the rare occurrence of an individual homozygous for the M¡¿¡sn allele allowed analysis of crl AT biosynthesis in crlAT synthesizing cells obtained from this individual to aid in understanding the pathogenesis of the liver disease associated with this ol AT deficiency allele.

82. 4.2 METHODS

The lndex case, a 47 year old male Caucasian presented to his local medical practitioner with a 20 pack-year smoking history and several years of increasing dyspnoea on exertion, lnitial investigations indicated reduced olAT serum levels and he was referred to the Pulmonary Branch, NHLBI for further investigation. Routine investigations were undertaken as previously outlined. ln addition, the patient underwent fibreoptic bronchoscopy to obtain alveolar macrophages as well as percutaneous liver biopsy after screening liver enzymes suggested an abnormality. Two generations of family members were available for phenotypic and genotypic analysis.

cllAT phenotype analysis of the lndex case and available family members was undertaken by a combination of crlAT serum levels (radial immmunodiffusion) and IEF with an immobilized pH gradient of 4.45 - 4.75.

Evaluation of the function of the Mmatron crlAT was carried out as previously described after purifying the Mmafton ol AT from the patient's serum.

Genotype analysis of the lndex case was achieved by cosmid cloning of genomic DNA followed by sub-cloning into the plasmid vector pUC19 to enable nucleotide sequencing using the di-deoxy chain termination reaction.

Genotypic inheritance of the Mmatron allele in the family of the lndex case was accomplished using PCR amplification of exon ll of the crlAT gene with subsequent analysis of the amplified DNA using [aszP]-ATP 5' end- labelled 21-mer oligonucleotide probes synthesized to correspond at their centre to either the M¡n¿¡¡en mutation or the normal sequence at the site of the Mmafton mutation (See Results).

To evaluate the consequences of the Mmalton mutation at the mRNA level, crlAT mRNA transcripts in alveolar macrophages from both the lndex case and a normal control were evaluated by Northern blot analysis to size the ol AT mRNA and cytoplasmic dot hybridization for quantitat¡on.

83 The biosynthesis of crlAT directed by the Mmahon gene was evaluated in blood monocytes obtained by monocytapheresis of the lndex case by pulse- chase protein labelling with ¡ssgl methionine and immunoprecipitation of labelled crlAT with an anti-cr,1AT antibody followed by SDS-polyacrylamide gel electrophoresis and fluorography. A normal control was used for comparison. In addition, parallel analyses were carried out utilizing an anti- human lysozyme antibody.

4.3 RESULTS

4.3.1. Patient Characteristics

The presence of emphysema in the lndex case was demonstrated by physical examination revealing : hyper-resonance to percussion and diminished breath sounds on auscultation at both lung bases; a chest x-ray showing bilateral diaphragmatic flattening and loss of vascular markings in the lower lung fields' ¿ 133f,s ventilation scan showing abnormal retention of gas in the lung bases; ¿ eemJç-¡acro-aggregated albumin perfusion scan demonstrating decreased vascularity in the same distribution; and pulmonary function tests revealing a VC ol 75/" predicted, a total lung capacity of 1 11% predicted, an FEVr of 68/" predicted, an FEVI/forced VC ratio of 69% predicted, and a diffusing capacity (corrected for volume and haemoglobin) of 70"/" predicted.

The presence of hepatic inflammation was documented by abdominal ultrasound demonstrating a diffuse echogenic pattern within the liver, a serum alanine aminotransferase level of 58 (normal is 3-44 units/litre), a serum aspartate aminotransferase level of 39 (normal is 8-31 units/litre), and a percutaneous liver biopsy showing portal inflammation (Figure 4.14). ln addition, liver biopsy specimens revealed abundant hyaline globules within peri-portal hepatocytes which stained with PAS stain and were resistant to digestion with diastase, characteristics of globules comprised of cr1AT. Furthermore, immunohistochemical analysis for crlAT using avidin-biotin- conjugated peroxidase labelled crlAT antiserum confirmed that the intrahepatocyte globules were crlAT (Figure 4.18).

84 FIGURE 4.1

Evaluation of Hepatic Tissue of an Mmarton Homozygote

Liver tissue was obtained from the Mr"¡on lndex case by percutaneous liver biopsy.

A. Portal inflammation and diastase resistant globules within periportal hepatocytes after staining with Periodic acid-Schiff and digestion with diastase (x 250).

B. lmmunohistochemical identification of cr,l AT demonstrating localization within intrahepatocyte globules (x 630). A. '\' -tr¡ -'.-'*ì¡F,þ ç - È-4.' t7: ¡ SL¡.

ea ;{ I - .'

Iz 4.3.2. ldentificat¡on of the crlAT M6¿¡¡s¡ Prote¡n

The M¡¿¡on crlAT protein was ¡dentified in the lndex case and his children (Figure 4.2) by its behaviour in an IEF gel with an immobilized pH gradient in which it migrated in a position cathodal to the normal M2 type alAT (Figure 4.3). lmmunofixation of the immobilized pH gradient gel confirmed the observed protein bands were crl AT (data not shown). Serum levels of ø1AT in the lndex case were only 2 pM (normal 20- 53 ¡rM), confirming the M,n¿¡on allele to be a severely deficient al AT allele.

4.3.3. Functional Activity of glAT Mms¡16¡

When Mmafton al AT was evaluated for functional activity, it was apparent that this variant olAT was as competent as an inhibitor of NE as was the normal M1(Valzls) type alAT (Mmalton 9.5 + 0.3 x1Q6 [$-1 s-1,

M1 (Val2t3) 9.3 + O.2 x 1 06 M-1 s-l, p > 0.5) (Figure 4.4).

4.3.4. ldentification of thê M¡¡¿¡1q¡ Mutation

Sequencing of 150 base pairs 5' to exon lc, exons lc-V, all exon-intron boundaries, and the 3'-flanking region of the cloned Mmatron gene of the lndex case demonstrated three differences from the common normal M1 (Valzts¡ crlAT gene. The M¡n¿¡1on gene contained two single point mutations (Glusz6 GAA ,, Asp GAÇ. and Arg1o1 CGT " His CAT) which are the mutations that distinguish the normal M2 slAT allele from the normal M1(Valzts) allele (sequence data not shown; Nukiwa et al, 1988). ln addition, the Mr"¡s¡ gênê was characterized by a triple nucleotide deletion in exon ll at the codon position corresponding to Phes2 (Figure 4.5). This causes an "in-phase" 5' frameshift resulting in an crlAT mRNA coding for an crlAT protein deficient in a single amino acid (i.e. an ø1AT molecule identical to the normal M2 molecule without Phe52). Together the sequence data demonstrated that the Mmatron mutation is a deletion mutation of an entire codon occurring on a background of a normal M2 allele.

87. FIGURE 4.2

Pedigree of a Family Carrying the crl AT Mmarton Allele

alAT phenotypes were determined by a combination of lEF, alAT serum levels and pedigree analysis. The lndex case is indicated by an arrow. The alAT phenotype is listed below each family member together with the serum cr,1AT level (pM) in boldface. Shading denotes the presence of the M¡¿¡1on allele. 1 2 MmaltonMmalton M1M2 2 35 / il

1 2 3 M2M.aron M1M-aron Ml Mmatton 25 23 25 FIGURE 4.3

IEF pattern of crl AT Mmarron in Serum

IEF was carried out using an immobilized pH gradient (pH 4-45 - 4.75)- The anode (+) is at the tope and cathode (-) at the bottom. The left-hand markers indicate the two major (M+ and Mo) IEF bands for the M1, M2 and M3 crl AT proteins. The phenotypes are indicated at the top of each lane.

Lane 1 M1 M1 Lane 2 M1 M2 Lane 3 M1 M3 Lane 4 M1M¡¿¡6¡ (family member llz) Lane 5 MmahonMm"¡g¡ (familY member lt) Lane 6 M2Mmahon (family member llt)

The right-hand arrows indicate the migration positions of the two major bands of the Mmatron protein. Note that the 4 and 6 bands of the Mmalton alAT (Lane 5) are more cathodal than M2. II" Il rIt M1M1 M1M2 M1M3 rvriv-;,' MmaltonM malton M2Mmutton

@

M4 + malton

1- M6 |= + malton

o 123456 FIGURE 4.4

Time-Dependent lnhibition of Neutrophil Elastase bY crl AT Mmarton

Purified qlAT M1(Valzt3) (o) or glAT Mmalton (') was incubated with an equivalent amount of neutrophil elastase for the times indicated and the residual elastase activity quantified. Shown for each time point is the mean È standard error for triplicate determinations. % lnhibition of neutrophil elastase oJ ol\) oooÀo)@ o

(tl

oJ I 3 o l J = 3= Ot gL o_ À' o (¡ ¡

q) O

oo) FIGURE 4.5

ldentification of the Mmarton Mutation ¡n Exon ll bY Sequence AnalYsis

shown are autoradiograms of sequencing gels. on the left is a cloned four M1(Val 213) allele and on the right the cloned Mmafton gene. ln each, the lanes representing the four nucleotides G, T, A and C respectively are shown' The exon ll nucleotide sequences are indicated as are the corresponding sequences for amino acids 49-55 on the left and 49-54 on the right. A triple nucleotide deletion (arrow) at the codon position for amino acid 52 the CharaCterizeS the Mmatton mutatiOn. The M,n"¡1s¡ rnUtâtiOn oCcurs on background of a normal M2 allele. ,-A Ar AI Asn 49 49 Asn L+ Ml(Val21s) Mmâkon 1J A-r GTAC GTAC T 50 50 lle cJltte T1 T lPhe 51 51 Phe 6J Deletion of normal 52 Phe TTC Phe52 Tr 53 Ser C lser se ç-t rC Cr 54 Fro C lPro 53 lcL¡ ¡J rG G 55 Val T v"r s+ L¿ G ] 4.3.5. Inheritance of the Mmatle¡ Gene

Evaluation of exon ll of the alAT genes amplified from genomic DNA by PCR from each family member of the lndex case and the lndex case himself using synthetic oligonucleotide gene probes centred at the region of exon ll coding for Phe52 and specific for the Mmalton mutation sequence or the corresponding normal sequence (Figure 4.6) demonstrated that the triple nucleotide deletion of the Mmatton gene had been inherited in an autosomal fashion (Figure 4.7). Amplified genomic DNA of the lndex case (lt) exhibited specific hybridization with thê M¡¿¡1on 52 region probe but not the normal 52 region probe indicating that this individual carried two copies of the Mmahon gene. The progeny of the lndex case (Family members llt, llz, lls) were demonstrated to be heterozygous for the Mmatton allele, exhibiting specific hybridization with both the Mmatton 52 region probe and the normal 52 region probe.

4.3.6. Evaluation of olAT mRNA Transcripts from Cells Expressing the Mmatton Gene

Northern blot analysis of the mRNA extracted from alveolar macrophages of the lndex case showed that the Mr"¡on ø1AT mRNA transcripts were comparable in size to alAT mRNA transcripts from alveolar macrophages of a normal M1(Valzts) homozygote control case (Figure 4.8) i.e. the triple nucleotide deletion in the olAT Mmatron gene does not appearto affect the normal splicing and processing of the transcript. Furthermore, quantification of olAT mRNA transcripts by cytoplasmic dot hybridization demonstrated that crlAT-expressing cells of the lndex case contained levels of slAT mRNA transcripts comparable to cells of the normal M1(Valzts¡ homozygote control (Figure 4.9). Thus, the abnormality in crl AT gene expression of the Mmatron crlAT gene does not appear to be at the level of mRNA stability, suggesting that the basis of the reduced serum alAT levels must be manifested at the translational or post-translational levels.

92 FIGURE 4.6

Strategy to Delermine Genotypic Inheritance of the Mmarton Allele

Synthetic oligonucleotide extension primers flanking exon ll of the alAT gene (exon ll 5' (+) and exon ll 3' (-) ) were utilized to amplify genomic DNA of 32P-labelled members of the family with the Mmahon allele by PCR. synthetic 21-mer oligonucleotide gene probes centred in exon ll about the M¡¿¡16¡ mutation or corresponding to the normal sequence in the region coding for amino acid residues 49-55 were used to evaluate the amplified DNA. Shown are the nucleotide sequences of the probes as well as the corresponding amino acid sequences. The triple nucleotide deletion in the Mmafton Sequence is underlined. bp = base pairs. I 13' ( :GGAGAGTTCAAGMCTGATGGTTTGAG 1r5',( + ):TCAGTGTTACTGATGTCGGCAAGTACT - )

870 bp

exon ll 3', 5'

3', s', 5' 3', il3'( - ) ils',(+ )

ì l' Asnae lleso Phesl Phes2 Ser53 Pror Valss TTC TCC CCA GTG Normal Probe'52 region NNT ATC TTC ' TTC TCC^CCA GTG AG Mmarton probe 52 region C AAT ATC prous -. fnr".t îrnts'i¡"-so Þi-,ã"t Sãit' Valsa Ser55 FIGURE 4.7

Demonstration of Genotyp¡c lnheritance of the Mmatton Allele

Strategy is depicted in Figure 4.6. Shown is the analysis of individual family members in the study kindred evaluated with either the normal 52 region probe or the Mmatton 52 region probe. The numbers referring to the family members are the same as in Figure 4.2. Probe

Family Normal Mmalton Member Phenotype 52 region 52 region

l1 MmaltonMmalton

ll'' M2Mmanon

I, MlM-¿¡en

lls Ml Mr¡¿¡¡6n

l2 Ml M2 FIGURE 4.8

Qualitative Analysis of Mmarton cxl AT mRNA TranscriPts

was Total cellular RNA (10 ¡rg/lane) extracted from alveolar macrophages 32P-labelled evaluated by Northern blot analysis and hybridization to a human crlAT cDNA probe. Shown is an autoradiogram of this analysis'

Lane 1 M1(Valzts) homozygote Lane 2 Mmahon homozygote lndex case

The size of the normal macrophage ø1AT mRNA transcript (1.8 kb) is indicated on the left. kb = kilobases M1M1 MmattonMmatton

kb

1.8>

1 2 FIGURE 4.9

Quantitative Analysis of Mmatton crl AT mRNA TranscriPts

Equal numbers of alveolar macrophages (5 x 1Oo cells) of either an M1 (Valzts¡ homozygote or the Mmatton homozygote lndex case were evaluated by 32P-labelled cytoplasmic dot-blot and hybridization to a human alAT cDNA probe. Densitometric quantification of autoradiographs (arbitrary units) is shown with an example of an autoradiogram of a serial dilution of cytoplasmic RNA in a dot-blot in the inset' ctlAT mRNA transcripts (ct1AT mRNA/1OG cells)

tu (¡) 5('l o)

!lr E3Or f

3 L o = f =

3 D- o f

3 D- o f 4.3.7 . Synthesis and Secretion of cr,lAT by Blood Monocytes Expressing the M¡s¡¡6¡ Gene

Evaluation of blood monocytes of the lndex case demonstrated an abnormal intracellular accumulation of crlAT during biosynthesis and secretion of reduced amounts of alAT (Figure 4.10)' Labelling of monocytes of an M1(Valzts¡ homozygote with [35S]-metnionine for t hour followed by a2 hour chase period demonstrated secretion of a normal 52 kDa mature form of cr,1AT specifically immunoprecipitated by an anti-cr1AT antibody (Lane 1). Likewise, monocytes of the homozygous M¡¿¡1sn lndex case also secreted a52 kDaform of clAT but in markedly reduced amounts (Lane 2). As a control, evaluation of the supernatants for the secretion of lysozome, another glycoprotein secreted by monocytes, demonstrated comparable levels of secretion by Mmatron cells compared with M1(Valzts¡ homozygote cells, i.e. the secretory defect manifested by monocytes of the Mmahon homozygote is restricted to cr1AT.

To evaluate the basis of this reduced secretion, the intracellular form of the newly synthesized olAT was examined. Analysis of cellular lysates of an M1(Valzts) homozygote immediately after a t hour pulse labelling with [ss51-methionine demonstrated that the normal intracellular form of a 1AT was dominated by a 50 kDa molecule specifically immunoprecipitated with an anti-a1AT antibody (Figure 4.10, Lane 3). This intracellular species of crlAT corresponds to the normal immature precursor form localized to the RER and characterized by high mannase oligosaccharide side chains (Lodish & Kong, 1984). ln the normal cells, following a 2 hour chase period, this precursor form of ol AT is not detected within the cells (Lane 5) i.e. it has been converted to the mature form of the protein with complex carbohydrate side chains and secreted (Lodish & Kong, 1984; Perlmutter et al, 1985b). ln marked contrast, whereas monocytes of the lndex case also exhibit a 50 kDa intracellular form of alAT after a t hour pulse labelling in amounts comparable to normal monocytes (Lane 4), aÍler a 2 hour chase period most of the olAT remained within the cell (Lane 6). Thus, the Mmatron protein accumulates abnormally with the cell during biosynthesis prior to secretion, explaining the reduced cellular secretion by o1 AT synthesizing cells.

97 FIGURE 4.10

Analysis of crl AT BiosYnthesis Associated with the cxl AT Mmarton Allele

Synthesis and secretion of olAT by monocytes of the Mt¿¡on homozygote lndex case compared to an M1(Valzts) homozygote control' shown are fluorograms of SDS-polyacrylamide gel analysis ot 1ssS¡-methionine labelled crlAT immunoprecipitated from supernatants (extracellular) or from cell lysates (intracellular) of monocytes that had been pulse labelled for t hour period with label with [ssg¡-methionine followed by a zero - or 2 hour chase free media. lmmunoprecipitations were carried out on samples of cell supernatants containing 1Oo dpm or cell lysates containing 102 dpm total TCA precipitable 3sS-labelled protein.

Lane 1 M1(Valzts) homozygote, extracellular 2 hour chase Lane 2 lndex case, extracellular 2 hour chase Lane 3 M1(Valzts) homozygote, intracellular, 0 hour chase Lane 4 lndex case, intracellular 0 hour chase. Lane 5 M1(Valzts) homozygote, intracellular, 2 hour chase Lane 6 lndex case, intracellular 2 hour chase

The positions of the normal 52 kDa plasma cl AT and the 50 kDa intracellular form of crlAT are indicated. The lower molecular mass band (43 kDa) is a non-specific band unrelated to alAT and not blocked by excess unlabelled o1 AT Extracellular lntracellular

t=0hr L=2hr

M1M1 MmattonMmarton M1M1 MmattonMmatton M1M1 MmattonMmatton

kDa 52+ ìt 50+ 5 lfrD - O. 43+ r.-tlD-

2 3 4 5 6 4.4 DISCUSSION

The presence of lung disease in the Mmatton homozygote lndex case is not unexpected given the profound deficiency of ø1AT in his serum and the significant smoking history and is in keeping with previous descriptions of lung destruction associated with this allele (Cox, 1975; Sproule et al, 1983; Allen et al, 1986). In addition, consistent with earlier reports, the lndex case had evidence of liver damage in association with accumulation of crlAT within hepatocytes (Reid et al, 1987).

Whereas the pathogenesis of the lung destruction consequent upon serum alAT deficiency is understood to result from insufficient crlAT in the lower respiratory tract to protect the alveolar walls from damage by NE (Gadek et al, 1981a; Janoff 1985), the pathogenesis of the liver disease associated with a subset of deficiency alleles is less well understood. lt has been suggested that the liver damage is due to the lack of protease inhibitor per se causing exposure of the liver to absorbed intestinal proteases (Udall et al, 1982). Alternatively, and more widely accepted, because the alAT alleles associated with liver disease are also associated with intracellular crl AT deposits within hepatocytes, it has been hypothesized that these alleles cause liver disease by directing synthesis of a variant protein which accumulates within cells and that this accumulated olAT damages hepatocytes (Gordon et al, 1972; Errington et al, 1982; Sharp, 1982; Roberts et al 1984; Eriksson, 1985; Carrell, 1986; Reid et al, 1987; Crowley et al, 1987). This latter hypothesis clearly fits for the common Z mutation in which abnormal biosynthesis of the aberranlZ type olAT has been demonstrated to result in intracellular olAT accumulation (Foreman et al, 1984; Perlmutter et al, 1985b; Verbanac & Heath, 1986), and this intrahepatocyte al AT accumulation has been demonstrated to be aetiologic of the associated liver injury in transgenic mice carrying the human crlAT Z gene (Dycaico et al, 1988; Carlson et al, 1989). The rarity of other crlAT mutations associated with liver disease has prevented the investigation of these naturally occurring mutations to add suppot"t to this concept.

Taking advantage of the identification of the rare homozygous inheritance of the ø1AT Mmaron allele, the study presented in this chapter

99. confirms the association of this rare deficiency allele with emphysema and liver disease, characterizes the function of the Mmafton crlAT protein, defines the M,n¿¡sn gene mutation, demonstrates that the allele is transcribed normally and proves that cells synthesizing the M¡¿¡1on protein have an abnormal intracellular accumulation of newly synthesized c1AT. This latter observation supports the concept that the accumulation of crlAT in crlAT synthesizing cells and engorgement of those cells with ø1AT is a likely cause rather than an effect of cell damage (Carrell, 1986).

ln addition to the data presented here, further studies of the Mmalton allele and alAT protein biosynthesis directed by it undertaken principally by Dr. D. Curiel using pulse-chase protein labelling of monocytes obtained from the Mmatron homozygote lndex case in the presence or absence of endoglucosaminidase H suggested the accumulation of the Mr"¡on alAT was at the level of the RER (Gross et al, 1983; Curiel et al, 1989c). Furthermore, in yifro studies of the M¡¿¡1en allele in mouse NIH-3T3 cells infected with the N2 retroviral vector containing either a normal crlAT cDNA or an Mmalton crlAT cDNA confirmed that the deletion of Phesz in crlAT gives rise to the crlAT secretory defect observed in vivo (Curiel et al, 1989c).

For the Z alAT allele, it has been suggested that the amino acid substitutiorì, Glu34z " Lys (Jeppsson, 1976; Yoshida et al, 1976; Carrell et al, 1982) disrupts an important intramolecular salt bridge (Glus¿z " Lys2so¡ leading to abnormal folding of the molecule into its tertiary structure during synthesis (Perlmutter et al, 1985b; Carrell ,1986; Brantly et al, 1988b). Consequently, there is impairment of transport of the newly synthesized Z molecule from the RER either because the slowly folding Z molecules aggregate through exposed hydrophobic residues normally hidden within the molecule or because the alteration in the three-dimensional structure itself prevents translocation. Recent in vitro experiments suggest that this is unlikely to be the whole explanation (Foreman, 1987; Sifers et al, 1989d; McQracken et al, 1989; Wu & Foreman, 1989) and suggest the possibility of a defect in receptor mediated transporl ol Z a1AT. Whatever the mechanism, the result is intracellular accumulation of the immature protein and reduced secretion of cr1AT.

1 00. Unlike lhe Z mutation, analysis of the likely three-dimensional structure of the Mmahon protein does not lead to clear hypotheses about the mechanism' of intracellular accumulation. Phe52 is a highly conserved amino acíd residue amongst the serpins, suggesting it is important (Huber & Carrell, 1989). Studies of other secretory proteins that are abnormally retained with the RER of synthesizing cells have delineated at least two different mechanisms of aberrant biosynthesis leading to intracellular accumulation; impaired cleavage of the amino-terminal signal sequence (Schauer et al, 1985; Lodish 1988) or alteration of the conformation of the newly synthesized protein (Gething et al, 1986; Copeland et al, 1986; Lehrman et al, 1987; Lodish, 1988). For M,n"¡on, the mutation causes a deleted residue in a hydrophobic region within the centre of the olAT molecule in B-sheet B (Loebermann et al, 1984; Huber & Carrell, 1989). This site is distant from the signal peptide cleavage sequence of the crlAT protein (Long et al, 1984; Loebermann et al, 1984), and as the newly synthesized crlAT in the cells of the M¡¡¿¡1e¡ homozygote was not larger than normal as would be expected if the signal peptide was retained, it is most likely that the abnormal accumulation of the newly synthesizêd M,¡"¡1on alAT derives from perturbation of the molecular conformation, impairing maturation from the RER to Golgi. Huber & Carrell (1989) have suggested this mutation would obstruct conformation of B-sheet B. Thus, despite very different types of mutations, the Z and Mmatron alleles exhibit similar pathophysiology of clAT deficiency at the cellular level.

Since the defects in alAT biosynthesis associated with olAT mutations are manifested in all alAT-synthesizing cells in a similar manner (Foreman et al 1984, Perlmutter et al, 1985b; Mornex et al, 1986), it is reasonable to assume that the pattern of abnormal intracellular accumulation of Mmatron slAT observed in mononuclear phagocytes is responsible for the intrahepatocyte accumulation of cr1AT. Therefore, all available evidence suggests that the accumulation of cr,1AT observed in the liver biopsy of the Mmatron homozygote results from de novo crlAT synthesis, not from accumulation of ctlAT taken up by the hepatocytes or because of co-existing liver disease. This is an important distinction since abnormal crlAT deposits in hepatocytes have been demonstrated to occur non-specifically in the absence of ol AT in association with cirrhosis (Pariente et al 1981),

101 . hepatocellular carcinoma (Palmer & Wolf, 1976), and passive hepatic congest¡on (Klatt et al, 1988). For the Mmahon ø1AT allele, as in the case of the c¡1AT Z allele, the intrahepatocyte ø1AT accumulation occurs as a direct consequence of the aberrant biosynthesis of the alAT protein, i.e. clAT accumulation is a primary event.

Although the Mr"¡1on crlAT protein appears similar to the Z ulAT in regard to the secretory defect causing intracellular accumulation and extracellular deficiency, unlike the Z a1AT, the Mmatron protein functions normally as an inhibitor of NE (Ogushi et al, 1987). This data suggests that the tertiary structure of the small amount of M¡¿¡e¡ protein that escapes the secretory block in crlAT synthesizing cells is not significantly abnormal and is able to act well as a substrate for the active site pocket of NE (Carrell, 1986).

lnterestingly, a recent analysis of an alAT deficiency variant in a Japanese, Mnichinan, has revealed that this crlAT allele also has deletion of the codon for Phes2 but in this case not on an M2 background, but on an M1(Valzt3) background. ln addition, there is another mutation in M¡¡ç¡¡¡¿¡, leading to the amino acid substitution, Glyt+aQGG "" Arg AGG (Matsunaga et al, 1990). This latter mutation which is not in a conserved residue among serpins is not thought to cause the alAT deficiency associated with M¡¡ç¡¡¡¿¡ (Huber & Carrell, 1989). The deletion of the codon for Phes2 must therefore have occurred more than once in evolution, suggesting this area may be one of increased mutational activity. Clinically, the Mnichinan lndex case, unlike the Mmafton lndex case did not have significant respiratory disease, however, liver biopsy revealed accumulation of olAT within hepatocytes as is seen with Mmatton (Matsunaga et al, 1990). Another interesting Japanese ø1AT variant Si¡y"r" is characterized by a nucleotide mutation Very near the M6¿¡1on and Mnichinan variants (Seyama et al, 1991). ln this case, a base substitution causes the change Serss TCC "" Phes3 TTC. The position of this mutation adds further credence to this area being a mutational hotspot. The S¡¡yama variant, like Mmatron is associated with profound crlAT deficiency and intracellular alAT accumulation. The function of the S¡¡y¿6" crlAT has not as yet been assessed.

102 Coincident with the study described here, two other groups have sequenced the Mmatton allele and confirmed the delet¡on of the codon for phenylalanine 52 on an M2 background (Fraizer et al, 1989a; Graham et al, 198s).

103 CHAPTER 5

ANALYSIS OF THE MOLECULAR BASIS OF THE ALPHA l.ANTITRYPSIN DEFICIENCY VARIANT lMoeuresoa

5.1 INTRODUCTION

olAT alleles which put an individual "at risk" for disease are, except for rare exceptions, characterized by deficient crlAT serum levels such that in the homozygous state or certa¡n heterozygous combinations the ø1AT levels are below the 11 pM threshold necessary for protecting the lungs from proteolytic degradation (Fagerhol & Hauge, 1969; Wewers et al, 1987a). The deficiency of serum crl AT of the two most common "at risk" variants, S and Z, result from intracellular degradation or intracellular accumulation of newly synthesized cr1AT, respectively (Carrell, 1986; Brantly et al, 1988a; Cox, 1989; Crystal et al, 1989; Curiel et al, 1989d; Crystal, 1990). Other rarer alleles cause crlAT deficiency by gene deletion, mRNA degradation, or an inability of the cll AT molecule to function normally as an inhibitor of NE (Crystal 1989). The purpose of the study in this Chapter was to analyze the molecular basis of a newly recognised crlAT deficiency variant, Wbethesda. Unlike the c¡1AT

Nullsranite 1¿¡¡s allele and Mr"¡¡on allele studied in Chapter 3 and Chapter 4 respectively, the Wbethesda allele was not discovered in the homozygous state. This necessitated that investigations of the biosynthetic abnormalities of crl AT directed by this allele were studied utilizing in vitro systems to avoid confusion with the other alAT allele carried by the lndex case.

1 04. 5.2 METHODS

The lndex case, a 41 year old Caucasian male mining worker was identified during screening for industrial lung disease. There was no history of lung disease. Three generations of family members were available for phenotypic and genotypic analysis.

crlAT phenotype was undertaken by a combination of crlAT serum levels (radial immunodiffusion), IEF and fami[ analysis. IEF was carried out in thin layer polyacrylamide at pH 4-5, with immunofixation to confirm identity of cr,1AT protein bands. ln addition, crlAT of the lndex case was purified and concentrated to highlight Wbethesda ol AT protein bands.

Sequence analysis of the Wberhesda crlAT allele was achieved by first isolating this allele by cloning into the cosmid C2RB, followed by subcloning into pUC19 and sequencing by the di-deoxy chain termination method.

Evaluation of inheritance of the al AT W6s1¡ss¿a allele was undertaken utilizing ASA with PCR, using PCR primers based on sequence analysis of this allele (see Results).

To assess the expression of ol AT by the W6s1¡ss¿a allele at the level of mRNA translation, â Wbethesda crlAT oDNA was created by PCR mutagenesis of an M1(Al¿zts) glAT 6DNA. The Woetnesda ü1AT cDNA or, as a control, an alAT M1(Valzts) cDNA, were then ligated into the plasmid pSP64(polyA) for use in an in vitrolranscription system to generate olAT mRNA of W6s¡¡ssda or M1(Valzts) type. Equal amounts of M1(Valzts¡ or W6s1¡ssoa mRNA (1 pg) were then used as templates for generation of corresponding crlAT protein in rabbit reticulocyte lysate using lssS]-methionine as a label. Resultant translation products were analyzed by SDS-polyacrylamide gel electrophoresis, fluorography and laser densitometry of fluorograms, as previously described.

Biosynthesis of human crlAT directed by the ø1AT Wbethesda allele was compared to that of the alAT M1(Valzts¡ and Z alleles by utilizing COS I monkey kidney cells transfected with the pSVL eukaryotic expression vector

1 05. carrying the respective clAT cDNAs. After 72 hours of continuous culture equal numbers of each transfected population were analyzed for human ctl AT mRNA content by RNA cytoblot and hybridization to a 32P-labelled human slAT cDNA, autoradiography and laser densitometry. Human cllAT protein synthesis by transfected cells containing equal amounts of olAT mRNA were evaluated by pulse-chase protein labelling with [32S]-methionine as previously described.

5.3 RESULTS

5.3.1. Patient Characteristics

Physical examination, pulmonary function tests, chest x- ray, ventilation and perfusion lung scans, biochemistry and liver function tests were all normal in the lndex case.

5.3.2. Phenotypic Analysis of crlAT Wbethesda

Family analysis showed that three members inherited the alAT Wbethesda allele, including the mother of the lndex case (lz), the lndex case (llr) and the daughterof the lndex case (lllr) (Figure 5.1). IEF revealed thatthe crlAT Wbethesda protein migrated cathodal to the S crlAT protein but anodal to the ZaIAT protein, consistent with a designation as a W crlAT variant (Figure 5.2). The W6slhesda olAT protein bands were confirmed as glAT by immunofixation. Because the W6s1¡es¿a bânds were difficult to visualize, alAT from the lndex case was purified and concentrated and then analyzed by IEF to highlight W6s1¡gs¿ac1AT. The serum alAT levels of the lndex case were within the low normal range (Figure 5.1). The intensity of the WbethesdacrlAT protein bands relative to the S alAT protein on IEF (Figure 5.2) suggested that Wbethesda is an crl AT deficiency variant.

5.3.3. Sequence of the cllAT Wb"th""d" Gene

DNA sequence analysis of 150 base pairs 5' to exon 1c and exons lc and ll-V of the olAT W6s1¡esoa allele isolated by cosmid cloning

106 FIGURE 5.1

Pedigree of a Family with the crl AT Wbethesda Allele.

The 3 generations (l to lll) are indicated on the left. Below each family member is a number of identification within the generation followed by ø1AT phenotype and serum crlAT levels in pM units. Shading denotes the ø1AT Wbethesda allele. Dotted lines indicate individuals unavailable for analysis. I

1 2 M3S Ml Wbethesda 54 33 \ il t

1 2 SWu"tn"r¿" 21 / ilt

1 Ml Wberhesda 31 FIGURE 5.2

Evaluation of crl AT Woerhesda by IEF

IEF was performed at pH 4-5. The anode (+) is at the top and the cathode(-) at the bottom. The major 4 and 6 protein bands of the normal M crl AT (lett) or wbethesdaalAT (right) are indicated by arrows. The cr,1AT phenotype of each Lane (1-6) is at the top. An asterisk (.) indicates purified, concentrated SWu"*'""6" samples to highlight Wuethesd"ol AT bands'

Lane 1 M1 M1

Lane 2 sz

Lane 3 SWb"th".d"*

Lane 4 72

Lane 5 SWuethesda

Lane 6 SWb"th"sd"* SWb"tn""o"* M 1 M 1 SZ SWo.tn""d"" ZZ SWo"tnt"o"

M bands Wbr,r,."6" bands 4> 6> +4 +6

12 3 4 5 6 and subsequent subcloning in the plasmid vector pUC19 showed that it differed from the normal M1(Al¿zts) crlAT allele by a single base substitution in exon V of thé ø1AT gene of guanine to adenosine in the codon coding for amino acid 336. This substitution causes the amino acid difference Ala336 G.Cf ,,,, Thr ACT (Figure 5.3). No other differences compared to the M1 (Al¿zts) allele were identified.

5.3.4 Confirmation of lnheritance of crlAT Wbethesda

Analysis of available family members for the al AT Wbethesda mutation by ASA with PCR using ASA primers based on the single nucleotide substitution identified by sequence analysis (Figure 5.4) showed a pattern consistent with autosomal inheritance (Figure 5.5). ln this regard, the allele specific primer with its 3' terminus corresponding to the normal base at the W6s16ssoa mutation site (Mws) amplified DNA of all family members and an M1(Valzts) homozygote control (lanes 1, 3, 5, 7 & 9), resulting in the expected 0.32 kb amplification product consistent with each having at least one normal base at this site [the S allele is identical to the M1(Valzls) at the site of the Wberhesda mutation (Long et al, 1984)1. However, using the primer specific for the Wuetnesoa mutation (WB) only the DNA of family members lz, llt (the lndex case) and lllr amplified (lanes 6,8 & 10 respectively) consistent with each being heterozygous for the Wuetn"r6" allele and the lndex case inheriting this allele from his mother.

5.3.5. Biosynthesis of crlAT Directed by the crlAT Wb"tt allele ""d" Evaluation of the translational capacity of cr 1 A T M1 (Valzts) and Wbethesda mRNA transcripts derived f rom in vitro lranscription of the respective crlAT cDNAs in the pSP64(polyA) vector showed each produced comparable amounts of the expecled 47 kDa non-glycosylated olAT protein (p > 0.4, Figure 5.6). However, analysis of human cr1 AT production in heterologous COS I cells transfected with the pSVL eukaryotic expression vector, containing either the crlAT W6s1¡ss6¿ CDNA or aS a control the crlAT M1(Valzls) cDNA reproduced the olAT secretory defect of the Wbethesda allele suggested by the IEF and serum clAT levels of the lndex case

109 FIGURE 5.3

ldentification of the crl AT Wbetnesda Mutation by DNA Sequence AnalYsis

Autoradiogram of sequencing gel around residue 336. On the left is an M1(Al¿zts) control and on the right the Wuetn""6" allele. ln each, four lanes represent G, A, T and C respectively. The exon V nucleotide and corresponding amino acid sequences are to the side of each allele. The Wbethesda allele differs from the M1(Al¿zts) allele by a single base (indicated . by ) and resultant amino acid change Ala336 GCT "" Thr ACT. The rest of sequence analysis of the crlAT Wbethesda allele was identical to the crlAT M1(Al¿zts) allele. rG c-ì r lv"t 303 ssg val I T cl Le c-l rc n lHi" sg+ ss4 His I n Lr Mt (Ala213) Wb.th."d. rl rA GATC A-l GATC n llys sss sgs l-vs-Lo I A cl rG n"l \ c lrnr sso sgo nla I C Lr rl rG ssz val I T Lo t-c ssa Leu I T Lc rA sgg rhr I C Lc FIGURE 5.4

Strategy for A llele Specific Amplification tó- ldentify the Wo.tn"soa Allele

Shown is the normal olAT M1(Al¿zts) and Wb"th""d" nucleotide and amino acid sequence around the site of the Wbethesda mutation (boxed area) in exon V. On the left, below the exon V schematic is the nucleotide sequence of the normal primer at the Wbethesda site (Mrys) and the W6s¡¡ss6s mutâtion specific primer (WB). The nucleotide difference is underlined. When combined with the common distal primer on the right, a 0.32 kb amplification product will occur only if the 3' terminal nucleotide of the allele specific primer is complementary to the temPlate DNA. Normal M'l (Ala213) GTG CAT AAG GCT GTG CTG ACC y"¡sss ¡1¡asaa ¡"uesa T¡rsse

Wb"th""d" GTG CAT AAG ACT GTG CTG ACC y"¡sss ¡¡aaar ¡"raae T¡rsss \ ) \ Site of Wor,n""da mutation

5', Exon V 3 l-0.32kb-l ll 5' primer þ> <4 g' primer Normal 5' primer (Mvys) 5' TCCAGGCCGTGCATAAGG 3' ;l I g' primer I common l- s' ¡c¡rrrAGcGAcccrAcAAAAcAGAc s' I Wbethesda 5' primer (WB) 5' TCCAGGCCGTGCATAAGA 3' J I FIGURE 5.5

lnheritance of crl AT Woerhesda ldentified by Allele Specific AmPlification

Results of ASA of an M1(Valzt3) homozygote control and available family members as shown at the top of the figure. Amplification products were analyzed by 1% agarose gel electrophoresis and ethidium bromide staining. Above each lane is the ASA primer used. The size of the expected amplification product, 0.32 kb is on the left. All samples amplified with the nOrmal primer at the crlAT Wbethesda mutation Site (M6¡3, Lanes 1, 3,5, 7 & 9) showing all have at least one allele which is normal at this site. Only DNA of family members lz, llt, and lllt, amplified with the W6s16""6¿ mutâtion specific primer (WB, lanes 2,4,6,8 & 10) showing only these individuals carry this allele, and that inheritance was from family member lz, to lll, to llll. l1 l2 il., llll M1 M1 M3S M1Wbethesda SWo.tn""o" M1Wb.rh""da

Primer Mwe WB M*" WB Mw" WB M*" WB Mwe WB

kb

0.32 > .D

2 3 4 5 6 7 I I 10 FIGURE 5.6

ln Vitro Translation of crl AT Woerhesda mRNA TranscriPts ,

Equal amounts (1 pg) of crlAT mRNA of the M1(Valzts) orwb"rhesda type, derived from in vitro transcription of the respective crlAT cDNAs were translated and labelled with [35S]-methionine in rabbit reticulocyte lysate. Translation products were analyzed by SDS-polyacrylamide gel electrophoresis, fluorgraphy and laser densitometry of resultant fluorgrams. Shown are the fluorgrams:

Lane 1 M1(Valzts¡

Lane 2 Wbethesda

Size of the non-glycosylated translation product, 47 kDa is on the left. M1(Val213) Wo",nesda kDa

47>

1 2 (Figure 5.7). Despite there being no significant difference in ø1AT mRNA content (p > 0.5) (Figure 5.74) the crl AT W6s1¡s"6" transfected cell population secreted significantly less of the 52 kDa extracellular form of human crlAT into the supernatant after a t hour pulse labelling with [ssg¡-methionine and 2 hour chase period than the M1(Valzts) transfected cell population (49.5 + 15.57o, p < 0.05) (Figure 5.78).

lnterestingly, examination of cellular lysates of the Wbethesda and M1(Valzl3) transfected populations, and in addition Z al AT cDNA transfected cells, immediately after the t hour pulse labelling (t = 0 hr) and after lhe 2 hour chase period (l = 2 hr) revealed that the pattern of the crlAT secretory defect associated with the crlAT Wue*resoa allele was unlike that with the crl{T Z allele (Figure 5.8). At t = 0 hr, though the crlAT M1(Valzts) and Z transfected populations had similar amounts of the 50 kDa intracellular form of a1AT, the alAT Wbethesda transfected cells had less (Figure 5.8, lanes 1 - 3). ln addition, ât t = 2hr,in contrast to the crlAT Z cDNA transfected cells in which the amount of intracellular crlAT had remained similarto that at t = O hr, the intracellularcrlAT in the crlAT M1(Valzts¡ and Wbethesda transfected cDNA populations had decreased (Figure 5.8, lanes 4- 6). To evaluate whether the deficiency of secreted crlAT observed in the crlAT Wberhesdatransfected populations could be due to differential catabolism during the chase period, samples of labelled ot,lAT Wbethesda protein were re- exposed to non-transfected COS I cells for an additional 24 hours. Analysis by immunoprecipitation, SDS polyacrylamide gel electrophoresis, fluorography and laser densitometry of equal volumes of this material and the same protein not re-exposed to COS I cells showed no difference (p > 0.3).

5.4 DISCUSSION

Epidemiologic data on the association of emphysema with al AT deficiency states suggests that risk for disease occurs when serum crlAT levels are less than 11 pM (Fagerhol & Hauge, 1969; Larsson et al, 1976; Wewers et al, 1987a; Hubbard and Crystal, 1988a). Analysis of alAT alleles within this risk group has revealed that various gene mutations (base

11 4. FIGURE 5.7

Biosynthesis of crl AT W¡etnesoa irl COS I Gells

A. crlAT mRNA content in COS I cells transfected with an qlAT M1(Valzts¡ or ølAT W6s1¡es¿a cDNA. Following transfection (72 hours), equal cell 32P-labelled numbers were analyzed by cytoblot and hybridization with a alAT çDNA probe, autoradiography and laser densitometry. Shown are the autoradiograms :

Lane I M1 (Valzts) transfected cells

Lane 2 Wbethesda transfected cells

B. Secretion of newly synthesized crlAT in COS I cells transfected with an crlAT M1(Valzts) or W6s¡¡esoa cDNA. Transfected COS I cells with equivalent olAT mRNA content were labelled with ¡ssg¡-methionine for t hour followed by a 2 hour chase period. The extracellular supernatant was evaluated after the 2 hour chase for human crlAT by immunoprecipitation, SDS-acrylamide gel electrophoresis, fluorography and laser densitometry of fluorograms. Shown are the fluorograms:

Lane 3 M1 (Valzte) transfected cells

Lane 4 Wberhesda transfected cells

The size of extracellular a1AT,52kDa, is on the left A. M 1 Wbethesoa M 1 Wbethesoa

kDa

52>

1 2 3 4

I i l i

I

I FIGURE 5.8

Pattern of the cxl AT Secretory Defect Associated with crl AT Woetnesoa

lntracellular human ø1AT production in cell lysates of crlAT M1(Val21s¡, Wbethesda , or Z cDNA transfected cells was analyzed as previously. Shown are the fluorograms.

Lanes 1-3, intracellular cx,1AT after the pulse labelling period (t = 0 hr);

Lane 1 M1(Valzts¡

Lane 2 Wbethesda

Lane 3 z

Lanes 4-6- lntracellularcrlATafterthe2 hourchase period (t =2 hr);

Lane 4 M1(Valzts¡

Lane 5 Wbethesda

Lane 6 z

The size of intracellular o1AT, 50 kDa, is on the left' i i I

I I rt rl. i

I,

t=0hr t=2hr

M1 \ffn*h"u¿* Z M1 Wb*hr*d" Z

kDa

6O*

1 & 3 4 5 6 substitutions, insertions, deletions and gene deletions) can cause the crl AT deficient state by equally varied effects on biosynthesis including : reduced slAT mRNA, intracellular protein degradation and; intracellular protein accumulation (Brantly et al, 1988a; Crystal et al, 1989; Cox, 1989; Crystal, 1eeo).

ln the present study of the gene and cr,1AT biosynthesis of crlAT Wberhesda, an glATvariant identified by IEF in an SWuethes¿" heterozygote, it has been shown that the Wbethesda allele differs from the normal M1(Alazts¡ allele by a single base substitution G to A in exon V, resulting in the amino acid substitution Ala336 G-CT "" Thr ACT. lnterestingly, this amino acid substitution occurs in the same secondary structure of crlAT as the common Z- type crlAT deficiency allele, B-sheet A, strand 5 (Loebermann et al, 1984).

The Wbethesda mutation results in the substitution of a neutral hydrophobic amino acid by a neutral polar amino acid and therefore the t overall charge of the Wberhesda crl AT should be minimally affected. Therefore, the marked cathodal shift in IEF somewhat surprising (Fagerhol & Cox, 1981; Arnaud, 1988). A similar phenomenon has been observed for the Siiy"r" variant and in this case it has been postulated that the amino acid substitution (Serss "" Phe) affects the integrity and organization of the alAT molecule thus influencing its IEF behaviour (Seyama et al, 1991). Similarly, the of a neutral Mminerat sprinss al AT variant (Gtyoz "" Glu) results in the substitution amino acid by an acidic or negatively charged amino acid (Curiel et al, 1990). Based on charge alone, the predicted IEF position would be in the I or F range whereas Mminerat sprinss is cathodal to M1. Relative to the charge change, this IEF position is similar to W6s16ss6¿ ârìd Siiy".". ln the case of Mmineralsprings, thê glAT molecule functions very poorly as an inhibitorof olAT presumably due to structural change (Curiel et al, 1990). lt will be interesting to see if future analyses of Wbethesda and S¡¡y¿r" crlAT showed that those are also dysfunctional cr1 AT variants.

To explore the biosynthetic defect of the Wberhesda allele, mutant al AT cDNAs in the pSVL expression vector were transfected into, and transiently expressed in, COS I monkey kidney cells. This system which is very convenient, albeit transient, has been used extensively to investigate the Z

117 mutat¡on and is believed to provide a reliable and reproducible model (McQracken et al, 1989; Davis et al, 1990; McCracken et al, 1991). Analysis of human olAT biosynthesis in COS lcells transfected with crlAT cDNAs of thê W6s1¡esda, or aS controls, M1(Valzls¡ orZ crlAT cDNAs showed that the single base substitution of the Wbethesda allele was able to cause an ol AT secretory defect. ln this regard, though the W6s¡¡s"6" allele directed deficient slAT levels of 49.5 + 15.5% of the normal M1(V¿lzts) control, the pattern of the biosynthesis was not intracellular alAT accumulation as seen with the common Z deficiency allele (Carrell, 1986; Brantly et al, 1988a; Crystal et al, t989; Cox, 1989; Crystal, 1990). Rather, in the context of normal translation of W6e1¡ss6¿ mRNA in vitro, the biosynthetic pattern of W6s1¡..¿a crlAT during pulse-chase labelling of transfected COS I cells showing reduced intracellular levels at t = O hr which reduced further at t = 2 hr suggests post-translation degradation of the W6s16esda crlAT protein. ln contrast, the Z alAT cDNA transfected cells had almost equal amounts of intracellular crlAT att = 0 hr and t - 2 hr consistent with the known biosynthetic defect ol Z a1AT, that of intracellular accumulation. Whereas lhe Z mutation, Glu342 ,," Lys results in accumulation of al AT within the cell, in W6s1¡ss6¿ cr1AT, the replacement of the small hydrophobic residue Ala336 by the bulkier residue Thr336 in the confines of a B-sheet may disrupt it, causing intracellular degradation (Leobermann et al, 1984). Thus, although the W6s1¡ss6¿ ând ZaIAT are deficiency alleles, and although in both, the gene mutations cause amino acid substitutions in the same secondary structure of cr1AT, the resultant crlAT biosynthetic defects are markedly different.

The amount of cr,lAT produced by COS lcells transfected by the Wbethesda cDNA compared to cells transfected with the M1(Valzts¡ cDNA is consistent with Serum levels and IEF in the SW¡erhesda lndex case. This suggests that although levels of ø1AT in â Wberhesda homozygote or SWu"th".6" heterozygote should be adequate to protect the lungs from damage by neutrophil elastase, the Wberhesda allele in heterozygous combination with a severely deficient variant such as lhe Z allele or a Null allele would give rise to cr,l AT levels less than 35% of normal and therefore CaUSe an inCreaSed risk Of emphysema, and aS SuCh W6"rhesda iS an "at risk" cx,1AT allele.

118. Although the amount of crlAT directed by the Wbethesda allele suggests that this allele should be characterized as an "at risk" allele for emphysema, it is much less likely that this allele would put the individual at r¡sk for liver disease. As discussed previously (Chapter 4), all available evidence suggests that only those alAT deficiency alleles associated with accumulation of crlAT within the hepatocytes cause liver disease. ln this regard, the evaluation of the Wbethesda allele in vitro has enabled important clinical predictions to be made.

1 19. CHAPTER 6

MOLECULAR HETEROGENEITY AMONGST THE P.FAMILY OF ALPHA I.ANTITRYPSIN VARIANTS

6.1 INTRODUCTION

The P-Family of qlAT variants is so named because of the position of migration of the P olAT on IEF of serum between the M and S variants (Fagerhol & Hauge, 1969; Cox et al, 1980; Fagerhol & Cox, 1981). At least 1O P variants have been described. lP, Pbud"p""r, Pcastoria, Pclifton, Poki, Ponomichi, Psainttouis, Pkyoto, Pweishi, Pyasugi (Fagerhol & Hauge, 1968; Miyake et al, 1979; Cox et al, 1980; Fagerhol & Cox, 1981 ; Cox, 1981; Pierce & Eradio, 1981; Hug,etal 1981; Coxetal, 1982; Yuasaetal, 1984; Yingetal, 1985a). The allelic frequency of each of these variants is < 0.001 (Fagerhol & Cox 1981 ;Brantly et al 1988a). lnterestingly, the original P variant is a deficiency variant with crlAT level of 25/" of normal (Fagerhol & Hauge, 1968; Fagerhol & Hauge, 1969).

It was the purpose of the studies in this chapter to begin to characterize the molecular basis of the rare P-family of olAT alleles. ln a preliminary report defining the DNA sequence of the two P variants described in this Study, the variantS were named Ptowe¡ and P""¡nl atbans based on the birthplace of the oldest living family member with that allele (Holmes et al, 1989b). Subsequent to that, a study by Faber et al, (1989), used oligonucleotide I'Prr probes based on the partial amino acid Sequence of the variant characterized by Weidinger and Jeppsson (see Discussion in Faber et al 1989) to define the nucleotide mutation of the "P" variant. This analysis found the same mutation aS the Pbwe¡ variant. Thus, the P¡s,rvs¡¡variant is likely

120. identical to the original "P" variant. However, the same mutation has been called Nullç¿¡6¡1¡ (Graham et al, 1989) highlighting some of the difficulties with a classification system based upon lEF. ln this case the "P" protein bands have not been identified, giving rise to the incorrect designation as a Null allele. To avoid further confusion until the whole family of P alleles is completely defined, the designation Ptowe¡ has been maintained for this allele in this thesis.

6.2 METHODS

The P¡e,¡sllZ lndex case, a 55 year old Caucasian female, had a 25 pack year smoking history. There was a five year history of increasing dyspnoea. She was referred to the Pulmonary Branch, NHLBI, for consideration of investigation and treatment after reduced serum levels of al AT were detected in her serum. Routine clinical investigations were undeftaken. The patient's clinical status precluded safe bronchoscopy and bronchoalveolar lavage, however blood monocytes were able to be obtained by monocytapheresis. Two generations of family members were available for phenotypic and genotypic analysis (Figure 6.1).

The M3Psainr atbans lndex case, a 63 old male American Black had no history of lung disease and was identified during an asbestos exposure screening programme. Two generations of family members were available for analysis (Figure 6.2).

crlAT phenotype analysis was undertaken by a combination of crlAT serum levels (nephelometry), IEF at pH 4-5 as well as IEF with an immobilized pH gradient at pH 4.45 - 4.75 and family studies.

Nucleotide sequence analysis of the P olAT alleles was undertaken using the asymmetric PCR amplification technique to generate single- stranded DNA from genomic DNA extracted white blood cells from the lndex case.

121 FIGURE 6.1

Pedigree of a Family Garrying the Prowe¡r crl AT

Generations are shown on the left (1, ll). Directly below each family member is a number for identification, followed by the al AT phenotype and serum crl AT in pM units. Shading denotesthe P¡s,/vs¡¡ cr1AT. The lndex case is indicated by an arrow. 6 7 I 1 2 3 4 5 PtonettZ M1Z Plo*eltZ M1Z M1Z M1 M3 M1Z Pto*ettZ 10\ 25 26 12 14 27 47 36

il

1 2 3 M3Z M3P¡e*s¡¡ MlZ 25 34 35 FIGURE 6.2

Pedigree of a Family Carrying the Psaint atbans cxl AT

Generations are shown on the left (1, ll). Directly below each family member is a number of identification, followed by the crlAT phenotype and serum ø1AT levels in pM units. Shading denotes the Pss¡¡l atbans variant and the lndex case is indicated by an arrow. Broken lines indicate family members unavailable for analysis. I

1 2 M3Ps¿¡nt arbans M1 M1 31 /40

r I r -l I ! I I I I I L -J L J 1 2 3 4 5 M1 M3 M1 M3 M1 M3 38 27 33 Genotypic inheritance of the P alleles and confirmation of nucleotide sequence analysis was undertaken by ASA with PCR. ASA primers to identify these alleles were based upon the sequence analysis and are shown in Table 6.1 (see "Results").

ln vitro translation experiments of crl AT mRNA transcribed from recombinant plasmids created to carry the P crlAT cDNAs or an crlAT M1(Valzts) gene were undertaken as previously described.

Evaluation of ø1AT biosynthesis directed by the Ptowe¡ or Ps¿¡n1 atbans crlAT genes compared with an ø1AT M1(Valzts) gene was undertaken in vitro by retroviral gene transfer of N2 retroviral constructs contain¡ng the respective cDNAs (see Figure 6.3) into murine fibroblasts to establish permanent cell lines producing the respective alAT variants.

To determine if the deficiency of slAT in the Ptowe¡ Z lndex case (see Results) was due wholly or in part to abnormal transcriptional processing of Ptowe¡ mRNA in vivo (the Z allele gives rise to a normal amount of o1 AT mRNA), blood monocytes harvested by monocytapheresis were examined by Northern blot and cytoblot analyses.

To evaluate whether the ø1AT synthesizing cells of the alAT-deficient PlowellZ lndex case (see Results) could be stimulated in vivo lo result in alAT levels above the 11 pM threshold level that defines the risk for lung disease (Wewers et al, 1987a) tamoxifen [trans(Z) isomer of triphenylethylene], 10 mg twice daily, was administered orally. This approach is based on the knowledge that tamoxifen therapy will elevate crlAT levels associated with the M or S c[lAT allele, but not those of the Z allele (Eriksson, 1983; Wewers et al, 1987c). Serial crlAT levels, pulmonary function, serum electrolytes, haematologic parameters, and liver function tests were followed during treatment.

124. FIGURE 6.3

N2 Retroviral Vector Constructs

Schematic of the N2 retroviral vector which contains 5' and 3' long terminal repeats (LTR) and a neomycin resistance gene (NEOn¡ for selection. The Ptowe', Psaint atbans and M1(Valzts¡ ø1AT CDNAS Were inserted, together with the SV40 late promotor, into the Xho I site of this vector. N2 expression vector Mt (Vat213) CDNA or SV40 crlAT cDNA Ptowe' oDNA or Hindlll Psainr atbans CDNA

3'LTR Xhol

5'LTR

300 bp 6.3 RESULTS

6.3.1 Patient Characteristics

The Plowe¡Z lndex case was documented to have emphysema by the following : physical examination showing hyper- resonance of the chest and diminished breath sounds; a chest x-ray showing hyperexpansion and flattened diaphragms; ¿ 133¡s ventilation scan showing decreased ventilation in the bases and a eemls-¡¿croaggregated albumin perfusion scan showing matched perfusion defects; and lung function tests demonstrating a VC 63% predicted, total lung capacity ol 124/" predicted, FEVr ol 32/" predicted, FEVr/forced VC 60% predicted, and single breath diffusion capacity (corrected for volume and hemoglobin) of 24/" predicted. Arterial blood gas analysis showed a POz 79 mmHg, pCOz 32 mmHg and pH 7.40. Liver function tests and routine blood studies were normal.

The M3Ps¿¡¡1 albans heterozygote lndex case had no physical examination evidence of lung disease; chest x'ray, pulmonary ventilation and perfusion scans, lung function, liver function and blood parameters were all normal.

6.3.2 olAT Phenotype Analysis

crlAT serum levels of the PlowelrZ individual were 10 ¡tM suggesting that the P¡s,/vs¡l olAT allele was an allele associated with deficient cr,1AT expression, i.e. Pbwe¡ iS an "at risk" allele fordeveloping lung disease. ln contrast, the M3Pe¿¡¡1 atbans lndex case had an crlAT level of 40 pM indicating that the Psaintatbans allele was a normal olAT allele. IEF analysis showed that both the P¡syvs¡¡ and Pr"¡n1 atbans al AT protein migrated significantly cathodal to M crlAT but anodal to V and S cr1AT, consistent with their designation as P crlAT variants (Figure 6.4). Furthermore, the observation that the Psaint atbans variant migrated just cathodal to the P¡e,¡s¡r variant (Figure 6.4) was highlighted by IEF with immobilized pH gradient utilized to maximize the IEF migration differences of these two P variants (Figure 6.5). The relative intensities in IEF of the P bands to each other and to other normal and deficient ø1AT variants provided further evidence that P¡e,¡e¡¡ is an al AT allele

126 FIGURE 6.4

Gharacterization of crl AT Variants Prowel and Psa¡nt aroans bY IEF at PH 4'5

The anode (+) is at the top and the cathode (-) at the bottom. on the left are indicated the 2 major M-type 4 and 6 crlAT bands and on the right the major 4 and 6 bands for the Ptowe¡ and Ps¿¡¡1 atbans variants. Note the small difference in migratiOn between these variants with P¡s*s¡¡ rnorê anodal than P""¡nt albans. At the top of each lane (1-7) is the crl AT phenotype'

Lane 1 M1V

Lane 2 sz

Lane 3 72

Lane 4 M1 M1

Lane 5 M1 M3

Lane 6 M3Ps¿¡¡l atbans

Lane 7 PlowellZ MlV SZ ZZ Ml M1 M1M3 M3Psainr arbans ProweilZ

M bands P bands

4+ Plowetl M1 { 6-> Psa¡nt albans Plowell \ Psa¡nt albans

12 3 4 5 6 7 FIGURE 6.5

IEF w¡th lmmobilized pH Gradient at pH 4.45 - 4.75 to Maximize SeParation BetWeen the Plowe¡¡ and Psa¡nt albans Variants

The anode (+) is at the top and the cathode (-) at the bottom. Above each lane (1-4) is indicated the al AT phenotype.

Lane 1 M1 M1

Lane 2 PbwellZ

Lane 3 M3Ps¿¡¡¡ atbans

Lane 4 SS

To the left of each lane is shown the position of the major crl AT 4 (top) and 6 (lower) bands of that phenotype. The Z al AT 4 and 6 bands (Lane 2) are not visible because of the pH range chosen for this analysis' M3 Plowell M1M1 z Psaint albans SS t M1 ? t M3t Ç Plowelt - Psaint albans t S.> M1 M3 t - a> ; Plowell - Psaint atbans ô - SÒ

IG Õ - 1 2 3 4 assoc¡ated with deficient crlAT expression, whereaS Psaint atbans appears to be an alAT allele associated with normal ctlAT expression.

The P¡s,¡s¡¡ vâriânt cr,1AT was identified in two siblings of the lndex case and a daughter (Figure 6.1). Only the M3Ps¿¡¡¡ atbans lndex case was found to carry the Ps¿¡¡1 atbansalAT by IEF analysis of available family members (Figure 6.2).

6.3.3. Genetic Basis of crlAT P¡e$¡glt ând crlAT Psaint atbans Alleles

crl AT gene sequence analysis of the ProwellZ individual showed that the Ptowe¡ allele differed from the normal M1(Valzts) allele by a single base substitution of thymine for adenine in the codon for amino acid 256, Asp2s6 GAT "" Val GfT (Figure 6.6). Two other mutations observed in the sequence of the olAT genes of the Plowe¡Z individual when compared to an M1(Valzts¡ homozygote were attributable to the Z allele, including -

(i) heterozygosity in the codon for amino acid 213 lValzlg GAT "" Ala GQG (data not shown)]; and

(ii) heterozygosity in the codon for amino acid 342 lGlus¿z GAG "" Lys AAG] (Figure 6.7).

Both of these mutations are known to occur in the Z al AT allele (Nukiwa et al,

1986b). Sequence analysis of the M3Ps¿¡¡1 albans DNA showed that the

Psaint atbans allele differed from the M1(Valzts) allele by two base substitutions. The first was a silent mutation in the codon for amino acid 256 [Aspzs0 GAT "" Asp GAQ (Figure 6.6). The second was a mutation of guanine to adenine in the codon for amino acid 341 [Asps+t GAC ," Asn AAC] (Figure 6.7). Also observed in the sequence of the M3Ps¿¡¡1 atbans DNA was Glu376 GAA >" Asp GAe., a mutation attributable to the normal M3 allele (data not shown) (Curiel et al, 1989a). Though the silent mutation at position 256 in the M3P"aintatbans DNA could be a new mutation in an M3 allele, this was shown not to be the

case with ASA by PCR of the Ps¿¡¡l atbans family (see below).

129 FIGURE 6.6

Gene Sequence AnalYsis of the Prowe¡rZ and M3P""¡nt arbans lndex Gases Around Codon 256

Autoradiograms of the sequencing gels showing sequence of region around amino acid residue 256. On the left is an M1(Valzts¡ homozygote control, in the centre is the PlowellZ lndex case, and on the right the M3P"aint albans lndex case. For each, the base represented by each lane (G, A, T, C) is shown. The base and corresponding amino acid sequence around this region in exon lll is on the left. The P¡6,¡ys¡¡Z lndex case differs from the M1(Valzts) control by a .) single base in the codon for amino acid 256 (indicated by Asp2s6 GAT Ð>> Val GIT. The M3P.ainr atbans lndex case differs from the M1(Valzts¡ control by a *), single silent base mutation (indicated by Asp2so GAT "" Asp GAe. Sequence of region around residue 256

213)M1 213) M1(Va (Va P o*"2 M3P"" n1 r 6"n" TC 254 Leu t? 254Leu ZS+ l-", [ ? Lc L¿ Lc t-c GA TC GATC \ t- 255 Pro |t- 255 Pro zss P'o [ 3 Lç a tç Lr J t-G f-G 256 Asp 256Asp/Val 256Aspl A ti 7 L.T L.t/c 7 t-G r-G rG 257 Glu 257 Glu ln zszeru n Le Le I t-G l-G 258Glv lc 258 Gly t3 zseerv e Lc Le I FIGURE 6.7

Gene Sequence AnalYsis of the Prowe¡rZ and M3P""inr arbans Index Gases Around Codon 341

Autoradiograms of sequencing gels showing sequence of the region around amino acid residue 341. On the left is an M1(V¿lzts) homozygote control in the centre is the ProwellZ lndex case, and on the right the M3P"aint albans lndex case. For each, the base represented by each lane (G, A, T, C) is shown. The ProwelrZ individual differs from the M1(Valzt3) control by a single base .) (indicated by that defines the c¡1 AT Z mutation (Glug¿z G.AG "" Lys AAG). The M3Psainr atbans analysis on the right shows the Psaint albans allele differs from the M1(V¿lzts) homozygote by a single base change in the codon for .), amino acid 341 (indicated by Asps+t qAC "' Asn AAC. 6.3.4 Confirmation and lnheritance of the Plowelt and P"r¡nt albans Gene Mutations

To confirm the mutations revealed by DNA sequence analysis and to demonstrate ¡nher¡tance of these mutations, ASA by PCR was utilized (see Table 6.1 for primers used). For each allele specific primer, DNA of an M1(Valzts) homozygote control, the P¡e,¡s¡¡Z lndex case and the M3Pss¡¡1 atbans individual were evaluated in parallel (Figure 6.8). Using the allele specific primer with its 3' terminus corresponding to the normal base at the P¡6,/r,s¡¡ mutation site (Mp¡), amplification reactions of all 3 DNA samples yielded an expected 0.56 kb fragment (Lanes 1, 3 & 5). ln contrast, only the PlowellZ DNA amplified with the primer specific for the Ptowe¡ mutation (PL) (Lanes 2, 4 & 6). Overall, the data with the Mp¡ and PL primers is consistent with the concept that only the P¡eyy"¡¡Z individual carries the P¡s,¡ys¡¡ mutation and this individual is heterozygous for this mutation. Likewise, only the

M3Ps¿¡¡1 atbans DNA amplified with the allele specific primer for the Psaint atbans codon 256 mutation (PS3) (Lane 12) whereas all 3 samples of DNA amplified with the normal primer at this site (Mps3) (Lanes 7,9 and 11). ln a similar fashion, only the M3Pss¡¡¡ atbans DNA amplified with the primer specific for the Psaint atbans codon 341 mutation (PSs) (Lane 18), whereas all 3 of samples amplified with the primer specific for the normal base at this site (Mpss) (Lanes 13, 15 and 17). Together, this data is consistent with only the M3Ps¿¡¡1 atbans individual carrying the Ps¿¡¡1 atbans mutation and that it was present in the heterozygous state. ln addition to the above analysis, ASA evaluation of all family members of the PlowellZ lndex case (Figure 6.1) showed inheritance of the P¡s*.¡¡ allele in the expected fashion (not shown). However, parallel analysis of the available family members of the M3Ps¿¡¡1 atbans lndex case could not confirm inheritance of this allele but this is almost cerlainly due to chance, as two of the children of the lndex case were not available for analysis (Figure 6.2). Furthermore, of the three children available, none had rrM3rr the Ps¿¡¡¡ atbans codon 256 silent mutation whereas they all had the mutation of Glu376 GAA "" Asp GAe. using ASA specific for this mutation, confirming that the silent codon 256 mutation is linked to the Psaint atbans allele, not the M3 allele in this family (data not shown).

132. TABLE 6.1

Allele Specific Amplification Primers

Allele Mutat¡on Normal Primer al Mulation Site'l Mutat¡on Specilic Primer'l Common Distal Extension Primer Expected Amplified Fraoment lboì

( (PL)5'TGCTGTAGTTTCCCCTCAA3' 5'AAATATGCCTGATGCTCCMC3' 561 P ¡owell Asp256 " Val Mp¡) 5' TGCTGTAGTTTCCGoToAI 3' GAT GIT

(ps3)s'GTGCTGTAGTTTCCCCTCGs' s',AAATATGCCTGATGCTCCAACs' 562 Psaint albans Asp256, 4sp (MpSs) 5' GTGCTGTAGTTTCCCCTCA 3' GAI " GAC

(PSs)5'AGGCTGTGCTGACCATCA3' 5'CAGAGfuqAACATCCCAGGGATTTACA3' 301 Psaint albans Asp341 u ¡s¡ ( Mpsj s' AGGGTGTGCTGACCATCQ 3' qAC , MC

* Primers Mp¡, PL, MpSg, PS3 are ant¡sense primers

I Site of mutationalditferences is underlined FIGURE 6.8

Genotyp¡c Analy_si9 by Allele Specific Amplification

The alAT phenotype is at the top, Above each lane (1-18) is the ASA primer used (see Table 6.1 for details of primers) for each of the mutations defined by sequence analysis of the P alleles; the P¡eç"¡¡ êXorì lll mutation (Lanes 1-6); P3¿¡¡1 exon V the Ps¿¡¡1 atbans exon lll mutation (Lanes 7-12) and; the albans mutation (Lanes 13-18). The size of the amplified segments are given at the left of each set of lanes; see Table 6.1 for the expected size of the fragments. The M1(Valzr3) homozygote control, PlowellZ and M3Psaint albans DNA all amplified with the normal primer at the Plowell mutation site (Mp¡; Lanes 1, 3 and 5) but only the P¡s,¡s¡Z DNA amplifies with the Plowell mutation specific primer (pL; Lane 4) consistent with the individual being heterozygous for the plowe' and other individuals not carrying this mutation. Likewise, all samples amplify with the normal primer for the Psaint albans exon lll mutation (Mpss; Lanes 7, 9 and 11) whereas only the M3P"aint albans DNA amplifies with the mutation specific primer (PS3; Lane 12) confirming that this mutation occurs only in this individual. Similarly, for the Psaint albans exon V, mutation all samples amplify with the normal primer specific for this site (Mp55; Lanes 13, specific 15 and 17) but only the M3P5¿¡¡1 atbans DNA amplifies with the mutation primer for this site (PSS; lane 18)' Phenotype

M1M1 Ptn*"ttZ M3Pra,nr orban. Primer M". PL M",- M". PL kb

0 56..

! 2 3 5 6

Phenotype

M1M1 Pto*ettZ M3P..¡n¡ "¡5¿¡g Primer Mpsg PS3 M"s. PS3 M"s. PS3 kb

0.56 -

7 8 9 10 11 12

Phenotype

M1M1 Pto*"llZ M3P"uint .rbun" Primer M".o pss Mpsu PSs Mr.u PSs kb 03* I rl

I13 14 15 16 TI17 1B 6.3.5 Biosynthesis of alAT Directed by Ptowelt and Ps"¡nt atbans ctlAT cDNAs

Analysis ol in vitro translation of M1(Valel3), P¡s'vs¡¡ or Psaint atbans al AT mRNA transcripts, showed that each directed the synthesis of the expecled 47 kDa non-glycosylated olAT protein in comparable amounts (p > 0.4) (Figure 6.9). Evaluation of olAT biosynthesis in NIH-3T3 cells infected with the N2 retroviral vector containing either an M1(Val2ts¡, P¡sr¡¡s¡¡ or PsainratbansolAT cDNA (Figures 6.10 & 6.11)showed that equal numbers of each polyclonal cell population produced comparable amounts of ol AT mRNA (p r 0.5) (Figure 6.10; lanes 1-3). ln addition, each polyclonal cell population secreted a 52 kDa alAT protein into the media at 2 hours aftera 0.5 hour pulse-labelling with l3sS]-metnionine and a 2 hour chase period (Figure 6.11, lanes 1-3). Quantitation of the amount of 3sS-labelled extracellular crlAT secreted revealed similar amounts for the NIH-

3T3/M1(Valzts¡ and NIH-3T3/Psainr atbans cell populations (p > 0.4) but that there was significantly less in the extracellular supernatant of the NIH- 3T3/P¡e,¡ys¡¡ populations (p < 0.001). Quantitatively, the level of ol AT secreted by the NIH-3T3/P¡6vys¡¡ cells was 24 + 12"/" that of the NIH-3T3/M1(Valzts¡ cells, a finding that is consistent with the crlAT serum levels in the lndex case.

To ascertain the intracellular mechanism causíng the deficiency of crlAT associated with the Ptowe¡ mutation, cell lysates were analyzed immediately after the 0.5 hour pulse and after the 2 hour chase (Figure 6.11 ; lanes 4-9). lmmediately after the pulse-labelling, like the situation in the extracellular supernatant after the chase period, the NIH- 3T3/M1(Valzts¡ and NIH-3T3/P'¿¡¡¡ atbans cells had similar amounts of the 50 kDa intracellular form of olAT (lanes 4 and 6) whereas the NIH-3T3/Plowerr cells had much less (lane 5). At 2 hours, no cr,1AT was detectable in any of the cell populations (lanes 7-9). Together these observations show that the deficiency of alAT associated with the ø1AT Ptowe¡ gene is not due to accumulation of the cll AT within the cells as is seen with lhe Z and M6¿¡1s¡ slAT mutations (Brantly et al, 1988a; Crystal et al, 1989; Cox,1989; Curiel et al, 1989c). To determine whether the crlAT deficiency associated with the Pbwe¡ gene could at least in part be due to differential turnover or uptake of the ulAT protein in the extracellular supernatant, [35S]-methionine labelled Ptowel

1 35. FIGURE 6.9

In Vitro Translation of Ml(Val213) r Ptowell and Psaint arbans mRNA Transcripts

Shown are fluorograms of the [3sS]-methionine labelled olAT protein translated in vitro from synthetic olAT mRNA transcripts derived from pSP64 (polyA) constructs as described in Methods'

Lane 1 Ml lvalzts) alAT mRNA

Lane 2 Pbwel crlAT mRNA

Lane 3 Psainr atbans ol AT mRNA

The size of the translated non-glycosylated ü1AT protein, 47 k)a is on the left M1(Val213) P¡owe¡ Psaint arbans kDa

47+ t

1 2 3 FIGURE 6.10

Analysis of Human crlAT mRNA Production by Modified NIH-3T3 Cells

NIH-gT3 fibroblasts were modified by retroviral transfer to contain an M1(Valzr3), Plowell or Ps¿¡¡1 atbans human alAT cDNA and equal numbers of each type of resultant polyclone were analysed by cytoblot mRNA analysis' Shown are autoradiograms of cytoblot mRNA analysis of equal numbers of each population hybridized with a 32P-labelled ol AT CDNA.

Lane I N I H-3T3/M1 (Valzts) cells

Lane 2 NIH-3T3iPlowel cêlls

Lane 3 N I H-ST3iPsaint atbans CêllS I

.;

M1(Va¡zts¡ Prowe¡ Psarnt arbans o o o o o o o o o o o o

.i¡r .Í * ilÞ

1 2 3 FIGURE 6.11

Analysis of Human cxl AT Biosynthesis by Modified NIH-3T3 Cells

NIH-3T3 fibroblasts were modified by retroviral transfer to contain an M1(Valzts), Ptowelt Or Pe¿¡n1 atbans human crlAT cDNA. Shown are the pulse- fluorograms of the l3sS]-methionine labelled human crlAT produced by chase labelling of resultant modified polyclonal populations, immunoprecipitated with an anti-human alAT antibody and electrophoresed in a 7.5/" SDS-polyacrylamide gel.

Lanes 1-3 - extracellular supernatant aftera 0.5 hour pulse labelling period (t = O hr) w¡th [35s]-methionine and a 2 hour chase period (l = 2 hr) of :

Lanel - NIH-3T3/M1(Valzts)cells

Lane2 - NIH-3T3/P¡s'¡vs¡¡ cells

Lane 3 - NIH-3T3/Psa¡nt albans cells

Lanes 4-6 - cellular lysates immediately after the 0.5 hour pulse labelling period (t = 0 hr) in: Lane 4 - NIH-3T3/M1(Valzts) cells Lane5 - NIH-3T3/Plowe¡¡ cells

Lane 6 - NIH-3T3/Psa¡nt albans cêlls

Lanes 7-9 - cellular lysates after the 2 hour chase period (l = 2 hour) in: LaneT - NIH-3T3/M1(Valzts) cells LaneS - NIH-3TS/Plowe¡¡ cells

Lane 9 - NIH-3T3/Psaint aluans cêllS

The sizes of extracellular and intracellular forms of olAT (52 and 50 kDa, respectively) are shown on the left. Extracellular lntracellular lntracellular | = 2hr t=0hr I = 2hr M1 Ptowe¡t Psaint arbans M1 Ptowett Psainr arbans M1 Prower Psa¡nr arbans

kDa

52> 50> O O (D-o-

1 2 3 4 5 6 7 I I olAT produced by modified NIH-3T3 cells was placed on confluent plates of unmodified NIH-3T3 cells Íor 24 hours, and then immunoprecipitated and analyzed by SDS-polyacrylamide gel electrophoresis, fluorography and laser densitometry. There was no significant difference in amount of P¡q,/vgll crl AT before and after plating (p > 0.5), suggesting differential extracellular catabolism was not a factor in the deficiency of olAT associated with the Pbwe¡ allele (McOracken et al, 1989).

6.3.6 crlAT mRNA Transcripts in the P¡s¡¡¡s¡¡Z Individual

Evaluation of crlAT mRNA transcripts by Northern blot analysis and hybridization with a 32P-labelled C[lAT cDNA of total cellular RNA extracted from monocytes of the PlowellZ individual and an M1(Valzts¡ homozygote control showed identical 1.8 kb c¡1AT mRNA transcripts in both, indicating that abnormal mRNA splicing does not contribute to the deficiency of crlAT associated with the cx,1AT Ptowe¡ allele in vitro (Figure 6.12). Furthermore, quantification of crlAT mRNA transcripts in the ProwerrZ monocytes and a M1(Valzts¡ homozygote control by cytoblot mRNA analysis showed they were comparable (p > 0.5).

6.3.7 Effect of Tamoxifen Therapy on crlAT Serum Levels in a P¡6$¡s¡¡Z lndividual

Administration of oral tamoxifen, 10 mg twice daily, to the Plowe¡Z lndex case over a 5 month period produced a rise in serum ctl AT levels from below the epidemiologic threshold level for disease to well above this level (Figure 6.13). The increase in crlAT was 48"/" above the baseline level. No adverse reactions were noted during the treatment period.

6.4 DISCUSSION

The severe lung disease in the PlowerrZ lndex case is not unexpected given the profound deficiency of crl AT and significant smoking history

1 39. FIGURE 6.12

Evaluation of the crlAT mRNA Transcripts of a PlowellZ HeterozYgote

Total cellular RNA was extracted from blood monocytes of an M1(Valzts¡ homozygote control or the PlowelrZ lndex case. The RNA samples (20 pg per lane) were analyzed by Northern blot analysis and hybridization to a 32P- labelled full length clAT cDNA probe. Shown is the autoradiogram of this analysis.

Lane 1 M1(Valzts) homozygote

Lane 2 PuweltZ

The expected ø1AT mRNA transcript size (1.8 kb) is indicated on the left M1M1 Pø*"rZ kb 1.8+> t

1 2 FIGURE 6.13

Evaluation of Tamoxifen in the P¡ower¡Z Heterozygote

Oral tamoxifen therapy (10 mg twice daily) was administered over a 5 month period. Shown is the effect of tamoxifen on serum crlAT levels. On the ordinate is serum crlAT level (pM) and on the abscissa time of tamoxifen therapy (months). The period of treatment is indicated by a hatched bar. The crl AT threshold protective level from emphysema (1 1 pM) is indicated. 16 amoxifen 1 bid

14 Pto*ettZ

¿J 1 2 Threshold o 1 0 protect¡ve o level B r

E 6 f oL U) 4

2

0 1 2 3 4 5 Tamoxifen therapy (months) (Larsson et al, 1976). lf, as Seems almost certain, Ptowett is identical to the original deficient P variant, then an association with lung disease and this allele is not unprecedented (Crawford et al, 1974). Like the lndex case in the study in this thesis, the patient of Crawford et al, (1974)had the P variant in heterozygous combination with the z variant and developed premature obstructive lung disease at the age of 34.

The P-variant crlAT alleles are rare, representing far less than 1"/" ol all olAT alleles (Brantly et al, 1988a; Crystal et al, 1989; Cox, 1989; Crystal l 969). They are of interest, however, in that they include both "normal" and "at risk" alleles, thus providing an opportunity to analyze, at the molecular level, how crlAT variants with the same general charge (and hence similar behaviour on isoelectric focusing gels) result in normal or deficient c1 AT levels. The ability to rapidly and conveniently sequence crl AT variants using PCR derived methodologies meansthat most if not all crlAT variants should be able to characterized at the level of their nucleotide sequence. This should avoid confusion in classification of certain alleles and allow rapid resolution of any ambiguities revealed bY lEF.

DNA sequence analysis revealed that P¡e,¡ys¡¡ differs from the normal M1(V¿lzts) allele by a single base substitution in exon lll, Asp256 GAT "" Val GIT. The Ps¿¡¡1 atbans allele was found to differ from the normal M1(Valzte¡ allele by 2 mutations, the first in the same codon in exon lll as the Plowell substitution, but a silent mutation, Asp2s0 GAT "" Asp GAQ, and the second in exon V in the codon for amino acid 341, Asp3+t qAC ,,,, Asn AAC. Both the pbwe' mutation and the Psaint atbans exon V mutation result in the substitution of negatively charged amino acid for a neutral amino acid, consistent with the cathodal position of these proteins relative to M olAT in IEF gels. Of the two p-variants, Ptowe¡ (the deficient allele) migrated on IEF analysis slightly anodal more to Ps¿¡nl atbans (the normal allele), Suggesting that overall, P¡e,¡ys¡¡ is slightly negatively changed than P""¡nr atbans.

The p""¡nr atbans amino acid substitution is near the end of B-sheet A strand 5, next to the site of lheZ crlAT amino acid substitution lGlus¿z "" Lys] (Loebermann et al, 1984). This is of interest, in that there is increasing evidence that the charge in the 342 region is very impoftant, particularly if the

142. normal negat¡ve charge a|342 is modified to a positive charge (Brantly et al, 1988b; Mcoracken et al, 1989; Sifers et al, 1989c). Thus, at least for the normalMl(V¿lzts)molecule,theAsp341,,,,Asn(negativechargetoneutral charge) is not sufficient to modify the influence of the normal Glu342. This concept was verified by the retroviral transfer studies in which the Ps¿¡¡1 albans cDNA directed the synthesis and secretion of similaramounts of crlAT asthe normal M1(V¿lzts) cDNA. Of particular note in this regard, McOracken et al, (1991) have examined the effects of various amino acid substitutions at positions 341 and 342 of ctl AT in COS I cells transfected with a number of mutant crlAT constructs and have concluded that the negative charge normally present at this position is not necessary for efficient secretion of a1 AT.

ln contrast, the Ptowe¡ substitution at amino acid 256 has a profound effect on the intracellular processing of the newly synthesizêd P¡srys¡¡ al AT molecule, resulting in a deficiency state. Analysis of the 3-dimensional structure of crlAT reveals that the P¡e,¡ys¡¡ amino acid change is in an external bend of the crlAT molecule between B-sheet B strand 3 and ø-helix G (Loebermann et al, 1984). This residue forms a salt-bridge with His231 and elimination of this destabilizes this area (Huber & Carrell, 1989). All evidence str:ongly suggests that like the S cl1AT mutation which also disrupts a salt- bridge and causes increased intracellular degradation of cr,1AT, the P¡s*"¡¡ mutation causes exaggerated intracellular degradation of the newly synthesized molecule resulting in extracellular deficiency'

To investigate in vitro, the effect of the Ptowett and P"¿¡n¡ albans mutations on slAT biosynthesis, mutant cDNAs were inserted into the N2 construct described by Garver et al, (1987b) and this viral construct was infected into NIH-3T3 cells as described (see "Methods"). This system had previously been used with success to evaluate the S and Mr"¡on crlAT variants (Curiel et al, l gggc; curiel et al, 1989d). Although more difficult to establ¡sh than transient expression in COS I cells, it eventually provided permanent polyclonal populations of mouse fibroblasts expressing human alAT which was convenient for subsequent analyses of biosynthesis. Retroviral transfer studies, consistent with the prediction of crystallography, demonstrated that the p¡evys¡¡ mutation causes a decrease in the secretion of newly synthesized

143 ü,14T. Furthermore, the intracellular pattern of alAT observed in cells containing the P¡6,¡,"¡l cDNA demonstrated a decreased level of intracellular cllAT seen immediately after the pulse labelling period, but no íntracellular ø1AT detected atter a 2 hour chase period, a pattern similar to the pattern observed in the S mutation (Curiel et al, 1989d). This pattern is unlike that observed with the Z mutation, in which there is an accumulation of newly synthesized crlAT in the synthesizing cells (Brantly et al 1988b; McOracken et al 1989; Sifers et al 1989c). With the observations that the P¡s'/vs¡¡ crl AT mRNA is translated ín vitro in a normal fashion and that Ptowe¡ mRNA appears to be of normal amount and normally spliced in vivo, the decreased level of intracellular P¡s*.¡¡ olAT is likely due to abnormal breakdown of the newly translated protein prior to secretion. While the mechanisms causing this are unknown, the amino acid substitution in P¡s,/vs¡¡ most likely results, like S cr1AT, in the Pbwe¡ alAT folding in an abnormal fashion, a process that is recognized by the synthesizing cell, leading to increased intracellular proteolysis (Curiel et al, 1989d).

A knowledge of the intracellular mechanisms causing the deficiency state associated with the P¡e,¡ye¡¡ mutation together with the knowledge that S crlAT levels can be increased with oestrogen type drugs lead to the concept that perhaps the consequences of the P¡eyy"¡¡ mutation could be overcome by hormone manipulation (Laurell et al, 1967; Eriksson, 1983; Wewers et al, 1987c). ln this regard, with the knowledge that the liver is the major source of alAT and that alAT levels increase in pregnancy, previous studies have attempted to raise clAT levels by using the oral tamoxifen, a drug which binds intracytoplasmic oestrogen receptors (Laurell et al, 1967; Eriksson, 1983; Wewers et al, 1987c). Results showed that tamoxifen was capable of significantly raising deficient olAT levels associated with the S allele. Since the clAT P¡e,¡ys¡¡ allele has a similaramino acid change asthe ø1AT S variant, and the structural consequences and the abnormal biosynthetic pattern of crlAT of these two crlAT variants appears similar, the effects of tamoxifen in the P¡s,¡e¡¡Z lndex case were assessed. Over a 5 month period of treatment, olAT serum level was elevated significantly and in amounts comparable to that previously reported for SZ individuals, further suggesting the general mechanisms causing the deficiency states associated with crlAT P¡er¡ys¡¡ âl'ê similar to those with ø1AT S (Eriksson 1983). This rise in ol AT is above the

144 threshold level for protection from emphysema and therefore theoretically should help retard the progression of the lung disease (Wewers et al, 1987a). However, tamoxifen has potentially significant side-effects and therefore careful consideration would need to be given before undertaking long-term treatment of ø1AT deficiency with this agent.

The fact that the Ptowe¡ variant does not accumulate intracellularly in the in vitro systems used suggests that it would not be likely to be associated with liver disease (Dycaico et al, 1988; Carlson et al, 1989). This aspect would not have been assessable rn vivo tor example by liver biopsy as the lndex case also has the Z allele and thus liver biopsies would likely reveal intrahepatocyte a 1AT globules regardless of the Ptowe¡ variants pathophysiology. This further illustrates the utility and relevance of investigating these variants in in vitro systems.

1 45. CHAPTER 7

CHARACTERIZATION OF THE NORMAL ALPHA l.ANTITRYPSIN VARIANT Vmun¡ch : A VARIANT ASSOCIATED WITH A UNIQUE PROTEIN ISOELECTRIC FOCUSING PATTERN

7.1 INTRODUCTION

The 12.2kb olAT gene, located on chromosome 14 a|q32.1, is highly polymorphic, encoding more than 90 different crlAT variants identified by either IEF and/or sequence analysis (Brantly et al, 1988a; Crystal et al, 1989; Cox, 1989; see Chapter 1). For a given variant, the IEF patterns reveal microheterogeneity consisting of 2 major bands (referred to as 4 and 6) and 3 minor bands (2,7 and 8). For the common M "normal" ol AT variants, this microheterogeneity is attributed to the differences in the 3 carbohydrate side chains and the length of the crlAT protein (see Chapter 1 and schematic, Figure 7.1) (Vaughan et al, 1982; Jeppsson et al, 1985; Hercz,1985).

The purpose of the present study was to characterize a newly identified alAT variant, Vmunich, an alAT variant with unique IEF characteristics in which the microheterogeneity pattern does not fit the "classic" pattern for the relative positions of the various bands defined for the common normal M variants. lnterestingly, analysis of the coding exons of the Vmunich gene yielded an explanation for this unique IEF banding pattern.

146 FIGURE 7.1

Microheterogeneity of the Serum crlAT IEF Pattern

Show is a schematic of the IEF pattern of the normal M olAT protein and the structure of the cx,1AT molecules corresponding to each IEF band. The serum crlAT protein associated with the homozygous inheritance of normal M cx,1AT gene demonstrates a microheterogeneity based on differences in carbohydrate side chains (indicated by vertical structures attached to Asn residues at positions 46,83 and 247 ol the olAT protein schematic) and the loss of the 5 terminal amino acids GlulAsp2Pro3cln4Glys. At the left is the pl (isoelectric point) of each band. ln the middle is the crlAT structure; note that there are two different carbohydrate side chain types - biantennary and triantennary. The 3 forms focusing at the lower pH range have 394 amino acids, while the 2 forms focusing at the high pH range are missing the 5 N- terminal residues of c1AT. At the right is a schematic of the focusing pattern and band designation with the anode (+) at the top and cathode (-) at the bottom. Bands 4 and 6 are the major bands, with 2, 7 and 8 bands less abundant (Figure based on Figure 4 of Jeppsson et al, 1985). IEF Band pattern designation pl Structure

@ 2 vvv[9¡46 Asn83 Lys""o 4.42 GlulAsp2prosGlnaclys - - -Asnza7- \/VV 4 4.4g GlulAsp2prooclnoclys [5¡a6 Asn83 Lys"tn - - -Asn247- -

VV 6 4.55 GlurAsp2prosGln4clys [s¡a6 [s¡83 Asn2a7-Lyssea - - - - VV 7 4.59 [g¡a6 [g¡83 fig¡2a7 Lystno - - - -

VVV I 4.67 fig¡a6 Asn83 ¡y53ea - - -¡g¡2a7- o 7 .2 M ETHODS

Three individuals of a family carrying the alAT V¡u¡¡"¡ allele were available for analysis, the lndex case and both parents. Serum and white blood cells from these individuals were forwarded to Pulmonary Branch, NHLBI, by Dr. S.D. Weidinger, Munich, after they were identified during screening for other reasons. There was no known lung or liver disease in family members.

The olAT phenotype was determined by a combination of IEF and qlAT serum levels of family members. To determine that Vmunich focused in the "V" range, the phenotypes evaluated in parallel as standards included M1V [the "V" allele is different from "Vmuni"h", see Fagerhol & Laurell, 1967; M1S, M3S, and M2Ps¿¡¡1 ¡eu¡s (Pierce & Eradio, 1981). ln addition, enhancement of the clAT protein bands was achieved by an immunofixation print. cr,lAT serum levels were measured by nephelometry. Nucleotide sequence analysis of the lndex case was undertaken using the asymmetric PCR amplification technique to generate single stranded DNA from genomic DNA extracted from his white blood cells.

To demonstrate genotypic inheritance of the Vmunich allele, ASA with PCR was undertaken on each family member using allele specific primers based on the sequence analysis of the Vmunich allele (see Results)'

7.3 RESULTS

7.3.1. ldentification of the crlAT Vmunich Variant

Family analysis revealed two individuals with the al AT Vmunich protein (Figure 7.2),lhe lndex case (llt), and the mother of the lndex case (lz). Both of these individuals had olAT serum levels in the normal range, consistent with Vmunich being an cr,1AT allele associated with normal crlAT synthesis and secretion. IEF analysis showed that the V¡r¡¡ch var¡ant was in the "V" range, just anodal to the V type crlAT but significantly cathodal

148 FIGURE 7.2

Pedigree of a Family Carrying the crl AT Vmun¡ch Allele

Generations I and ll are shown on the left. Below each family member is a number for identification followed by the crlAT phenotype and cr,1AT serum levels in pM units. The lndex case is indicated by an arrow. Shading denotes the Vrrn¡"n allele. 1 2 M1M2 Ml Vmun¡ch 32 29 il

/ 1 Ml Vmun¡ch 46 to the Psaint ¡6u¡s Vâriânt (Figure 7.3). ln addition, IEF showed that the M1V¡u¡¡çn lndex case appeared to lack Vrrn¡"¡ 7 and 8 crlAT protein bands when compared to those of the MIV individual. The absence of Vmunich 7 and I bands was more apparent with immunofixation with a rabbit anti-human ulAT antibody performed to highlight the cllAT protein bands (Figure 7.4).

7.3.2 Elucidation of the Molecular Basis of the Vmunich Variant

DNA sequence analysis of the lndex case revealed that the M1 allele of this individual was an M1(Valzt3) allele, and that the V¡u¡¡.¡ allele differed from the normal M1(Valzts¡ allele by a single nucleotide mutation of adenosine to cytidine in the codon for amino acid 2 of the mature alAT protein (Figure 7.5). While the normal M1(Valzt3) homozygote control had the sequence Asp2 GAT, the M1(Valzts¡y'un¡"¡ heterozygote individual had both Asp2 GAT and Ala2 GQT, showing that the Vmunich allele coded for alanine at amino acid position 2. This mutation explains two interesting features of the Vmunich variant; the cathodal position of the major Vmunich 4 and 6 c¡1AT protein bands on IEF and the reason for absence of the V¡u¡¡¿¡ 7 and 8 bands in the predicted position (see Discussion).

7.3.3 Gonfirmation of lnheritance of the Vrun¡cn Allele

Analysis of the genomic DNA of each family member and an M1(Valzts¡ homozygote control by ASA with PCR showed that the Vmunich allele of the lndex case was inherited in a codominant fashion from the mother (Figures 7.6 &.7.7). Using the ASA primerwith its 3'terminus based on the normal Sequence at the Vmunich mutation site, (Myy) in combination with a common distal primer (Figure 7.6), amplification of the DNA of the normal M1(Valzte) homozygote control as well as all family members was observed, consistent with each having at least one normal allele (Figure 7.7,lanes 1,3, 5 and 7). Furthermore, the ASA primer with its 3' end complementary to the Vmunich mutation (VM) in combination with the common distal primer did not amplify the DNA of the normal control or the father (Figure 7.7,lanes 2 and 4), but did amplify the DNA of the lndex case and the mother (Figure 7.7, lanes 6 and 8) confirming they both carry the Vrrn¡"h mutation and that the lndex case had inherited the Vmunich allele from the mother.

150 FIGURE 7.3

ldentification of cxlAT V'unich by IEF at pH 4.2 - 4.9

The anode is at the top (+) and the cathode at the bottom (-). The five normal crlAT bands (2, 4,6, 7 and 8) are shown on the left. For the major 4 and 6 bands, the positions of the M1, M3, and M2 alAT proteins is indicated. On the right, the five corresponding V al AT bands are identified. For the major 4 and 6 bands, both the Vmunich and V bands are shown, indicating the small difference in position of the major 4 and 6 bands of these two variants. At the top of each lane is shown the cllAT phenotype'

Lane 1 M1V

Lane 2 M1V¡u¡¡s6 (lndex case)

Lane 3 M1V

Lane 4 M1S

Lane 5 M3S

Lane 6 M2Psaint louis (

M1V Ml Vmun¡ch M1V M1 S M3S M2Pe¿¡n1 ¡eu¡g @ Normal bands V bands l---+ w

1 +- ----2 4 3 2 7 Vmunich 1 4 6 T V , ,> 7 Vmun¡ch 6 V --D +- ---7 $ ---+ r-b

+- ---- I 123456 FIGURE 7.4

IEF w¡th lmmunofixation to Gharacterize crl AT Vmun¡cn

The anode is at the top (+) and the cathode (-) at the bottom. Normal a,1AT bands (4,6,7 and 8) are shown on the left and the V crlAT bands on the right (4, 6, 7 and 8) showing absence of V.rn¡"¡ 7 and I bands in the predicted position (indicated by "7*" and "8*"). At the top of each Lane is the al AT phenotype.

Lane 1 M1 M1

Lane 2 M1V

Lane 3 M1V¡u¡¡ç¡ M1M1 M1V Ml V-unich

@ Normal bands V bands 4> +4 9= +6 8> +7* +B*

12 3 FIGURE 7.5

DNA Sequence Analysis of the Ml Vmun¡cn lndex Case

Shown are autoradiograms of the sequencing gels. On the left is an M1(Valzts) homozygote control and on the right the M1(V¿l2ts)Vmun¡crr lndex case. tn each, the four Lanes represent the nucleotides G, A, T and C respect¡vely. The exon ll nucleotide sequences are indicated as are the corresponding sequences for amino acids -2lo 5 (amino acids -1 and -2 are part of the leader peptide of alAT). The MlVmunich DNA differs from the M1(Valzts) homozygote by having both the normal adenosine and a cytidine at the second nucleotide (indicated by """) of the codon for amino acid 2 showing the lndex case is both Asp and Ala at this codon. Therefore, the Vmunich allele differs from the M1(Valzts) allele by Asp2 GAT "" Ala GOT. The rest of the sequence analysis of the M1V¡nu¡¡srr DNA revealed no other mutations compared to the normal M1(V¿lzts) allele. rc -2 Leulr Le rG Mt (val213) M11val213) 1 Ata Ic M1(Val213) Vmunicrr Lr G ATC GATC rG (-D 1 Glu ln l¡ a r- Lc '-aD t-G Ala 2 2 Aspl¡ Lr 7 \ rc c-l Ë 3 3 Pro lc clPro Lc CJ rc c-l 4 4 Gtn ln Aleln Le eJ rG G-l GU 5 5 Gry L: ÎJ FIGURE 7.6

Scheme for Demonstration of lnheritance of the cxl AT Vmun¡ch Allele by Allele Specific AmPlification

Shown is a schematic with the nucleotide and amino acid sequence of the normal M1(Valzts) allele and the Vmunich mutation in the codon for amino acid 2 in exon ll (boxed area). The diagram of exon ll indicates the site of the Vmunich mutation. Below this on the left is the sequence of the normal allele specific 5' primer (Mvu) and the Vmunich allele specific primer (VM) with the mutational difference underlined. ln combination with the common 3' distal primer on the right, a 0.5 kb amplification product will be generated if the 3' terminus of the 5'allele specific primer is complementary to the template DNA. NormalMl (Val213) CTG GCT GAG GAT CCC CAG GGA Lou-2 Ala-l Glul Asp' Proo Gtnl Gtt'

Vmunich CTG GCT GAG GCT CCC CAG GGA Leu-2 Ala-r Glui Ala2 Pro3 Gln4 Glys I a ¿ ¿ \ ¿ \ \ \ Site of Vmun¡ch mutation

5', * Exon ll 3',

0.5 5'primørft 4"' primer I Normal 5' primer (My^r) I 5' TGTCTCCCTGGCTGAGGA 3' +{ I Common 3'Pdmer I 5' (VM) f+ 3' GCACCTCTTCCCATGAGTTCCC Vmuntc¡ 5'primer I 5' TGTCTCCCTGGCTGAGGC 3' +l FIGURE 7.7

Demonstration of lnheritance of the Vmun¡ch AIlele

Result of ASA for an M1M1 control and for each family member as indicated at the top of the Figure. Above each Lane is the allele specific primer used and on the left is indicated the size of the amplification product, 0.5 kb. The M1M1 control amplifies only with primer MVM, the normal primer at the Vmunich mutation site (Lanes 1 and 2) as does family member ll an M1M2 heterozygote (Lanes 3 and 4). Family members lz and ltr amplify with both primers Myy, ând VM (Lanes 5 and 6 and Lanes 7 and I respectively) consistent with each being an MlVmunich heterozygote and showing that inheritance of the Vmunich allele in family member llt is from family member lz. l1 l2 llt M1 M1 M1 M2 Ml Vmunich M1V-rn¡ç¡ Primer Mv" VM Mu¡n VM Mvu vM Mvu vM

kb

II 0.5 > -- - -

12 34 56 7 I 7.4 DlscussloN

IEF plays a major role in the categorization of the polymorphic protein crlAT (Cox et al, 1980; Fagerhol & Cox, 1981). All previously described normal and deficient crlAT variants categorizedby IEF have conformed to the typical pattern of two major 4 and 6 protein bands and three minor bands, the anodal 2 band and the two cathodal 7 and I bands (Vaughan et al, 1982; Jeppsson et al, 1985; Hercz,1985; Brantly et al, 1988a; Cox, 1989). ln this study is described V¡u¡¡ç¡, a normal level al AT variant whose depafture from the typical IEF pattern can be explained by its molecular basis.

DNA sequence analysis revealed that Vn.,un¡cn differs from the normal M1(Valzts¡ alAT allele by a single nucleotide substitution of adenosine by cytidine causing an amino acid change Asp2 GAT "" Ala GQT. The substitution of a negatively charged aspartic acid by a neutral alanine, creating a more positively charged protein, accounts for the cathodal position of the major 4 and 6 bands of the Vmunich variant. Furthermore, the amino acid substitution within the first 5 amino acids of the Vmunich protein explains why, in an M1(Valzts¡yru¡¡s¡ heterozygote, Vmunich 7 and I bands are missing in the expected position. As the only difference between Vrrn¡cn ârìd M1(Valzts) is within the first 5 N-terminal amino acids, and since the cr1 AT 7 and 8 bands are comprised of molecules missing the first 5 amino acids of the mature crl AT protein, removal of this one difference between the V¡y¡¡s¡ ârìd M1(Valzts¡ proteins causes the Vrrn¡"¡ 7 and 8 bands to focus with the normal M-type 7 and I bands (Figure 7.8).

Whetheror notthe cr,lAT protein comprising the 7 and 8 bands on IEF is function al in vivo is unknown. Fufihermore, it is not known what causes the cleavage of the mature ø1AT molecule between the fifth and sixth amino acids to form the 7 and 8 bands (Hercz, 1985). lnterestingly, the deduced amino acid sequence of the oDNA for the rabbit o1-antiproteinase F which has recently been isolated and which has high homology with human clAT (74/") is missing the first 5 amino acids suggesting that these may not be necessary for the function of crl AT (Saito & Sinohara, 1991).

156 FIGURE 7.8

Basis of the Unique IEF Pattern of crl AT Vmunich

On the left is the IEF pattern for the M1(Valzts) protein, with anode (+) at the top and cathode (-) at the bottom, showing the 2 major bands (4 and 6) and 3 minor bands (2,7 andS). On the right, the V¡u¡¡ch crlAT is shown to have these Same bands but the Vmunich 7 and I bands focus with the normal M- type 7 and 8 bands (arrows). This is explained (far right) by the only amino acid difference between M1(Valzts¡ and Vrrn¡"¡ being in the first 5 amino acids (residue 2, indicated by "*"), which are cleaved to form the 7 and 8 crlAT bands, and thus the proteins forming lhe 7 and I bands for M1(Valzts¡ and Vmunich crl AT are identical. muntcn ¡M1(Val213¡ o +

2- cru'fr'pro'crnocryu H 4- - -Y" -:#--.r.'"'

4- cru,o15,pro"Grn.crv, H 6- - -V -V--.r"*. VVV 7- |t- ¡"nae ASn83 LYs3e4 - - -Asñ247- 6- cru'fr,pro"Grn.o,r, - H -V" -V- - .r."". VVV 8- )e- Asn46 Asn83 LYs3e4 o - - -Asn247- Although charge and IEF pattern of the major Vmunich ollAT protein is not very different from that of the S o¿lAT deficiency variant, V,¡u¡¡6¡ is assoc¡ated with normal cl AT levels. This is consistent with the position of the amino acid substitution within the 3-dimensional structure of the al AT molecule. Whereas the S variant substitution (Glu26a "" Val) is within an a helix and is thought to disrupt a salt bridge between this residue and residue 387, the Vmunich mutation is not within any defined critical area of the mature crlAT protein (Owen & Carrell, 1976; Loebermann et al, 1984).

Due to the highly polymorphic nature of o1AT, it is possible that further interesting variants affecting the pattern of IEF microheterogeneity will be found. ln this regard, careful analysis of the changes in pattern may allow deduction of the likely area ol mutation either within the coding sequence for the first 5 amino acids of the crlAT molecule or in the codons for the 3 asparagine residues (Asn 46, 83, 247) lo which the carbohydrate side chains are attached (Carrell & Owen, 1979). Although amino acid variation can affect microheterogeneity as illustrated in this study, by far the most common change in microheterogeneity is due to changes in the carbohydrate side chain synthesis during stress, inflammation and oestrogen administration (Vaughan et al, 1982). During these states there is a pronounced rise in triantennary carbohydrate side chains resulting in prominence of lhe 2 and 4 bands. Any apparent variation of microheterogeneity therefore needs to be considered in light of the individual's clinical condition.

158 CHAPTER 8

CONCLUDING REMARKS

The molecular basis of six rare crlAT variants has been examined in this thesis.

As expected from the crlAT phenotype, the genetic basis of these crlAT variants was diverse and included mutations in three of the four protein coding exons of cr1AT. These mutations included nucleotide substitutions both silent and causing amino acid substitutions as well as a variant with an in frame triple nucleotide deletion. ln all of the variants, determination of the genetic basis allowed subsequent identification of the variant with rapid, convenient methods of mutation detection. ln both the initial sequencing of the variants, subsequent mutation determ¡nation and later creation of mutant olAT cDNAs forthe various mutations, methodologies centred around the newly described PCR technique were developed, modified and applied to the study of crlAT rapidly reducing the time needed for these experiments. As is required with all methods involving PCR, appropriate measures were taken to avoid erroneous results due to the limitations of this technology induced by its extreme sensitivity and the possibility of PCR induced mutations in amplified DNA. ln this regard, appropriate controls were always used in ASA analyses, all mutant clAT cDNas were sequenced to ensure no extra PCR induced mutations and in sequence analyses, ambiguities were resolved by sequencing in both sense and antisense directions. The ability to sequence cr,1AT variants so rapidly and conveniently using PCR methods will likely make attainable the prediction of Carrell & Owen, (1979) -

"The number of varied mobility of allelic variants makes their phenotyping and classification a difficult task. However, past experience with abnormal haemoglobins has shown the

1 59. futility of putting too much effort into such a classification. The result will have only limited meaning, since each group is likely to contain various mutants whose abnormalities, by coincidence, result in the same electrophoretic properties. Classification, in the long term, must await the identification of the molecular defects involved.".

ln the case of the Vmunich crlAT variant the gene sequence completely explained the unusual IEF phenotype and confirmed previous explanations of the mechanism of crlAT microheterogeneity with lEF.

The analyses of crlAT biosynthesis allowed a number of conclusions to be drawn. Conceptually, the Null olAT alleles can be classified into those with detectable crlAT mRNA and those without alAT mRNA in cells which normally produce cr1AT. This has importance in the question of potential liver disease - those without ø1AT mRNA need not further be considered for possible liver disease. ln addition, the Null crlAT variants may provide a convenient model for studies of mRNA stability. The findings of crlAT accumulation in cr1 AT synthesizing cells in the Mmatron olAT variant together with an association with liver disease adds weight to the now widely held belief that only those crl AT variants which accumulate intracellularly are associated with liver disease. This knowledge allows the use oÍ in vitro systems of clAT biosynthesis to predict the possibility of liver disease arising in association with that variant as was done for the W6s1¡ss6¿ ârìd Ptowe¡ alleles. Both of the systems used - transient transfection studies in COS lcells and permanent retroviral infection studies in NIH-3T3 fibroblasts were effective in biosynthetic analyses. The permanent cell lines though difficult to establish are subsequently very convenient for multiple analyses whereas the transfection studies were simple to start though each experiment was labor intensive. The use of newer vectors to establish permanent cell lines after simple transfection may prove to have the advantages of both of the systems used in this thesis.

ln the case of the P¡s,rv.¡¡ variant, the rn vifro studies of ø1AT biosynthesis were able to predict the success of a potential treatment to raise crlAT levels, in effect as coined by Eriksson, (1989) bringing lessons learned from the gene to the bedside.

1 60. I

BIBLIOGRAPHY

AAGENAES, Ø., MATLARY, A., ELGJO, K., MUNTHE, E. & FAGERHOL, M. (i972). Neonatal cholestasis in alpha-1-antitrypsin deficient children. Acta. Pediatr. Scand. 61 :632.

ABBOTT, C.M., McMAHON, C.J., WHITEHOUSE, D'8. & POVEY, S. (1988). Prenatal diagnosis of alpha-1-antitrypsin deficiency using polymerase chain reaction. Lancet i : 763.

ABBOUD, R.T., FERA, T., RICHTER,4., TABONA, M.Z. & JOHAL, S. (1985). Acute effect of smoking on the functional activity of alpha 1-protease inhibitor in bronchoalveoiar lavage fluid. Am. Rev. Respir. Dis. 131 :79.

ABE, T., TAKAHASHI, H., HOLMES, M.D., CURIEL, D.T. & CRYSTAL, R.G. (1989). Ribonuclease A cleavage combined with the.polymerase chain ieactiôn for detection of lhe Z mutation of the alpha-1-antitrypsin gene. Am. J. Respir. Cell Mol. Biol. 1 :329.

ADHYA, S. & GOTTESMAN, M. (1978). Control of transcription termination. Ann. Rev. Biochem. 47 :967.

ALLEN, M.8., WARD, A.M. & PERKS, W.H. (1986). Mmaron Z phenotype :case report and family study. Thorax 41 :568.

ALLEN, R.C., HARLEY, R.A. & TALAMO, R.C. (1974). A new method for detôrmination of alpha 1-antitrypsin phenotypes using isoelectric focusing on polyacrylamide gel slabs. Am. J. Clin. Pathol' 62 :732.

ALPER, C.A., RAUM, D., AWDEH,2.L., PETERSEN, 8.H., TAYLOR, P.D. & STARZL, T.E. (1980). Studies of hepatic synthesis in vivo of plasma proteins, including orosomucoid, transferrin, cr1-antitrypsin, C8, and factor 81. Clin. lmmunol. lmmunopathol. 16:84.

ANDERSEN, M.M. (1983). LeukocYte- of prealbumin, albumin, oxosomuc haptoglobin with human lYmPhoc promyelocytic leukaemic cell line ( 49.

161 ARCHIBALD, 4.1., McCLENAGHAN, M., HORNSEY, V., SIMONE, J.P. & CLARK, A.J. (1990). High level expression of biologically active human alpha-1-antitrypsin ín the milk of transgenic mice. Proc. Natl. Acad. Sci. USA. 87 : 5178-5182. ARNAUD, P., CHAPUIS-CELLIER, C. & CREYSSEL, R. (1974)- Polymorphisme de l'alpha-1-antitrypsin plasmatique (systeme Pi). Mise en iividence par electrofocalisation sur gel de polyacrylamide. C.R. Soc. Biol., Paris 168 :58.

ARNAUD, P., WILSON, G.8., KOISTINEN, J. & FUDENBERG, H.H. (1977). lmmunofixation after electrofocusing : lmproved method for specific detection of serum proteins with determination of isoelectric points. 1. lmmunofixation print'technique for detection of alpha-1-protease inhibitor. J. lmmunol. Methods 16:221. ARNAUD, P., GALBRAITH, R.M., GALBRAITH, G.M.P., ALLEN, R.C. & FUDENBERG, H.H. (1978). A new allele of human cr1-antitrypsin : PiNhampton. Am. J. Hum. Genet. 30 :635.

ARNAÚD, P., GALBRAITH, R.M., FAULK, W.P.& BLACK, C. (1979). Pi phenotypes of alpha-1-antitrypsin in Southern England : identification of M subtypes and implications for genetic studies. Clin. Genet. 15 : 406.

ARNAUD, P. & CHAPUIS-CELLIER, C. (1988). ø1-antitrypsin. Methods Enzymol. 163 : 337.

ATTENBURROW, A.A. (1985). Modification of the alpha-1-antitryps¡n phenotype in neonatal hepatitis. Arch. Dis. Child. 60 :378.

AXELSSON, U. & LAURELL, C.B. (1965). Hereditary variants of serum alpha 1-antitrypsin. Am. J. Hum. Genet. 17 :466.

BANDA, M.J., RICE, 4.G., GRIFFIN, G.L. & SENtOR, R.M. (1988a). The inhibitory complex of human o1-proteinase inhibitor and human leukocyte elastase is a neutrophil chemoattractant. J. Exp. Med. 167 : 1608.

BANDA, M.J., RICE, 4.G., GRIFFIN, G.L. & SENIOR, R.M. (1988b). cr1- proteinase inhibitor is a neutrophil chemoattractant afle1 proteolytic inactivation by macrophage elastase. J. Biol. Chem. 263 : 4481"

BAO, J.J., SIFERS, R.N., KIDD, V.J., LEDLEY, F.D. & WOO, S.L.C. (1987). Molecular evolution of serpins : homologous structure of the human a1- antichymotrypsin and a1-antitrypsin genes. Biochemislry 26 : 7755.

BAO, J.J., REED-FOURQUET, L., SIFERS, R.N., KIDD, V!. & WOO, S.L'C. liSAa¡. Molecular structure and sequence homology of a gene related to a1-antitrypsin in the human genome. Genomics 2 : 165.

162 BARAN, K., GORKA, M., POTEMPA, J. & PORWIT-BOBR, Z. (1989). Chemoattractant activity of Staphylococcus aureus serine proteinase modified human plasma alpha-1-proteinase inhibitor. Antionie Van Leeuwenhoek 56 : 361.

BARBEY-MOREL, C. & PERLMUTTER, D.H. (1991). Effect of pseudomonas elastase on human mononuclear phagocyte alpha-1-antitrypsin expression. Pediatr. Res. 29 : 133.

BARBEY-MOREL, C., PIERCE, J.4., CAMPBELL, E.J. & PERLMUTTER, D.H. (1987). Lipopolysaccharide modulates the expression of cr1 proteinase inhibitor and other serine proteinase inhibitors in human monocytes and macrophages. J. Exp. Med. 166 : 1041.

BARTMANN, K., FOOKE-ACHTERRATH, M., KOCH, G., NAGY, I., SCHUTZ,I., WEIS, E. &Z|ERSK|, M. (1985). Heterozygosity in the Pi-system as a pathogenetic cofactor in chronic obstructive pulmonary disease (COPD). Eur. J. Respir. Dis. 66 :284.

BATES, P.F. & SWIFT, R.A. (1983). Double cos site vectors :simplified cosmid cloning. Gene 26 :137.

BATHURST, 1.C., STENFLO, J., ERRINGTON, D.M. & CARRELL, R.W. (1983). Translation and processing of normal (PiMM) and abnormal (P\ZZ) human cr1-antitrypsin. FEBS. Lett. 1 53 :270.

BATHURST, t.C., TRAVIS, J., GEORGE, P.M. & CARRELL, R.W. (1984). Structural and functional characterization of the abnormal Z ø1-antitrypsin isolated from human liver. FEBS. Lett. 177 :179.

BATHURST, I.C., ERRINGTON, D.M., FOREMAN, R.C., JUDAH, J.D. & CARRELL, R.W. (1985). Human Z o1-antitrypsin accumulates intracellularly and stimulates lysosomal activity when synthesised in the Xenopus oocyte. FEBS. Lett. 183 : 304.

BAUMHUETER, S., MENDEL D.8., CONLEY, P.8., KUO, C.J., TURK, C., GRAVES, M.K., EDWARDS, C.A., COURTOIS, G. & CRABTREE, G.R. (1990). HNF-1 shares three sequence motifs with the POU domain proteins and is identical to LF-BI and APF. Genes Dev. 4 :372.

BEATTY, K., BIETH, J.& TRAVIS, J. (1980). Kinetics of association of serine proteinases with native and oxidized alpha 1-proteinase inhibitor and alpha 1-antichymotrypsin. J. Biol. Chem. 255:3931. BEATTY, K., ROBERTIE, P., SENIOR, R.M. & TRAVIS, J. (1982). Determination of oxidized alpha 1-proteinase inhibitor in serum. J. Lab. Clin. Med. 100 : 186.

1 63. BEATTY, K., MATHESON, N. & TRAVIS, J. (1984). Kinetic and chemical evidence forthe inability of oxidized ø1-proteinase inhibitorto protect lung elastin from elastolytic degradation. Hoppe-Seyler's Z. Physiol. Chem. 365:731.

BECKMAN, G., STJERNBERG. N.L. & EKLUNG, A. (1984a). ls the PiF allele of cr1-antitrypsin associated with pulmonary disease? Clin. Genet. 25 : 491.

BECKMAN, G., BECKMAN, 1., BJELLE, A. & DAHLQVIST, S.R. (1984b). Alpha-1-antitrypsin types and rheumatoid arthritis. Glin. Genet. 25 : 496.

BELL, H., SCHRUMPF, E. & FAGERHOL, M.K. (1990). Heterozygous MZ alpha-1-antitrypsin deficiency in adults with chronic liver disease. Scand. J. Gastroenterol. 25 : 788.

BERG, N.O. & ERIKSSON, S. (1972). Liver disease in adults with alpha-1- antitrypsin aeficiency. N. Engl. J. Med. 287 :1264.

BHAN, A.K., GRAND, R.J., COLTEN, H.R. & ALPER, C.A. (1976). Liver in cr1- antitrypsin deficiency : morpologic observations and in vitro synthesis of a1-antitrypsin. Pediatr. Res. 10 :35.

BIETH, J.G. (19S6). Elastases : catalytic and biological properties. ln : Mecham, R., ed. Regulation of matrix accumulation. Academic Press, New York. 217.

BILLINGSLEY, G.D. & COX, D.W. (1980). Functional assay of o1-antitrypsin in obstructive lung disease. Am. Rev. Respir. Dis. 121 :161.

BILLINGSLEY, G.D. & COX, D.W. (1982). Functional assessment of genetic variants of a1 antitrypsin. Hum. Genet. 61 : 118.

BILLINGSLEY, G.D., WALTER, M.A. & COX, D.W. (1989). Physical linkag eof alpha 1-antitrypsin and alpha 1-antichymotrypsin by pulsed field gel electrophoresis. Am. J. Hum. Genet. 45 (Suppl.) : 4130.

BIRNBOIM, H.C. & DOLY, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucl. Acids Res. 7 : 1513. BJELLQVIST, 8., EK, K., RIGHETTI, P.G., GIANAZZA, E., GORG, A., WESTERMEIER, R & POSTEL, W. (1982). lsoelectric focusing in immobilized pH gradients : principle, methodology and some applications. J. Biochem. Biophys. Methods 6 :317. BLACK, L.F. & KUEPPERS, F. (1978). Alpha-1-antitrypsin deficiency in nonsmokers. Am. Rev. Respir. Dis. 117 :421.

BLEUMINK, E. & KLOKKE, H.A. (1984). Protease-inhibitor deficiencies in a patient with Weber-Christian panniculitis. Arch. Dermatol. 120 :936.

164. BLIN, N. & STAFFORD, D.W. (1976) lsolation of high-molecular-weight DNA. Nucl. Acids Res.3 :2303.

BODE, W., WEI, A., HUBER, R., MEYER, E., TRAVIS, J. & NEUMANN, S. (1986). X-ray crystal structure of human leukocyte elastase (PMN elastase) and the third domain of turkey ovomucoid inhibitor. EMBO. J. 5 : 2453.

BOLLEN, A., HERZOG, A., CRAVADOR, A., HÉRION, P., CHUCHANA, P., VANDER STRATEN, A., LORIAU, R., JACOBS, P. & VAN ELSEN A. (1983). Cloning and Expression in Escherichía coli of full-length complementary DNA coding for human o1-antitrypsin. DNA 2 :255.

BOOMSMA, D.I., ORBELEKE, J.F., MARTIN, N.G., FRANTS, R.R. & CLARK, P. (1991). Alpha-1-antitrypsin and blood pressure. Lancet 37 :1547.

BOSWELL, D.R., JEPPSSON, J.O., BRENNAN, S.O. & CARRELL, R.W. (1983). The reactive site of alpha-1-antitrypsin is C-terminal, not N- terminal. Biochim. Biophys. Acta. 744 : 212.

BOUDIER, C., PELLETIER, 4., PAULI, G. & BIETH, J.G. (1983). The functional activity of alpha-1-proteinase inhibitor in bronchoalveolar lavage fluids from healthy human smokers and non-smokers. Clin. Chim. Acta. 132 : 309.

BOUTIN, 8., FENG, S.H. & ARNAUD, P. (1985). The genetic polymorphism of alpha 2-HS glycoprotein : study by ultrathin-layer isoelectric focusing and immunoblot. Am. J. Hum. Genet. 37:1098.

BRANDRUP, F. & ØSTERGAARD, P.A. (1978). o1-Antitrypsin deficiency associated with persistent cutaneous vasculitis. Arch. Dermatol. 114 : 921.

BRANTLY, M., NUKIWA, T. & CRYSTAL, R.G. (1988a). Molecular basis of alpha 1 antitrypsin deficiency. Am. J. Med. 84 (suppl 6A): 13.

BRANTLY, M., COURTNEY, M. & CRYSTAL, R.G. (1988b). Repair of the secretion defect in the Z form of alpha 1-antitrypsin by addition of a second mutation. Science 242 :1700.

BRANTLY, M.L., PAUL, L..D., MILLER, 8.H., FALK, R.T., WU, M. & CRYSTAL, R.G. (1988c). Clinical features and History of the destructive lung disease associated with alpha-1-antitrypsin deficiency of adults with pulmonary symptoms. Am. Rev. Respir. Dis. 138:327.

BRANTLY, M.1., WITTES, J.T., VOGELMEIER, C.E., HUBBARD, R.C., FELLS, G.A. & CRYSTAL, R.G. (1991). Use of a highly purified alpha-1- antitrypsin standard to establish ranges for the common normal and deficient alpha-1-antitrypsin phenotypes. Chest. 100 : 703.

1 65. BREIT, S.N., ROBINSON, J.P. & PENNY, R.(1983). The effect of alpha-1- antitrypsin on phagocyte function. J. Clin. Lab. lmmunol. 1O:147.

BREIT, S.N., WAKEFIELD, D., ROBINSON, J.P., LUCKHURST, E., CLARK, P. & PENNY, R.(1985). The role of s1-antitrypsin in the pathogenesis of immune disorders. Clin. lmmunol. lmmunopathol. 35 : 363.

BRENNAN, S.O. & CARRELL, R.W. (1986). cr1-antitrypsin Christchurch. 363 Glu ¡,,' Lys : mutation at the P5, position does not affect inhibitory activity. Biochim. Biophys. Acta 873 : 13.

BREWERTON, D.4., WEBLEY, M., MURPHY, A.H. & MILFORD WARD, A. (1978). The cr1-antitrypsin phenotypeMZ in acute anterior uveitis. Lancet i:1103.

BRODBECK, R.M. & BROWN, J.L. (1992). Secretion of alpha-1-proteinase inhibitor requires an almost full length molecule. J. Biol. Chem. 267 :294.

BROWN, W.T., MAMELOK, A.E. & BEARN, A.G. (1979). Anterior uveitis and alpha-1-antitrypsin. Lancet ii : 646.

BRUCE, R.M., COHEN, 8.H., DIAMOND, E.L., FALLAT, R.J., KNUDSON, R.J., LEBOWITZ, M.D., MITTMAN, C., PATTERSON, C.D. & TOCKMAN, M.S. (1984). Collaborative study to assess risk of lung disease in Pi MZ phenotype subjects. Am. Rev. Respir. Dis. 130 : 386.

BUIST, A.S., BURROWS, 8., COHEN, A., CRYSTAL, R.G., FALLAT, R.J., GADEK, J.E. & TURINO, G.M. (1989). Guidelines for the approach to the patient with severe hereditary alpha 1-antitrypsin deficiency. Am. Rev. Respir. Dis. 140 :1494. BULLA, G.4., DESIMONE, V., COFìTESE, R. & FOURNIER, R.E. (1992). Extinction of alpha-1-antitrypsin gene expression in somatic cell hybrids : evidence for multiple controls. Genes Dev. 6 : 316.

BUNDY, H.F. & MEHL, J.W. (1959). Trypsin inhibitors of human serum. ll. lsolation of the alpha 1-inhibitor. J. Biol. Chem. 234:1124. BURKE, J.4., KIESEL, J.L. & BLAIR, J.D. (1976). Alpha-1-ant¡trypsin deficiencyand liverdisease in children. Am. J. Dis. Child. 130:621.

BUTEL, J.S., SEPULVEDA, A.R., FINEGOLD, M.J. & WOO, S.L.C. (1990). SV4O large T antigen directed by regulatory elements of the human alpha-1-antitrypsin gene. lntervirol. 31 : 85.

CALLEA, F., FEVERY, J., MASSI, G., LIEVENS, C., DE GROOTE, J. & DESMET, V.J. (1984). Alpha-1-antitrypsin (cr1AT) and its stimulation in the liver of PiMZ phenotype individuals. A "recruitment-secretory block" ("R-SB") phenomenon. Liver 4 :325.

1 66. CAMUS, L. & GLEY, E. (1897). Action du serum sanguin sur quelque ferments digestifs. C.R. Soc. Biol. (Paris). 49 :825.

CARLSON, J. & ERIKSSON, S. (1985). Chronic'cryptogenic' liver disease and malignant hepatoma in intermediate alpha-1-antitrypsin deficiency identified by a Pi Z-specific monoclonal antibody. Scand. J. Gastroenterol. 20 : 835.

CARLSON, J.4., ROGERS, 8.8., SIFERS, R.N., HAWKINS, H.K., FINEGOLD, M.J. & WOO, S.L.C. (1988). Multiple tissues express alpha-1-antitrypsin in transgenic mice and man. J. Clin. lnvesl.82 :26.

CARLSON, J.4., ROGERS, 8.8., SIFERS, R.N., FINEGOLD, M.J., CLIFT, S.M., DEMAYO, F.J., BULLOCK, D.W. & WOO, S.L.C. (1989) Accumulation of Pi Z alpha 1-antitrypsin causes liver damage in transgenic mice. J. Clin. lnvest. 83: 1183.

CARP, H. & JANOFF, A.(1979). ln vitro suppression of serum elastase- inhibitory capacity by reactive oxygen species generated by phagocytosing polymorphonuclear leukocytes. J. Clin. lnvest. 63 : 793.

CARP, H. & JANOFF, A. (1980). Phagocyte-derived oxidants suppress the elastase-inhibitory capacity of alpha-1-proteinase inhibitor in vitro. J. Clin. lnvest. 66 :987. CARP, H., MILLER, F., HOIDAL, J.R. & JANOFF, A. (1982). Potential mechanism of emphysema : a1-Proteinase inhibitor recovered from lungs of cigarette smokers contains oxidized methionine and has decreased elastase inhibitory capacity. Proc. Natl. Acad. Sci. USA. 79 :2041.

CARRACEDO, A. & CONCHEIRO, L. (1982). The typing of alpha-1-antitrypsin in human bloodstains by isoelectric focusing. Forensic Sci. lnt. 19 : 181.

CARRELL, R.W. (1984). cr1-Antitrypsin, emphysema and smoking. N.Z. Med. J.97:327.

CARRELL, R.W. (1986). Alpha 1 antitrypsin :molecular pathology, leukocytes andtissue damage. J. Clin. 1nvest.78,1427.

CARRELL, R.W. & OWEN, M.C. (1979). Alpha 1-antitrypsin : structure, variation and disease. Essays Med. Biochem. 4: 83.

CARRELL, R.W. & OWEN, M.C. (1985). Plakalbumin, cr1-antitrypsin, antithrombin and the mechanism of inflammatory thrombosis. Nature 317 :730.

CARRELL, R.W., JEPPSSON, J-O., VAUGHAN, 1., BRENNAN, S.O., OWEN, M.C. & BOSWELL, D.R. (1981). Human a1-antitrypsin : carbohydrate attachment and sequence homology. FEBS. Lett. 135 : 301.

167 CARRELL, R.W., JEPPSSON, J-O, LAURELL, C-8, BRENNAN, S.O., OWEN, M.C., VAUGHAN, L. & BOSWELL, D.R. (1982). Structure and var¡ation of human alpha 1-antitrypsin. Nature 298 :329.

CARRELL, R.W., AULAK, K.S. & OWEN, M.C. (1989). The molecular pathology of the serpins. Mol. Biol. Med. 6 : 35.

CASKEY, C.T. (1987). Disease diagnosis by recombinant DNA methods. Science 236 :1223.

CHAABANI, H., MARTIN, J.P., FRANTS, R.R. & LEFRANC, G.(1984). Genetic study of Tunisian Berbers. ll. Alpha-1-antitrypsin (Pi) polymorphism : report of a new allele (Pi S Berber). Exp. Clin. lmmunogenet. 1 : 19. cHAl, K.X., MA, J.X., MURRAY, S.R., CHAO, J. & CHAO, L. (1991). Molecular cioning and analysis of the rat kallikrein-binding protein gene. J. Biol. Chem. 266 : 16029.

CHAN, S.K. & REES, D.C. (1976). Linear structure of the oligosaccharide chains in cr1-protease inhibitor isolated from human plasma. J. Biol. Ghem. 251 : 471.

CHANDRA, T., STACKHOUSE, R., KIDD, V.J., ROBSON, J.H. & WOO, S.L.C. (1983). Sequence homology between human a1-antichymotrypsin, a1- antitrypsin, and antithrombin ll. Biochemislry 22: 5055'

CHANG, J.C. & KAN, Y.W. (1979). ßo-thalassemia, a nonsense mutation in man. Proc. Natl. Acad. Sci. USA 76 :2886.

CHARLIONET, R., MARTIN, J.P., SESBOÜE, R., MADEC, P.J. & LEFEBVRE, F. (1979). Synthesis of highly diversified carrier ampholytes : evaluation of ihe resolving power of isoelectric focusing in the Pi system (alpha-1- antitrypsin genetic polymorphism). J. Chromatgr. 176 : 89.

CHARLIONET, R., SESBOÜÉ R., MORCAMP, C., LEFEBVRE, F. & MARTIN, J- P. (1981). Genetic variants of serum alpha-1-antitrypsin (Pi types) in Normans. Hum. Hered.31 : 104.

CILIBERTO, G., DENTE, L. & CORTESE, R. (1985). Cell-specific expression of atransfected human a1-antitrypsin gene. Cell. 41 :531.

CLARK, P. (1982). Alpha-1-protease inhibitor phenotypes in Australia. Hum. Hered. 32 :225.

CLARK, R.A., STONE, P.J., HAG,4.E., CALORE, J.D. & FRANZBLAU, C. (1 981 ). Myeloperoxidase-catalyzed lnactivation of cr1 -protease inhibitor by human neutrophils. J. Biol. Chem. 256 :3348.

COAN, M.H. (1988). Purification of alpha-1-proteinase inhibitor. Preparation and propdrties'of a therapeutic concentrate. Am. J. Med. 84 (suppl. 6A) : 32.

1 68. COHEN, 4.8., GIRARD, W,, McLARTY, J., STARCHER, 8., STEVENS, M., FAIR, D.S., DAVIS, D,, JAMES, H., ROSENBLOOM, J. & KUCICH, U. (1990). A controlled trial of colchicine to reduce the elastase load in the lungs of cigarette smokers with chronic obstructive pulmonary disease. Am. Rev. Respir. Dis. 142:63.

COLP, C., TALAVERA, W., GOLDMAN, D., GREEN, J., MULTZ, A. & LIEBERMAN, J. (1990). Profile of bronchospastic disease in Puerto Rican patients in New York City. A possible relationship to alpha-1-antitrypsin variants. Arch. lnter. Med. 150 :2349.

CONSTANS, J., VIAU, M. & GOUAILLARD, C. (1980). PiM4: an additional PiM subtype. Hum. Genet.55:119.

COOK, P.J.L. (1975). The genetics of a1-antitrypsin:a family study in England and Scotland. Ann. Hum. Genet., Lond. 38:275.

COPELAND, C.S., POMS, R.W., BOLZAU, E.M., WEBSTER, R.G. & HELENIUS, A. (1986). Assembly of influenza virus hemagglutinin trimers and its role in intracellulartranspotl. J. Cell Biol. 103:1179. COSTA, R.H., GRAYSON, D.R. & DARNELL Jr., J.E. (1989). Multiple hepatocyte-enriched nuclear factors function in the regulation of transthyretin and cr1-antitrypsin genes. Mol. Cell Biol. 9 : 1415.

COURTNEY, M., BUCHWALDER, 4., TESSIER, L-H., JAYE, M., BENAVENTE, A., BALLAND, A., KOHLI, V., LATHE, R., TOLSTOSHEV, P. & LECOCQ, J- P. (1984). High-level production of biologically active human alpha 1- antitrypsin in Escherichia Coli. Proc. Natl. Acad. Sci. USA 81 : 669.

COURTOIS, G., MORGAN, J.G., CAMPBELL, 1.4., FOUREL, G. & CRABTREE, G.R. (1987). lnteraction of a liver-specific nuclear factor with the fibrinogen and alpha-1-antitrypsin promoters. Science 238 :688.

COURTOIS, G., BAUMHUETER, S. & CRABTREE, G.R. (1988). Purified hepatocyte nuclear factor 1 interacts with a family of hepatocyte-specific promoters. Proc. Natl. Acad. Sc. USA. 85 : 7937.

COX, D.W. (1975) A new deficiency allele of alpha 1-antitrypsin : PiMmalton. ln : Peeters (ed) Protides of the Biological Fluids. Pergamon Press, Oxford , 375.

COX, D.W. (1981) New variants of c1-Antitrypsin : Comparison of Pl typing techniques. Am. J. Hum. Genet. 33 :354.

COX, D.W. (1987). DNA polymorphisms associated with alpha-1-antitrypsin and their clinical application. ln : Mittman, C. (ed). Pulmonary emphysema and proteolysis : 1986. Academic Press, New York, 119.

1 69. COX, D.W. (1989) Alpha 1-antitrypsin deficiency. ln : Stanbury, J.8., Wyngaarden, J.B, Fredrickson, D.S., Goldstein, J.L., Brown, M.S. (eds). The metabolic basis of inherited disease. McGraw-Hill Book Co., lnc., New York, 1450.

COX, D.W. & CELHOFFER, L. (1974). lnherited variants of a1-antitrypsin : a new allele PiN. Can. J. Genet. Cytol. 16:297.

COX, D.W. & HUBER, O. (1980). Association of severe rheumatoid arthritis with heterozygosity for cr,1-antitrypsin deficiency. Clin. Genet. 17 : 153.

COX, D.W. & SMYTH, S. (1983). Risk for liver disease in adults with alpha-1- antitrypsin deficiency. Am. J. Med. 74 :221. COX, D.W. & BILLINGSLEY, G.D. (1984). Oxidation of plasma alpha-1- antitrypsin in smokers and nonsmokers and by an oxidizing agent. Am. Rev. Respir. Res. 130 :594.

COX, D. & BILLINGSLEY, G.D. (1986). Restriction enzyme Maelll for prenatal diagnosis of c1-antitrypsin deficiency. Lancet ä :741 .

COX, D.W. & COULSON, S.E. (1987). Bgl ll polymorphism for the cr1- antitrypsin-related gene on chromosome 14. Nucl. Acids Res. 1 5 : 4701.

COX, D.W. & MANSFIELD, T. (1987). Prenatal diagnosis of cr1 antitrypsin deficiency and estimates of fetal risk for disease. J. Med. Genet. 24 :52.

COX, D.W. & LEVISON, H. (1988). Emphysema of early onset associated with a complete deficiency of alpha 1-antitrypsin (Null homozygotes). Am. Rev. Respir. Dis. 137:371.

COX, D.W., HOEPPNER, V.H. & LEVISON, H. (1976). Protease inhibitors in patients with chronic obstructive pulmonary disease : the alpha-1- antitrypsin heterozygote controversy. Am. Rev. Respir. Dis. 1 13 : 601.

COX, D.W., JOHNSON, A.M. & FAGERHOL, M.K. (1980). Report of nomenclature meeting for alpha 1-antitrypsin : lnserm, Rouen/Bois- Guillaume - 1978. Hum. Genet. 53 : 429.

COX, D.W., BILLINGSLEY, G.D. & SMYTH, S.(1981). Rare types of alpha 1- antitrypsin associated with deficiency. ln : Allen, R.C., Arnaud, P. (eds) : Electrophoresis '81, de Gruyer, New York, 505. COX, D.W., MARKOVIC, V.D. & TESHIMA, l.E. (1982a). Genes for immunoglobulin heavy chains and for alpha 1-antitrypsin are localized to specific regions of chromosome 14q. Nature. 297 : 428.

COX, D.W., SMYTH, S. & BILLINGSLEY, G. (1982b) Three new rare variants of a1-antitrypsin. Hum. Genet. 61 : 123.

170. COX, D.W., WOO, S.L.C. & MANSFIELD, T. (1985). DNA restriction fragments associated with a1-antitrypsin indicate a single origin for deficiency allele Pl Z. Nature 316 : 79.

COX, D.W., BILLINGSLEY, G.D. & CALLAHAN, J.W. (1986). Aggregation of plasma Z lype cr1-antitrypsin suggests basic defect for the deficiency. FEBS. Lett. 205 :255.

COX, D.W., NAKAMURA, Y. & GEDDE-DAHL Jr, T. (1990) Report of the committee on the genetic constitution of chromosome 14. Cytogenet. Cell Genet. 55 : 183.

COX, R.A. (1968). The use of guanidinium chloride in the isolation of nucleic acids. Methods Enzymol. 128:120. cRAWFORD, LP., DAWSON, A. & STEVENSON, D.D. (1974). Alpha-1- antitrypsin deficiency associated with PZ and MP phenotypes. Am. J. Med. 57 :210.

CRESTEIL, D., CICCARELLI, E., SONI, T., ALONSO, M.4., JACOBS, P., BOLLEN, A. & ALVARES, F. (1990). BiP expression is not increased by the accumulation ol PiZ alpha-1-antitrypsin in the endoplasmic reticulum. FEBS. Lett.267 :277.

CROCE, C.M., SHANDER, M., MARTINIS, J., CICUREL, L., D'ANCONA, G.G., DOBLY, T.H. & KOPROWSKI, H. (1979), Chromosomal location of the genes for human immunoglobulin heavy chains. Proc. Natl. Acad. Sc. USA. 76 :3416.

CROWLEY, J.J., SHARP, H.1., FREIER, E., ISHAK, K.G. & SCHOW, P. (1987). Fatal liver disease associated with cr1-antitrypsin deficiency PiMl / PiM6u¿xs. Gastroenterol. 93 :242.

CRYSTAL, R.G. (1989) The o1-antitrypsin gene and its deficiency states. Trends. Genet5:511.

CRYSTAL, R.G. (1990) Alpha 1-antitrypsin deficiency, emphysema and liver disease. Genetic basis and strategies for therapy. J. Clin. lnvest. 85 : 1 343.

CRYSTAL, R.G., BRANTLY, M.L., HUBBARD, R.C., CURIEL, D.T., STATES, D.J. & HOLMES, M.D. (19S9). The alpha 1 antitrypsin gene and its mutations : clinical consequences and strategies for therapy. Chest 95 : 1 96.

CURIEL, D., LAUBACH, V., VOGELMEIER, C., WURTS, L. & CRYSTAL, R.G. (19S9a). Characterization of the sequence of the normal alpha-1- antitrypsin M3 allele and function of the M3 protein. Am. J. Respir. Cell Mol. Biol. 1 :471.

171 CURIEL, D., BRANTLY, M., CURIEL, E., STIER, L. & CRYSTAL, R.G.(1989b). Alpha 1-antitrypsin deficiency caused by the alpha 1-antitrypsin Null¡¿11¿s¡¿ gênê : an insertion mutation rendering the alpha 1-antitrypsin gene incapable of producing alpha 1-antitrypsin. J. Clin. lnvest.83:1144.

CURIEL, D.T., HOLMES, M.0., OKAYAMA, H., BRANTLY, M.L., VOGELMEIER, C., TRAVIS, W.D., STIER, 1.E., PERKS, W.H. & CRYSTAL, R.G. (1989C) Molecular basis of the liver and lung disease associated with the cr1 - antitrypsin deficiency allele Mmahon. J. Biol. Chem, 264 :13938. cuRlEL, D., CHYTIL,4., COURTNEY, M. & CRYSTAL, R.G. (1989d). Serum a1-antitrypsin deficiency associated with the common S-type (Glu26a ¡ Val) mutation results from intracellular degradation of a1-antitrypsin prior to secretion. J. Biol. Chem. 264 : 10477.

CURIEL, D., STIER, L. & CRYSTAL, R.G. (1989e). Gene therapy for a1 - antitrypsin deficiency using lymphocytes as vehicles for o1-antitrypsin delivery. Clin. Res. 37 :5784.

CURIEL, D.T., VOGELMEIER, C., HUBBARD, R.C., STIER, L.E. & CRYSTAL, R.G. (1990). Molecular basis of cr,1-antitrypsin deficiency and emphysema associated with the cr1-antitrypsirì Mmineratsprinss allele. Mol. Cell Biol. 10 : 47.

DAAR, l.O. & MAQUAT, L.E. (1988). Premature translation termination mediates triosephosphate isomerase mRNA degradation. Mol. Cell Biol. I :802. DARLINGTON, G.J., ASTRIN, K.H., MUIRHEAD, S.P., DESNICK, R.J. & SMITH, M. (19S2). Assignment of the human alpha 1-antitrypsin gene to chromosome 14 by somatic cell hybrid analysis. Proc. Natl. Acad. Sci. USA. 79 :87O.

DAVIES, J. & JIMENEZ, A.(1980). A new selective agent for eukaryotic cloning vectors. J. Trop. Med. Hyg.29 : 1089.

DAVIS, E.S., HARPFZ, N. & JOHNSON, E.M. (1990). Expression of PiM- and PiZ-mutated forms of the human a1-antitrypsin gene in transfected monkey COS I cells. J. Biol. Chem. 265 : 22153.

DE SIMONE, V. & CORTESE, R. (1989). A negative regulatory element in the promoter of the human o1-antitrypsin gene. Nucl. Acids Res. 17 :9407.

DE SIMONE, V., CILIBERTO, G., HARDON, E., PAONESSA, G., PALLA, F., LUNGBERG, L. & CORTESE, R. (1987). Cis- and trans- acting elements responsible for the cell-specific expression of the human cr1-antitrypsin gene. EMBO. J. 6 :2759.

172 DE LlLLlS, R.4., BALOGH, K., MERK, F.B.& CHIRIFE, A.M. (1972). Distinctive hepatic cell globules in adult alpha-1-antitrypsin deficiency. Arch. Path. 94 :308.

DERMER, S.J., JOHNSON & E.M. (1988). Methods in laboratory investigation :rapid DNA analysis of o1-antitrypsin deficiency : application of an improved method for amplifying mutated gene sequences. Lab. lnvest. 59 :403.

DOHERTY, J.8., ASHE, 8.M., ARGENBRIGHT, L.W., BARKER, P.L., BONNEY, R.J., CHANDLER, G.O., DAHLGREN, M.E., DORN, D.P. Jr., FINKE, P.E', FIRESTONE, R.4., FLETCHER, D., HAGMANN, W.K., MUMFORD, R., O'GRADY, L., MAYCOCK, 4.L., PISANO, J.M., SHAH, S.K., THOMPSON, K.R. & ZIMMERMAN, M. (1986). Cephalosporin antibiotics can be modified to inhibit human leukocyte elastase. Nature 332 :192. DRY, P.J. (1991). Rapid detection of alpha-1-antitrypsin deficie_ncy by analysis of a POR-induced Taql restriction site. Hum. Genet. 87 :742. du BOIS, R.M., BERNAUDIN, J-F., PAAKKO, P., HUBBARD, R., TAKAHASHI, H., FERRANS, V. & CRYSTAL, R.G. (1991). Human neutroph¡ls express the a1-antitrypsin gene and produce g1-antitrypsin. Blood 77 : 2724.

DUBEL, J.R., FENWICK, R. & HEJTMACIK, J.F. (1989). Application of denaturing gradient gel electrophoresis (DGGE) to prenatal diagnosis of alpha-1-antitrypsin. Am. J. Hum. Genet. 45 (suppl) : 4184.

DUBEL, J.R., FENWICK, R. & HEJTMANCIK, J.F. (1991). Denaturing gradient gel electrophoresis of the alpha 1-antitrypsin gene : application to prenatal diagnosis. Am. J. Med. Genet. 41 : 39.

DYCAICO, M.J., GRANT, S.G.N., FELTS, K., NICHOLS, W.S., GELI¡R, S.A., HAGER, J.H., POLLARD, 4.J., KOHLER, S.W., SHORT, H.D., JIRIK, F.R', HANAHAN, D. & SORGE, J.A. (1988). Neonatal hepatitis induced by alpha 1-antitrypsin : a transgenic mouse model. Science 242 : 1409.

DYKES, D.D., MILLER, S.A. & POLESKY, H.F. (1984). Distribution of cr1- antitrypsin variants in a US white population. Hum. Hered. 34 : 308.

DZIEGIELEWSKI, K.M., SAUNDERS, N.R., SCHEJTER, E.J., ZAKUT, H., ZEVIN-SONKIN, D., ZISLING, R. & SOREZ, H. (1986). Synthesis of plasma proteins in fetal, adult and neoplastic human brain tissue. Der. Biol. 115:93.

EBERT, R.V. & PIERCE, J.A. (1963). Pathogenesis of pulmonary emphysema. Arch. lntern. Med. 111 : 34.

EGL|TIS, M.A., KANTOFF, P., GILBOA, E. & ANDERSON, W.F. (1985). Gene expression in mice after high efficiency retroviral-mediated gene transfer. Science 230 : 1395.

173 EHLEN, T. & DUBEAU, L. (1989) Detection of ras point mutations by polymerase chain reaction using mutation specific, inosine-containing oligonucleotide primers. Biochem. Biophys. Res. Commun. 160 : 441.

ELBEIN, A.D. (1987). lnhibitors of the biosynthesis and processing of N-linked oligosaccharide chains. Ann. Rev. Biochem. 56 : 497.

ENGH, R., LOEBERMANN, H., SCHNEIDER, M., WIEGAND, G., HUBER, R. & LAURELL, C-8. (1989) The S variant of human cr1-antitrypsin, structure and implications for function and metabolism. Protein Eng. 2 : 4O7. ERIKSSON, S. (1964). Pulmonary emphysema and alpha 1-antitrypsin deficiency. Acta Med. Scand. 175 : 197.

ERIKSSON, S. (1965). Studies in alpha 1-antitrypsin deficiency. Acta Med. Scand. 177 (Suppl.) 432 : 5.

ERIKSSON, S. (1983) The effect of tamoxifen in intermediate alpha 1 antitrypsin deficiency associated with the phenotype P|SZ. Ann. Clin. Res. 15:95.

ERIKSSON, S.G. (1985). Liver disease in a1-antitrypsin deficiency. Scand. J Gastroenterol. 20 : 907.

ERIKSSON, S. (1989). Alpha-1-Antitrypsin Deficiency : Lessons Learned from the Bedside to the Gene and Back Again. Chest 95 : 181.

ERIKSSON, S. & LAURELL, C-8. (1963). A new abnormal serum globulin cr1-antitrypsin. Acta Chem. Scand. (Suppl.) : 150.

ERIKSSON, S. & HÄGERSTRAND, l. (1974). Cirrhosis and malignant hepatoma in a1-antitrypsin deficiency. Acta Med. Scand. 195 : 451. ERIKSSON, S. & LARSSON, C. (1975). Purification and part¡al characterization of PAS-positive inclusion bodies from the liver in alpha-1- antitrypsin deficiency. N. Engl. J. Med. 292:176.

ERIKSSON, S., ALM, R. & ASTEDT, B. (1978). Organ cultures of human fetal hepatocytes in the study of extra- and intra-cellular o1-antitrypsin. Biochim. Biophys. Acta. 542:496.

ERIKSSON, S., LINDEL, S-E. & WIBERG, R. (1985). Effects of smoking and intermediate cr1-antitrypsin deficiency (PiMZ) on lung function. Eur. J. Respir. Dis. 67 :279.

ERIKSSON, S., CARLSON, J. & VELEZ, R.(1986). Risk of cirrhosis and primary liver cancer in alpha-1-antitrypsin deficiency N. Engl. J. Med. 314 :736.

174 ERRINGTON, D.M., BATHURST, I.C., JANUS, E.D. & CARRELL, R.W. (1982) ln vitro synthesis of M and Z forms of human alpha-1-antitrypsin. FEBS Lett. 148 : 83.

ERRINGTON, D.M., BATHURST, l.C. & CARRELL, R.W. (1985). Human o1- antitrypsin expression in Xenopus oocytes secretion of the normal (PiM) and abnormal (PiZ) forms. Eur. J. Biochem. 153 : 361.

EVALD, T., DIRKSEN, A., KEITTELMANN, S., VISKUM, K. & KOK-JENSEN, A. (1990). Decline in pulmonary function in patients with alpha-1-antitrypsin deficiency. Lung 168(suppl) : 579.

FABER, J-P., WEIDINGER, S., GOEDDE, H-W. & OLEK, 1(. (1989) The deficient alpha 1 antitrypsin phenotype Pi P is associated with an A-to-T transversion in exon lll of the gene. Am. J. Hum. Genet. 45 : 161.

FABER, J-P., WEIDINGER, S. & OLEK, K. (1990). Sequence data of the rare deficient alpha-1-antitrypsin variant Pl Zaugsburg. Am. J. Hum. Genet. 46 :1158.

FAGERHOL, M.K. (1967). Serum Pi types in Norwegians. Acta. Path. Microbiol. Scand. 70 : 421.

FAGERHOL, M.K. (1969). Quantitative studies on the inherited variants of serum alpha 1-antitrypsin. Scand. J. Clin. Lab. lnvesl.23:97.

FAGERHOL, M.K. (1972). Genetics of the Pi System. ln : Mittman, C. (ed). Pulmonary Emphysema and Proteolysis. Academic Press, New York , 123.

FAGERHOL, M.K. & BRAEND, M. (1965). Serum prealbumin : polymorphism in man. Science 149 : 986. FAGERHOL, M.K. & LAUREL, C.B. (1967). The polymorphisms of "prealbumins" and alpha 1-antitrypsin in human sera. Clin. Chim. Acta 16 :199.

FAGERHOL, M.K. & TENFJORD, O.W. (1968). serum Pi types in some European American, Asian and African populati ons. Acta Path. Microbiol. Scandinav. 72 : 601 .

FAGERHOL, M.K. & HAUGE, H.E. (1968) The Pi phenotype MP : discovery of a ninth allele belonging to the system of inherited variants of serum a1- antitrypsin. Vox Sang 15 :396.

FAGERHOL, M.K. & GEDDE-DAHL, T. (1969). Genetics of the Pi serum types Hum. Hered. 19 : 354.

FAGERHOL, M.K. & HAUGE, H.E.(1969) Serum Pi types in patients with pulmonary disease. Acta Allergol. 24 : 1O7.

175 FAGERHOL, M.K. & COX, D.W. (1981 ) The Pi polymorphism : genetic, biochemical, and clinical aspects of human alpha 1-antitryPsin. Adv. Hum. Genet. 11 : 1. FEARNLEY, I.R., SPALTON, D.J., MILFORD WARD, 4., SLAVIN, B. & MUNCEY, F. (1988). cr1-antitrypsin phenotypes in acute anterior uveitis. Br. J. Ophthalmol. T2: 636.

FELDMANN, G., BIGNON, J., CHAHINIAN, P., DEGOTT, C. & BENHAMOT, J- P. (1974). Hepatocyte ultrastructural changes in cr1-antitrypsin deficiency. Gastroenterol. 67 : 1214.

FELDMANN, G., MARTIN, J-P., SESBOUE, R., ROPARTZ, C., PERELMAN, R., NATHANSON, M., SERINGE, P. & BENHAMOU, J-P. (1975). The ultrastructure of hepatocytes in alpha-1-antitrypsin deficiency with the genotype Pi--. Gut. 16 : 796.

FISHER, R.L., TAYLOR, L. & SHERLOCK, S. (1976). ø1-Antitrypsin deficiency in liver disease : the extent of the problem. Gastroenterol. 71 : 646.

FOREMAN, R.C. (1987) Disruption of the Lys - 290 - Glu - 342 salt bridge in human alpha-1-antitrypsin does not prevent its synthesis and secretion. FEBS Lett. 216 : 79.

FOREMAN, R.C., JUDAH, J.D. & COLMAN, A. (1984). Xenopus oocytes can synthesise but do not secrete lhe Z variant of human ø1-antitrypsin. FEBS. Lett. 168 : 84.

FORTIN, P.R., FRASER, R.S., WATTS, C.S. & ESDAILE, J.M. (1991). Alpha-1- antitrypsin deficiency and systemic necrotizing vasculitis. J. Rheumatol. 18:1613.

FRAIZER, G., HARROLD, T.R., SIEWERTSEN, M. & COX, D.W. 098_8)_ Deletion/frameshift mutation in the alpha 1 antitrypsin null allele, Pl" QO botton. Am. J. Hum. Genet. a3 (suppl) : 4183.

FRAIZER, G.C., HARROLD, T.R., HOFKER, M.H. & COX, D.W. (1989a). ln: frame single codon deletion in the Mmatton deficiency allele of alpha 1 ant¡trypsin. Am. J. Hum. Genet. 44:894.

FRAIZER, G.C., SIEWERTSEN, M., HARROLD, T.R. & COX, D.W, (1989b). Deletion/frameshift mutation in the alpha-1-antitrypsin null allele. Hum. Genet. 83 :377.

FRAIZER, G.C., SIEWERTSEN, M.4., HOFKER, M.H., BRUBACHER, M.G' & COX, D.W. (1990). A null deficiency allele of alpha 1 antitrypsin, QOtuow¡g"rr¿1s¡ with altered tertiary structure. J. Clin. lnvest. 86 : 1878.

FRANTS, R.R. & ERIKSSON, A.W. (1980). A new unstable Pl M Variant of ct1- antitrypsin in a Finnish isolate. Hum. Hered. 30 :333. 176. FRANTS, R.R., NOORDHOEK, G.T. & ERIKSSON, A.W. (1978). Separator isoelectric focusing for identification of alpha 1-antitrypsin (P:M) subtypes. Scand. J. Clin. Lab. lnvest. 38:457.

FRANTS, R.R., HOOGHWINKEL, G.J. & ERIKSSON, A.W. (1981). Comparison of the Pl patterns of alpha-1-antitrypsin on acid starch gel electrophoresis and isoelectric focusing. Scand. J. Clin. Lab. lnvest. 41 : 15.

GADEK, J.E. & CRYSTAL, R.G. (1982). o1-antitrypsin deficiency. ln : Stanbury, J.8., Wyngaarden, J.8., Fredrickson, D.S., Goldstein, J.L., Brown, M.D. (eds). Metabolic basis of inherited disease. McGraw Hill, New York , 1450. GADEK, J.E., FELLS, G.A. & CRYSTAL, R.G. (1979). Cigarette smoke inhalation decreases cr1-antitrypsin activity in rat lung. Science 206 : 1313.

GADEK, J.E., KEOGH, B.A. & CRYSTAL, R.G. (1980a). Opportunities forthe specific therapy of destructive lung disease. Bull. Europ. Physiopath. Resp. 16 :389.

GADEK, J.E., FULMER, J.D., GELFAND, J.4., FRANK, M.M., PETTY, T.L. & CRYSTAL, R.G. (1980b). Danazol-induced augmentation of serum o1- antitrypsin levels in individuals with marked deficiency of this antiprotease. J. Clin. lnvest. 66 :82.

GADEK, J.E., FELLS, G.4., ZIMMERMAN, R.1., RENNARD, S.I. & CRYSTAL, R.G. (1981a) Antielastases of the human alveolar structures : implications for the protease-antiprotease theory of emphysema. J. Clin. lnvest. 68 : 889.

GADEK, J.E., KLEIN, H.G., HOLLAND, P.V. & CRYSTAL, R.G. (1981). Replacement therapy of alpha 1-antitrypsin deficiency. J. Clin. lnvest. 68 : 1 158.

GAILLARD, ¡., CLAUSER, E. & CORVOL, P. (1989). Structure of human angiotensinogen gene. DNA 8 : 87.

GANROT, P.O. & BJERRE, B. (1967). o1-antitrypsin and a2-macroglobulin concentration in serum during pregnancy. Acta. Obst. Gynec. Scand. 46 : 1.

GARVER, R.l. Jr., MORNEX, J-F., NUKIWA, T., BRANTLY, M., COURTNEY, M', LECOCQ, J-P. & CRYSTAL, R.G. (1986). Alpha-1-antitrypsin deficiency and emphysema caused by homozygous inheritance of non-expressing alpha-1-antitrypsin genes. N. Engl. J. Med. 314:762.

GARVER, R.l.Jr., CHYTIL, 4., COURTNEY, M. & CRYSTAL, R.G. (1987a). Clonal gene therapy : transplanted mouse fibroblast clones express human cr1-antitrypsin gene in vivo. Science 237 :762.

177. GARVER, R.l. Jr., CHYTIL, 4., KARLSSON, S., FELLS, G.A., BRANTLY, M.L., COURTNEY, M., KANTOFF, P.W., NIENHUIS,4.W., ANDERSON, W.F. & CRYSTAL, R.G. (1987b). Production of glycosylated physiologically "normal" human alpha 1-antitrypsin by mouse fibroblasts modified by insertion of a human alpha 1-antitrypsin cDNA using a retroviral vector. Proc. Natl. Acad. Sci. USA 84 : 1050.

GAUTIER, M., MARTIN, J-P. & POL|Nl, G. (1977). ln vitro synthesis of alpha-1- antitrypsin in long term monolayer human liver cell cultures. Biomedicine 27:116.

GEDDE-DAHL, T., FAGERHOL, M.K., COOK, P.J.L. & NOAKES, J. (1972). Autosomal linkage between the Gm and Pc loci in man. Ann. Hum. Genet. 35 :393.

GEORGE, P.M., VISSERS, M.C.M., TRAVIS, J., WINTERBOURN, C.C., & CARRELL, R.W. (1984). A genetically engineered mutant of cr1-antitrypsin protects connective tissue from neutrophil damage and may be useful in lung disease. Lancet ä:1426.

GETHING, M-J., McCAMMON, K. & SAMBROOK, J. (1986). Expression of wild-type and mutant forms of influenza hemagglutinin :the role of folding in intracellular transport. Cell 46 : 939.

GHSIN, V., CRKVENJAKOV, R. & BYUS, C. (1974). Ribonucleic acid isolation by cesium chloride centrifugation. Biochemistry 13 : 2633.

GIBBS, R. & CASKEY, C.T. (1987). ldentification and localization of mutations at the Lesch-Nyhan locus by ribonuclease A cleavage. Science 236 : 303.

GlsHEN, P., SAUNDERS, A.J., TOBIN, M.J. & HUTCHISON, D.C. (1982). Alpha-1-antitrypsin deficiency : the radiological features of pulmonary emphysema in subjects of Pi type Z and Pi type SZ : a survey by the British Thoracic Association. Clin. Radiol. 33 :371.

GLASER, C.8., KARIC, 1., FALLAT, R.J. & STOCKERT, R. (1977). Alpha-1- antitrypsin. plasma survival studies in the rat of the normal and homozygote deficient forms. Biochim. Biophys. Acta. 495 : 87.

GLASGOW, J.E., BAGDASARIAN, A. & COLEMAN, R.W. (1982). Functional alpha-1-protease inhibitor produced by a human hepatoma cell line. J. Lab. Clin. Med. 99 : 108.

GLUZMAN, Y. (1981). SV40-transformed simian cells support the replication of early SV40 mutants. Cell 23 :175.

GOETZL, E.J. (1975). Modulation of human neutrophil polymorphonuclear leukocyte migration. by human plasma alpha-globulin inhibitors and synthetic esterase inhibitors. J. lmmunol. 29 : 163.

178 GORG, A., POSTEL, W., WESER, J., WEIDINGER, S., PATUTSCHNICK, W., & CLEVE, H. (1983). lsoelectric focusing in immobilized pH gradients for the determination of the genetic Pi (ø 1-antitrypsin) variants. Electrophoresis 4 : 153.

GORG, A., POSTEL, W., WESER, J., PATUTSCHNICK, W. & CLEVEL, H. (1985). lmproved resolution of Pl (cr1-antitrypsin) phenotypes by a large- scale immubilized pH gradient. Am. J. Hum. Genet. 37 :922.

GORDON, H.W., DIXON, J., ROGERS, J.C., MITTMAN, C. & LIEBERMAN, J. (1972). Alpha-1-antitrypsin (o 1AT) accumulation in livers of emphysematous patients with al AT deficiency. Hum. Pathol. 3 : 361 .

GOVINDARAJAN, S., ASHCAVAI, M. & PETERS, R.L. (1981). a1-antitrypsin phenotypes in hepatocellular carcinoma. Hepatol. 1 : 628.

GRAHAM,4., J Val); PiM¡¡¿¡e¡ (Phe sl > deletion) and Pil (Argss t Cys). Hum. Genet. 84 :55.

GRAHAM, 4., KALSHEKER, N.4., BAMFORTH, F.J., NEWTON, C.R. & MARKHAM, A.F. (1990a). Molecular characterisation of two alpha-1- antitrypsin deficiency variants : proteinase inhibitor (Pi) Null¡s'/vps¡1 (Gly1 1S---Ser) and (Pi) Zwrexham (Ser-19---Leu). Hum. Genet. 85 : 537.

GRAHAM,4., HAYES, K., WEIDINGER, S., NEWTON, C.R., MARKHAM, A.F. & KALSHEKER, N.A. (1990b). Characterisation of the alpha-1-antitrypsin M3 gene, a normal variant. Hum. Genet. 85 : 381.

GRAHAM, K.S., LE, A. & SIFERS, R.N.(1990). Accumulation of the insoluble PiZ variant of human a1-antitrypsin within the hepatic endoplasmic reticulum does not elevate the steady-state level of grp78/BiP. J. Biol. Chem. 265 :20463.

GRAHAM, R. & VAN DER EB, A. (1973). A new technique for assay of infectivity of human adenovirus 5 DNA. Virol. 52:456.

GRAYSON, D.R., COSTA, R.H., XANTHOPOULOS, K.G. & DARNELL, J.E. (1988). One factor recognizes the liver-specific enhancers in a1- antitrypsin and transthyretin genes. Science 239 :786.

GREGERSEN, N., WINTER, V., PETERSEN, K.8., KOCH, J., KOLVRAA, S., RUDIGER, N., HEINSVIG, E.M. & BOLUND, L. (1989). Detection of point mutations in amplified single copy genes by biotin-labelled oligonucleotides : diagnosis of variants of alpha-1-antitrypsin. Clin. Chim. Acta. 182:151.

179. GROSS, P., BUBYAK, M., TOLKEN, E. & KASHAK, M. (1964). Enzymatically produced pulmonary emphysema. A preliminary report. J. Occup. Med. 6 : 481.

GROSS, V., ANDUS, T., TRAN.THI, T-A., SCHWARZ, R.T., DECKER, K. & HEINRICH, P.C. (1983). 1-deoxynojirimycin impairs oligosaccharide processing of ø1-proteinase inhibitor and inhibits its secretion in primary cultures of rat hepatocytes. J. Biol. Chem. 258 : 12203.

GUNNING, P., PONTE, P., OKAYAMA, H., ENGEL, J., BLAU, H. & KEDES, L. (1983). lsolation and characterization of full-length cDNA clones for human a, p- and y-actin mRNAs : skeletal but not cytoplasmic actins have amino-terminal cysteine that is subsequently removed. Mol. Cell Biol. 3 : 787.

GUZDEK, 4., POTEMPA, J., DUBIN, A. & TRAVIS, J. (1990). Comparative properties of human alpha-1-proteinase inhibitor glycosylation variants. FEBS. Lett.272:125.

GYLLENSTEN, U.B. & ERLICH, H.A. (1988). Generation of single-stranded DNA by the polymerase chain reaction and its application to direct sequencing of the HLA-DQA locus. Proc. Natl. Acad. Sci. USA 85 :7652.

HAHN, M. (1897). Zur kenntniss der wirkungen des extravascularen blutes. Berl. Klin. Wochenschr. 34 :499. HAKANSSON, L. & VENGE, P. (1983). Partial characterization and identifícation of chemokinetic factors in serum. Scand. J. lmmunol. 18 : 531.

HARDON, E.M., FRAIN, M., PAONESSA, G. & CORTESE, R. (1988). Two distinct factors interact with the promoter regions of several liver-specific genes. EMBO J.7 :1711.

HARRIS, P.I., CHAPMAN, D., HARRISON, R.4., SMITH, K.F. & PERKINS, S.J. (1990). Conformational transition between native and reactive center cleaved forms of alpha-1-antitrypsin by Fourier transform inf rared spectroscopy and small-angle neutron scattering. Biochemistry 29 : 1377.

HEJTMANCIK, J.F., WARD, P.4., MANSFIELD, T., SIFERS, R.N., HARRIS, S. & COX, D.W. (1986). Prenatal diagnosis of a1-antitrypsin deficiency by restriction f ragment length polymorphisms and comparison with oligonucleotide probe analysis. Lancet ä :767.

HERCZ, A. (1985) Proteolytic cleavages in o1-antitrypsin and microheterogeneity. Biochem Biophys Res Commun 128 : 199.

HERCZ, A. & HARPAZ, N. (1980). Characterization of the oligosaccharides of liver Z variant ø1-antitrypsin. Can. J. Biochem. 58 : 644.

1 80. HERCZ, A., KATONA, E., CUTZ, E., WILSON, J.R. & BARTON, M. (1978). a1- antitrypsin : the presence of excess mannose in Z variant isolated from liver. Science 201 :1229.

HIRSCHBERG, C.B. & SNIDER, M.D. (1987). Topography of glycosylation in the rough endoplasmic reticulum and Golgi apparatus. Ann. Rev. Biochem. 56 :63.

HJALMARSSON, K. (1988). Distribution of alpha-1-antitrypsin phenotypes in Sweden. Hum. Hered. 38:27.

HODGES, J.R., MILLWARD.SADLER, G.H., BARBATIS, C. & WRIGHT, R. (1981). Heterozygous MZ Alpha-1-Antitrypsin Deficiency in Adults with Chronic Active Hepatitis and Cryptogenic Cirrhosis. N. Engl. J. Med. 304 : 557.

HODGSON, l. & KALSHEKER, N. (1986). RFLP for a gene-related sequence of alpha 1-antitrypsin (cr1AT). Nucl. Acids Res. 14 :6779.

HODGSON, l. & KALSHEKER, N. (1987). DNA polymorphisms of the human cr1 antitrypsin gene in normal subjects and in patients with pulmonary emphysema. J. Med. Genet. 24:47.

HOFKER, M.H., NELEN, M., KLASEN, E.C., NUKIWA, T., CURIEL, D., CRYSTAL, R.G. & FRANTS, R.R. (1988). Cloning and characterization of an alpha 1-antitrypsin like gene 12Kb downstream of the genuine alpha 1-antitrypsin gene. Biochem. Biophys. Res. Commun. 155 :634.

HOFKER, M.H., NUKIWA, T., VAN PAASSEN, H.M.B., NELEN, M., KRAMPS, J.4., KLASEN, E.C., FRANTS, R.R. & CRYSTAL, R.G. (1989). A Pro "" Leu substitution in codon 369 of the alpha-1-antitrypsin deficiency variant Pl M¡ss¡¡s¡. Hum. Genet. 81 :264. HOLMES, M., CURIEL, D., BRANTLY, M. & CRYSTAL, R.G. (1989a). Characterization of the intracellular mechanism causing the alpha-1- antitrypsin Nullsranite ra¡s deficiency state. Am. Rev. Respir. Dis. 140 : 1 662. HOLMES, M., BRANTLY, M., WURTS, L. & CRYSTAL, R.G. (1989b) Characterization of the sequence differences among the P-family of cr1- antitrypsin alleles. Am. Rev. Respir. Dis. 139 : 4202.

HOLMES, M.D., BRANTLY, M.1., CURIEL, D.T., WEIDINGER, S. & CRYSTAL, R.G. (1990a). Characterization of the normal cl1-antitrypsin allele Vmunich l a variant associated with a unique protein isoelectric focusing pattern. Am. J. Hum. Genet. 46 : 810.

HOLMES, M.D., BRANTLY, M.1,, FELLS, G.A. & CRYSTAL, R.G. (1990b). ø1- antitrypsin Wbethesda : molecular basis of an unusual o1-antitrypsin deficiency variant. Biochem. Biophys. Res. Commun. 17O:1013.

181 . HOLMES, M.D., BRANTLY, M.L. & CRYSTAL, R.G. (1990c). Molecular analysis of the heterogeneity among the P-family of alpha-1-antitrypsin alleles. Am. Rev. Respir. Dis. 1 42 :1185.

HOOD, J.M., KOEP, L.J., PETERS, R.L., SCHRÖTER, G.P.J., WEIL, R., REDEKER, A.G. & STARZL, T.E. (1980). Liver transplantation for advanced liver disease with alpha-1-antitrypsin deficiency. N. Engl. J. Med. 302 :272.

HORNE, S.1., TENNENT, R.K. & COCKCROFT. D.W. (1982). A new anodal alpha-1-antitrypsin variant associated with emphysema : Pi Bsaskaroon. Am. Rev. Respir. Dis. 125 :594.

HORNE, S.1., TENNENT, R.K., COCKCROFT, D.W., COTTON, D.J. & DOSMAN, J.A. (1986). Pulmonary function in Pi M and MZ grainworkers. Chest. 89 :795. HSU, S-M. & REE, H.J. (1980). Self-sandwich method. An improved immunoperoxidase technique for the detection of small amounts of antigen. Am. J. Clin. Pathol.74:32.

HUBBARD, R.C. & CRYSTAL, R.G. (1988). Alpha-1-antitrypsin augmentation therapy for alpha-1-antitrypsin deficiency. Am. J. Med. 84 (suppl. 6A) : 52.

HUBBARD, R.C., OGUSHI, F., FELLS, G.4., CANTIN,4.M., JALLAT, S., COURTNEY, M. & CRYSTAL, R.G. (1987). Oxidants spontaneously released by alveolar macrophages of cigarette smokers can inactivate the active site of o1-antitrypsin, rendering it ineffective as an inhibitor of neutrophil elastase. J. Clin. lnvest. 80 : 1289.

HUBBARD, R.C., SELLERS, S., CZERSKI, D., STEPHENS, L. & CRYSTAL, R.G. (1988). Biochemical efficacy and safety of monthly augmentation therapy for q1-antitrypsin deficiency. JAMA 260 : 1259.

HUBBARD, R.C., CASOLARO, M.A., MITCHELL, M., SELLERS, S.E., ARABIA, F., MATTHAY, M.A. & CRYSTAL, R.G. (1989a). Fate of aerosolized recombinant DNA-produced cr,1-antitrypsin : use of the epithelial surface of the lower respiratory tract to administer proteins of therapeutic importance. Proc. Natl. Acad. Sci. USA. 86 : 680.

HUBBARD, R.C., BRANTLY, M.1., SELLERS, S.E., MITCHELL, M.E. & CRYSTAL, R.G. (1989b). Anti-neutrophil-elastase defenses of the lower respiratory tract in c1-antitrypsin deficiency directly augmented with an aerosol of cr1-antitrypsin. Ann. lntern. Med. 1 1 1 : 206.

HUBBARD, R.C., McELVANEY, N.G., SELLERS, S.E., HEALY, J.T.,CZERSKI, D.B. & CRYSTAL, R.G. (1989c). Recombinant DNA-produced 4,1- antitrypsin administered by aerosol augments lower respiratory tract antineutrophil elastase defenses in individuals with cr1-antitrypsin deficiency. J. Clin. lnvest. 84 : 1349.

182. HUBER, R. & CARRELL, R.W. (1989). lmplications of the three-dimensional structure of o1-antitrypsin for structure and function of serpins. Biochemislry 28: 8952.

HUG, G., CHUCK, G., BOWLES, B.& FAGERHOL, M.K. (1980a). Pizpratt'a new alpha-1-antitrypsin allele in an American Negro family. Hum. Genet. 56:107.

HUG, G., CHUCK, G., SLEMMER, T.M. & FAGERHOL, M.K. (1980b). Pi Ecincinnati : a new alpha-1-antitrypsin allele in three Negro families. Hum. Genet. 54 : 361.

HUG, G., CHUCK, G. & FAGERHOL, M. (1981) PiPcl¡rton : a new a1-antitrypsin allele in an American Negro Family. J. Med. Genet. 18 : 43.

HUG, G., CHUCK, G. & BOWLES, B. (1982a). Alpha-1-antitrypsin phenotype : transient cathodal shift in serum of infant girl with urinary cytomegalovirus and fatty liver. Pediatr. Res. 16 : 192.

HUG, G., CHUCK, G. & McGlLL, M.(1982b). Piwfinneytwon : anewalpha-1- antitrypsin allele in an American Negro family. Hum. Hered. 32 :280.

HULTCRANTZ, R. & MENGARELLI, S. (1984). Ultrastructural liver pathology in patients with minimal liver disease and a1-antitrypsin deficiency : a comparison between heterozygous and homozygous patients. Hepatol. 4 :937.

HUMPHRIES, R.K., LEY, T.J., ANAGNOUS, N.P., BAUR, A.W. & NEINHUIS, A.W. (1984). ßo-39 thalassemia gene : a premature termination codon causes ß-mRNA deficiency without affecting cytoplasmic ß-mRNA stability. Blood 64 :23.

HUNNINGHAKE, G.W. & CRYSTAL, R.G. (1983). Cigarette smoking and lung destruction : accumulation of neutrophils in the lungs of cigarette smokers. Am. Rev. Respir. Dis. 128 : 833

HUNNINGHAKE, G.W., GADEK, J.E., KAWANAMI, O., FERRANS, V.J. & CRYSTAL, R.G. (1979). lnflammatory and immune processes in the human lung in health and disease : evaluation by bronchoalveolar lavage. Am. J. Pathol. 97 :149. HUNT, L.T. & DAYHOFF, M.O. (1980) A surprising new super family containing ovalbumin, antithrombin-lll, and alpha 1 -proteinase inhibitor. Biochem. Biophys. Res. Commun. 95 :864. HUTCHISON, D.C. (1990). The epidemiology of alpha-1-antitrypsin deficiency. Lung 168(suppl) : 535. HUTCHINSON, D.C.S., TOBIN, M.J. & COOK, P.J.L. (1983). Alpha-l- antitrypsin deficiency : clinical and physiological features in heterozygotes of Pi type SZ. Br. J. Dis. Chest 77 :28.

1 83. IBARGUEN, E., GROSS, C.R., GAV|K, S.K. & SHARP, H.L. (1990). Liver disease in alpha-1-antitrypsin deficiency : prognostic indicators. J. Pediatr. 117 :864.

ISAACSON, P., JONES, D.8., MILLWARD-SADLER, G.H., JUDD, M.A. & PAYNE, S. (1981). Alpha-1-antitrypsin in human macrophages. J. Clin. Pathol. 34 :982.

ISH-HOROWICZ, D. & BURKE, J.F. (1981). Rapid and efficient cosmid cloning. Nucl. Acids Res.9 :2989.

JACOBSSON, K. (1953). Electrophoretic demonstration of two trypsin inhibitors in human blood serum. Scand. J. Clin. Lab. lnvest. 5 : 97.

JACOBSSON, K. (1955). Studies on fibrinogen ll. Studies on trypsin and plasmin inhibitors in human blood serum. Scand. J. Clin. Lab. lnvest. 14 (Suppl.) : 57.

JALLAT, S., CARVALLO, 0., TESSIER, 1.H., ROECKLIN, 0., ROITSCH, C., oGUSHl, F., CRYSTAL, R.G. & COURTNEY, M. (1986). Altered specificities of genetically engineered q1-antitrypsin variants. Protein Engineering. 1 :29.

JANOFF, A. (1985) Elastases and emphysema. Current assessment of the protease-antiprotease hypothesis. Am. Rev. Respir. Dis. 132:417.

JANOFF, A. & SCHERER, J. (1968). Mediators of inflammation in leukocyte lysosomes. lX. Elastinolytic activity in granules of human polymorphonuclear leukocytes. J. Exp. Med. 128 : 1137. JANOFF, A. & DEARING, R. (1980). Prevention of elastase-induced experimental emphysema by oral administration of a synthetic elastase inhibitor. Am. Rev. Respir. Dis. 121 :1025.

JANOFF, A., SLOAN, 8., WEINBAUM, G., DAMIANO, V., SANDHAUS, R.4., ELIAS, J. & KIMBELL, P. (1977). Experimental emphysema induced with purified human neutrophil elastase. Tissue localization of the instilled protease. Am. Rev. Resp. Dis. 115:461.

JANOFF,4., CARP, H., LEE, D.K. & DREW, R.T. (1979). Cigarette smoke inhalation decreases alpha-1-antitrypsin activity in rat lung. Science 206 : 1313.

JANOFF, 4., GEORGE-NASCIMENTO, C. & ROSENBERG, S. (1986). A genetically engineered, mutant human alpha-1-proteinase inhibitor is more resistant than the normal inhibitor to oxidative inactivation by chemicals, enzymes, cells and cigarette smoke. Am. Rev. Respir. Dis. 133 :353.

JANUS, E.D. (1988). Alpha-1-antitrypsin Pi types in COPD patients. Chest. 94 : 447.

184 JANUS, E.D. & BURDON, J.G.W. (1990). Alpha-1 -antitrypsin deficiency and emphysema : new horizons in treatment Aust. N.Z. J. Med. 20:755.

JANUS, E.D., JOYCE, P.R., MERLE-SHEAT, J. & CARRELL, R.W. (1975)' Alpha-1-antitrypsin variants in New Zealand. N.Z. Med. J. 82 : 289.

JANUS, E.D., PHILLIPS, N.T. & cARRELL, R.W. (1985). smoking, lung function, and cr,1-antitrypsin deficiency. Lancet i : 152.

JAYARAMAN, K. (1987). A rapid isolation procedure for oligonucleotide purification. Biotechniques 5 : 627.

JEPPSSON, J-O. (1976). Amino acid substitution Glu > antitrypsin Pi Z. FEBS Lett 65 : 195.

JEPPSSON, J-O. & FRANZEN, B. (1982). Typing of genetic variants of alpha 1-antitrypsin by electrofocusing. Clin. Chem' 28:219.

JEPPSSON, J.O. & SVEGER, T.(1984). Typing of genetic variants of cr1- antitrypsin from dried blood. J. Clin' Lab' lnvesl. 44 :413. JEppssoN, J-o. & LAURELL, C-8. (1988). The amino acid substitutions of human ø1-antitrypsin M3, XandZ. FEBS' Lett. 231 :327.

JEPPSSON, J-O., LARSSON, C. & ERIKSSON, S. (1975)' Characterization of g1-antitrypsin in the inclusion bodies from the liver in cr1-antitrypsin deficiency. N. Engl. J. Med. 293 :576.

JEPPSSON, J-O., LAURELL, C-8. & FAGERHOL, M.K. (197_8a). Properties of isolated human alpha 1-antitrypsin of Pi types M, S and Z. Eur. J. Biochem. 83 : 143.

JEPPSSON, J-O., LAURELL, C-8., NOSSLIN, 8., COX, D.W. (1978b). Catabolic rate of g1-antitrypsin of Pi types S, and Mmalton and of asialylated M-protein in man. Clin' Sci. Mol. Med.55 :103. JEPPSSON, J.O., FRAN_Z_ËNr 8., SVEGER, T., CORDESIUS, E., STRöMBERG, p. & GUSTAVil, B. (1979). Prenatal exclusion of alpha-1- antitrypsin deficiency in a high-risk fetus. N. Engl. J' Med. 300 : 1441- 8.. LOFBERG, L., FRANZEN, JEPPSSON,..J.O., CORDESIUST _E., GUSTAVI!, 8., STRöMBERG, P. & SVEGER, T. (1981). Prenatal diagnosis of alpha- 1-ântitrypsin deficiency by analysis of fetal blood obtained at fetoscopy' Pediatr. Res. 15 :254.

JEPPSSON, J-O., LILJA, H. & JOHANSSON, M. (1985) lsolation and characterization of two minor fractions of a1-antitrypsin by high- performance liquid chromatographic chromatofocusing. J. Chromatogr. 327 :173.

185 JOHNSON, A.M. (1976). Genetic typing of a1-antitrypsin by immunofixation electrophoresis. ldentification of subtypes of Pi M. J. Lab. Clin. Med. 87 : 152.

JOHNSON, D.A. & TRAVIS, J. (1976). Human alpha-1-proteinase inhibitor mechanism of action : evidence for activation by limited proteolysis. Biochem. Biophys. Res. Commun. 72 :33.

JOHNSON, D. & TRAVIS, J. (1978). Structural evidence for methionine at the reactive site of human a1-proteinase inhibitor. J. Biol. Chem.253 :7142.

JOHNSON, D. & TRAVIS, J. (1979). The oxidative inactivation of human cr1- proteinase inhibitor. J. Biol. Chem. 254:4022.

JOHNSON, P.H., CADIOU, H. & HOPKINSON, D.A. (1991). Detection of the common alpha-1-antitrypsin variants by denaturing gradient gel electrophoresis. Ann. Hum. Genet.55 : 183.

JONES, D.K., GODDEN, D. & CAVANAGH, P. (1985). Alpha-1-antitrypsin deficiency presenting as bronchiectasis. Br. J. Dis. Chest. 79 : 301. JONES, 8.A., VERGALLA, J., STEER, C.J., BRADLEY-MOORE, P.R. & VIERLING, J.M. (1978). Metabolism of intact and desialylated a1 - antitrypsin. Clin. Sci. Mol. Med.55: 139.

JOSLIN, G., FALLON, R.J., BULLOCK, J., ADAMS, S.P. & PERLMUTTER, D.H. (1991). The SEC receptor recognizes a penlapeptide neod^omain of àtpna-t-antitrypsin-protease complexes. J. Biol. Chem. 266 : 11282.

KALSHEKER, N. & SWANSON, T. (1990). Exclusion of an exon in monocyte alpha-1-antitrypsin mRNA after stimulation of U937 cells by interleukin-6. Biochem. Biophys. Res. Commun. 172:1116.

KALSHEKER, N. & MORGAN, K. (1990). The c¡1-antitrypsin gene and chronic lung disease. Thorax 45 :759. KALSHEKER, N.4., HODGSON, I.J., WATKINS, O.1., WHITE, J.P., MORRISON, H.M. & STOCKLEY, R.A. (1987). Deoxyribonucleic acid (DNA) polymorphism of the alpha-1-antitrypsin gene in chronic lung disease. Br. Med. J.. 294 : 151 1 . KAY, M.4., BALEY, P., ROTHENBERG, S., LELAND, F., FLEJVIING, L., PONDER, K.P., LIU, T., FINEGOLD, M., DARLINGTON, G., POKORNY, W. & WOO, S.L.C. (1992). Expression of human alpha-1-antitrypsin in dogs after autologous transplantation of retroviral transduced hepatocytes. Proc. Natl. Acad. Sci. USA. 89 :89.

KELLER, G., PAIGE, C., GILBOA, E. & WAGNER, E.F. (1985). Expression of a foreign gene in myeloid and lymphoid cells derived from multipotent haematopoietic precursors. Nature 318 : 149.

186 KELLERMANN, G. & WALTER, H. (1970). lnvestigations on the population genetics of the cr1-antitrypsin polymorphism. Humangetik 10 : 145.

KELLY, C.P., TYRRELL, D.N., McDONALD, G.S.A., WHITEHOUSE, D.B' & PRICHARD, J.S. (1989). Heterozygous FZ cr1-antitrypsin deficiency associated with severe emphysema and hepatic disease : a case report and family study. Thorax 44:758.

KELLY, R.B. (1985). Pathways of protein secretion in eukaryotes. Science 230 :25.

KELSEY, G.D., POVEY, S., BYGRAVE, A.E. & LOVELL-BADGE, R.H. (1987). Species and tissue-specific expression of human alpha 1-antitrypsin in transgenic mice. Genes Dev. 1 : 161.

KELSEY, G.D., PARKAR, M., POVEY, S. (1988). The human a1-antitrypsin related sequence gene : lsolation and investigation of its expression. Ann. Hum. Genet. 52 : 151.

KEYEUX, G., GILGENKRANTZ, S., LEFRANC, G. & LEFRANC, M.P. (1989). Molecular characterization of a ring chromosome 14 showing that the Pl locus is centromeric to the D14Sl and IGH loci. Hum. Genet. 82 : 219. KHAGHANI, A., BANNER, N., OZDOGAN, 8., MUSUMECI, F., THEODOROPOULOS, S., ARAVOT, D., FITZGERALD, M. & YACOUB, M. (1991). Medium-term results of combined heaft and lung transplantation for emphysema. J. Heart Lung Transplant 10(1) : 15.

KIDD, V.J., WALLACE, R.8., ITAKURA, K. & WOO, S.L.C. (1983) Alpha 1- antitrypsin deficiency detection by direct analysis of the mutation in the gene. Nature 304 :230.

KIDD, V.J., GOLBUS, M.S., WALLACE, R.8., ITAKURA, K. & WOO, S.L.C. (1984). Prenatal diagnosis of a1-antitryps¡n deficiency by direct analysis of the mutation site in the gene. N. Engl. J. Med. 310 :639.

KLASEN, E.C. & RIGUTTI, A. (1982). lsoelectric focusing of q,1-antitrypsin (Pi) using restricted pH-range carrier ampholytes in combination with a highly cross-linked gel and a separator. Electrophoresis 3 : 168.

KLASEN, E.C., BOS, A. & SIMMELINK, H.D. (1982). Pi (cr1-antitrypsin) subtypes : frequency of Pi M4 in several populations. Hum. Genet. 62 : 1 39.

KLASEN, E.C., BIEMOND, l. & LAROS, C.D. (1986). Alpha-1-antitrypsin deficiency and the flaccid lung syndrome. The heterozygote controversy. Clin. Genet. 29 :211.

187. KLASEN, E.C., HOFKER, M.H., VAN PAASSEN, H.M., VERLAAN DE VRIES, M., BOS, J.L. & FRANTS, R.R. (1987). Detection of alpha-1-antitrypsin deficiency variants by synthetic oligonucleotide hybridization. Clin. Chim Acta. 170 :201.

KLATT, E.C., KOSS, M.N., YOUNG, T.S., MACAULEY, L. & MARTIN, E. (1988). Hepatic hyaline globules associated with passive congestion. Arch. Pathol. Lab. Med. 112 : 510.

KLEINERMAN, J., RANGA, V., RYNBRANDT, D., IP, M.P.C., SORENSEN, J. & POWERS, J.C. (1980). The effect of the specific elastase inhibitor, alanyl alanyl prolyl alanine chloromethylketone, on elastase-induced emphysema.. Am. Rev. Respir. Dis. 121 :381.

KOJ,4., REGOECZI, E., TOEWS, C.J., LEVEILLE, R. & GAULDIE, J.(1978). Synthesis of antithrombin lll and alpha-1-antitrypsin by perfused rat liver. Biochim. Biophys. Acta. 539 :496.

KOOPMAN, P., POVEY, S. & LOVELL-BADGE, R.H. (1989). Widespread expression of human alpha-1-antitrypsin in transgenic mice revealed by in situ hybridization. Genes. Dev. 3 : 16-25. KREIG, P.A. & MELTON, D.A. (1984). Functional messenger RNAs are produced by SP6 in vitro transcription of cloned cDNAs. Nucl. Acids Res. 12 :7057.

KRlVlT, W., MILLER, J., NOWICKI, M. & FREIER, E. (1988). Contribution of monocyte-macrophage system to serum alpha-1-antitrypsin. J. Lab. Clin. Med. 1 12 :437.

KUEPPERS, F. (1978). lnherited differences in alpha-1-antitrypsin. ln : Litwin, S.D. (ed). Genetic determinants of pulmonary disease. Marcel Dekker, New York , 23.

KUEPPERS, F. & BEARN, A. (1966). A possible experimental approach to the association of hereditary alpha 1-antitrypsin deficiency and pulmonary emphysema. Proc. Soc. Exp. Biol. Med. 121 :1207.

KUEPPERS, F. & FALLAT, R.J. (1969) Alpha-1-antitrypsin deficiency : a defect in protein synthesis. Clin. Chim. Acta. 24 :401.

KUEPPERS, F. & BLACK, L.F.(1974) Alpha 1-antitrypsin and its deficiency. Am. Rev. Respir. Dis.1 10 : 176.

KUEPPERS, F. & DONHARDT, A. (1974). Obstructive lung disease in heterozygotes for alpha-1-antitrypsin deficiency. Ann. lntern. Med. 80 : 209.

KUEPPERS, F. & CHRISTOPHERSON, M.J. ( 1978). Alpha-1-antitrypsin : further genetic heterogeneity revealed by i soelectric focusing. Am. J. Hum. Genet. 30 :359.

188 KUEPPERS, F., BRISCOE, W.A, & BEARN, A.G. (1964). Hereditary deficiency of alpha 1-antitrypsin. Science 146 : 1678.

KUEPPERS, F., FALLAT, R. & LARSON, R.K. (1969). Obstructive lung disease and o1-antitrypsin deficiency gene heterozygosity. Science 165 : 899.

KUEPPERS, F., UTZ, G. & SIMON, B. (1977). Alpha-1-antitrypsin deficiency with M-like phenotype. J. Med. Genet. 14 : 183.

KUGLER, W., WAGNER, U. & RYFFEL, G.U. (1988). Tissue-specificity of liver gene expression : a common liver-specific promoter element. Nucl. Acids Res. 16:3165.

KÜHNL, P. & SPIELMANN, W.(1979). PiT : A New allele in the alpha-1- antitrypsin system. Hum. Genet. 50 :221. KURACHI, K., CHANDRA, T., FRIEZNERDEGEN, S.J., WHITE, T.T., MARCHIORO, T.L., WOO, S.L.C. & DAVIE, E.W' (1981)' Cloning and sequence of the cDNA coding for alpha 1-antitrypsin. Proc. Natl. Acad. Sc. USA 78 :6826.

LAEMMLI, U.K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 :680'

LAl, E.C., KAO, F.T., LAW, M.L. & WOO, S.L.C. (1983)' Assignment of the alpha 1-antitrypsin gene and a sequence related g_ene to^ human chromosome 14'by mõlecular hybridization. Am. J. Hum. Genet. 35 :385.

LARSSON, C. (1978). Natural history and life expectancy in severe alpha-1- antitrypsin deficiency , Pi Z. Acla Med. Scand.204:345.

LARSSON, C., DIRKSEN, H., SUNDSTRÖM, G. & ERIKSSON, S. (1976_), Lung functíon studies in asymptomatic individuals with moderately (Pi SZ) and severely (Pi Z) reduced levels of o1-antitrypsin. Scand. J. Resp. Dis. 57:267. LARSSON, C., ERIKSSON, S. & DIRKSEN, H. (1977). Smoking and intermediate alpha-1-antitrypsin deficiency and lung function in middle- aged men. Br. Med. J.2:922.

LAURELL, C-8. (1965). Electrophoretic microheterogeneity of serum a1- antitrypsin. Scand. J. Clin. Lab. lnvesl. 17 :271.

LAURELL, C-8. (1966). Quantitative estimation of proteins by electrophoresis in agarose-gel containing antibodies. Anal. Biochem. 15 :45.

LAURELL, C.B. & ERIKSSON, S. (1963). The electrophoretic alpha 1-globulin pattern of serum in alpha 1-antitrypsin deficiency. Scand. J. Clin. Lab. lnvest. 15 : 132.

1 89. LAURELL, C-8. & PERSSON, U. (1973). Analysis of plasma o1-antitrypsin variants and their microheterogeneity. Biochim. Biophys. Acta. 310 : 500.

LAURELL, C-8. & JEPPSSON, J-O. (1975). Protease inhibitors in plasma. Putnam F. (ed) ln : The Plasma Proteins. Vol. l. Academic Press, New York ,229. LAURELL, C-8., KULLANDER, S. & THORELL, J. (1967) Effect of administration of a combined estrogen-progestin contraceptive on the level of individual plasma proteins. Scand, J. Clin. Lab. lnvesl.21 :337.

LAURELL, C-8., NOSSLIN, B. & JEPPSSON, J-O. (1977). Catabolic rate of cr1-antitrypsin of Pi type M and Z in man. Clin. Sci. Mol. Med. 52 : 457.

LE, A., FERRELL, G.A., DISHON, D.S., LE, Q.O. & SIFERS, R.N. (1992). Soluble aggregates of the human PiZ alpha-1-antitrypsin variant are degraded within the endoplasmic reticulum by a mechanism sensitive to inhibitors of protein synthesis. J. Biol. Chem. 267 :1072.

LE, 4., GRAHAM, K.S. & SIFERS, R.N. (1990). lntracellular degradation of the transport-impaired human PiZ u1-antitrypsin variant : biochemical mapping of the degradative event among compartments of the secretory pathway. J. Biol. Chem.265:14001.

LEDLEY, F.D. & WOO, S.L. (1986). Molecular basis of alpha-1-antitrypsin deficiency and its potential therapy by gene transfer. J. lnherited Metab. Dis. 9 :85. LEDLEY, F.D., DARLINGTON, G.J., HAHN, T. & WOO, S.L.C. (1987). Retroviral gene transfer into primary hepatocytes : implications for genetic therapy of liver-specific functions. Proc. Natl. Acad. Sci. USA. 84 : 5335.

LEE, C.C., KUEPPERS, F., HARPEL, B. & RODGERS, G. (1981). Alpha-1- antitrypsin (Pi) types in Korean and Chinese populations. Hum. Genet. 57 :327. LEHRMAN, M.4., SCHNEIDER, W.J., BROWN, M.S., DAVID, C.G., ELHAMMER, 4., RUSSELL, D.W., GOLDSTEIN, J.L. (1987) Alu-Alu recombination deletes splice acceptor sites and produces secreted low density lipoprotein-receptor in a subject with familiall hypercholesterolemia. . J. Biol. Chem.262: 401.

LEICHT, M., LONG, G.L.,CHANDRA, T., KURACHI, K., KIDD, V.J., MACEJT., M., DAVIE, E.W. & WOO, S.L.C. (1982). Sequence homology and structural comparison between the chromosomal human alpha-1- antitrypsin and chicken ovalbumin genes. Nature 297 :655. LEWIS, H.J., IAMMARINO, R.M., SPERO, J.A. & HASIBA, V. (1978). Antithrombin Pittsburgh : an cr1-antitrypsin variant causing hemorrhagic disease. Blood 51 :129.

1 90. LEWIS, M., KALLENBACK, J.,ZALTZMAN, M., LEVY, H., LURIE, D.,.BAYNES, R.,'KtñG, p. & MEYERS, A. (1985). Severe deficie.n.cy of alpha-1- aniitrypsin associated with cutaneous vasculitis, rapidly progressive glomérulonephritis, and colitis. Am. J' Med. 79 :489. Ll,' Y., SHEN, R.F., TSAI, S.Y. & WOO, S.L.C. (1988). Multiple hepatic trans- aôting factors are required for in vitrotranscription of the human alpha-1- antitrypsin gene. Mol. Cell Biol. I :4362.

LIEBERMAN, J. (1969). Heterozygous and homozygo_u.s glpla-.1-antitrypsin deficiency in þatienis with pulrirónary emphysema' N. Engl' J. Med. 281 : 279. LIEBERMAN, J. (1972). Digestion of antitrypsin-deficient lungs by leukoprotôases. ln' : Mitman, C. (ed). Pulmonary emphysema and proteolysis. Academic Press, New York, 183'

LIEBERMAN, J. (1973). Heat lability of alpha-1-antitrypsin variants. Chest 64 :579.

LIEBERMAN, J. (19S3). Alpha-1-antitrypsin deficiency and related disorders. Princip. Prac. Med. Genet. 2 : 911.

LIEBERMAN, J., GAIDULIS, L. & KLOTZ, S.D. (1976). A new deficient variant of 9¿1-antitrypsin (M6r"1"). lnability to detect the heterozygous state by antitrypsin phenotyping. Am. Rev' Resp. Dis. 113:31.

LIEBERMAN, J., WINTER, B. & SASTRE, A. (1986). Alpha-1-antitrypsin Pi- types in 965 COPD patients. Chest 89 : 371.

LITTLETON, E.T., BEVIS, L., HANSEN, L.J., PEAKMAN, M., MOWAT, 4.P., MtEL|-VERGÂNt, G. & VERGANI, D. (1991). Alpha-1-antitrypsin deficiency, compiement activation, and chronic liver disease. J. Clin. Pathol. 44 :855.

LLOYD, C. & TRAVIS, J. (1989). Rapid screening for deficiency of alpha 1- proteinase inhibitor. Clin. Chem. 35 : 1971.

LODISH, H.F. (1988). Transport of secretory an_d membrane glycoproteins trom the rouigh enOoplasmic reticulum to the Golgi. J. Biol. Chem. 263 : 2107. LODISH, H.F. & KONG, N. (1984) Glucose removal from N-linked oligosaccharides is required for efficienct maturation of certain secretory gly"coproteins from the iough endoplasmic reticulum to the golgi complex. J. Cell Biol. 98 :1720. LOEBERMANN, H., LOTTSPEICH, F., BODE, W. & HUBER, R' (1982)' lnteraction of human c¡1-proteinase inhibitor with chymotrypsinogen A and crystallization of a proteolytically modified cr1-proteinase inhibitor. Hoppe-Seyler's Z. Physiol. Chem. 363 : 1377.

191 . LOEBERMANN, H., TOKUOKA, R., DE¡SENHOFER, J. & HUBER, R. (1984) Human alpha 1-prote¡nase inhibitor : crystal structure analysis of two crystal modifications, molecular model and preliminary analysis of the implications for function. J. Mol. Biol. 177 :531.

LONG, C., CHANDRA, T., WOO, S.L.C., DAVIE, E.W. & KURACHI, K.(1984). Complete sequence of the cDNA for human alpha 1-antitrypsin and the gene for the S variant. Biochemislry 23 : 4828.

LONGSTRETH, G.F., WEITZMAN, S.4., BROWNING, R.J. & LIEBERMAN, J. (1975). Bronchiectasis and homozygous alpha-1-antitrypsin deficiency. Chest. 67 :233.

MACKAY, I.R., OLIPHANT, R.C., LABY, 8., SMITH, M.M., FISHER, J.N., MITCHELL, R.J., PROPERT, D.N. & TAIT, B.O. (1990). An immunologic and genetic study of asthma in workers in an aluminum smelter. J. Occup. Med. 32 :1022.

MACKIEWICZ, 4., GANAPATHI, M.K., SCHULTZ, D., BRABENEC, 4., WEINSTEIN, J., KELLY, M.F. & KUSHNER, l. (1990). Transforming growth factor beta 1 regulates production of acute-phase proteins. Proc. Natl. Acad. Sci. USA. 87 :1491.

MAKINO, S. & REED, C.E. (1970). Distribution and elimination of exogenous alpha 1-antitrypsin. J. Lab. Clin. Med.75 :742.

MALONE, M., MIELI-VERGANI, G., MOWAT, A.P. & PORTMANN, B. (1989). The fetal liver in P\ZZ alpha-1-antitrypsin deficiency : a report of five cases. Pediatr. Pathol. 9 :623. MANCINI, G., CARBONARA, A.O. & HEREMANS, J. F. (1e65). lmmunochemical quanititation of antigens by single radial immunodiffusion. lmmunochemistry 2 : 235.

MANGIONE, S., KUEPPERS, F., PUGLIA, C. & GREENSPON, L.W. (1991). Erythrocytes prevent inactivation of alpha 1-antitrypsin by cigarette smoke. Eur. Respir. J. 4 :26.

MANIATIS, T., FRITSCH, E.F. & SAMBROOK, J. (1982). Molecular cloning : a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

MAQUAT, 1.E., KINNIBURGH,4.J., RACHMILEWITZ, E.A. & ROSS, J. (1981). Unstable ß-globin mRNA in mRNA-deficient ßo thalassemia. Cell27 : 543.

MARTIN, J-P., VANDEVILLE, D. & ROPARTZ, C. (1973). PiB, a new allele of cr1 antitrypsin genetic variants. Biomedicine 19 : 395.

MARTIN, J-P., SESBOUE, R., CHARLIONET, R. & ROPARTZ, C. (1975). Does alpha-1-antitrypsin P¡ Null phenotype exist? Humangenetik 30 : 121.

192. MASS,8., IKEDA, T., MARCO, V., MERANZE, D.R.& KIMBEL, P. (1972). lnduction of experimental emphysema : cellular and species specificity. Am. Rev. Resp. Dis. 106 : 384.

MASSOUD, M., BISCHOFF, R., DALEMANS, W., POINTU, H., ATTAL, J., SCHULTZ, H., CLESSE, D., STINNAKRE, M.G., PAVIRANI, A. & HOUDEBINE, L.M. (1990). Production of human proteins in the blood of transgenic animals. C.R. Acad. Sci. 311 :275.

MATHESON, N.R., WONG, P.S. & TRAVIS, J. (1979). Enzymatic inactivation of human alpha-1-proteinase inhibitor by neutrophil myeloperoxidase. Biochem. Biophys. Res. Commun. 88 :402.

MATHESON, N.R., WONG, P.S., SCHUYLER, M. & TRAVIS, J. (1981). lnteraction of human a 1-proteinase inhibitor with neutrophil myeloperoxidase. Biochemistry 20 : 331.

MATHESON, N.R., GIBSON, H.1., HALLEWELL, R.A., BARR, P.J. & TRAVIS, J. (1986). Recombinant DNA-derived forms of human o1-proteinase inhibitor. J. Biol. Chem. 261 :10404.

MATSUNAGA, E., SHIOKAWA, S., NAKAMURA, H., MURAYAMA, T., TSUDA, K. & FUKAMAKI, Y. (1990). Molecular analysis of the gene of the cr1- antitrypsin deficiency variant, Mnichinan. Am. J. Hum. Genet. 46 :602.

McCRACKEN, 4.4., KRUSE, K.B. & BROWN, J.L. (1989). Molecular basis for defective secretion of the Zvarianl of human alpha 1-proteinase inhibitor : secretion of variants having altered potential for salt bridge formation between amino acids 290 and342. Mol. Cell Biol. 9 : 1406. McCRACKEN,4.4., KRUSE, K.8., VALENTINE, J., ROBERTS, C', YOHANNES, T.Z. & BROWN, J.L. (1991). Construction and expression of a1-proteinase inhibitor mutants and the effects of these mutations on secretion of the variant inhibitors. J. Biol. Chem. 266 : 7578. McPHlE, J.1., BlNNlE, S. & BRUNT, P.W. (1976). Alpha-1-antitrypsin deficiency. An infantile liver disease. Arch. Dis. Child. 51 : 584. MEGA, T., LUJAN, E. & YOSHIDA, 4., (19S0a). Studies on the oligosaccharide chains of human cx,1-protease inhibitor. L lsolation of glycopeptides. J. Biol. Chem. 255 :4053.

MEGA, T., LUJAN, E. & YOSHIDA, A.(1980b). Studies on the oligosaccharide chains of human a1-protease inhibitor. ll. Structure of oligosaccharides. J. Biol. Chem. 255 : 4057.

MEIJERS, J.C. & CHUNG, D.W. (1991). Organization of the gene coding for human protein C inhibitor (plasminogen activator inhibitor-3). Assignment of the gene to chromosome 14. J. Biol. Chem. 266 : 15028.

1 93. MELTON, D.4., KREIG, P.4., REGABLIATI, M.R., MANIATIS, T., ZINN, K. & GREEN, M.R. (1984). Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids conta¡ning a bacteriophage SP6 promotor. Nucl. Acids Res. 12 :7035.

MIER, J.W., DINARELLO, C.4., ATKINS, M.8., PUNSAL, P.I. & PERLMUTTER, D.H. (1987). Regulation of hepatic acute phase protein synthesis by products of lnterleukin 2 (ll 2) stimulated human peripheral blood mononuclear cells. J. lmmunol. 139 : 1268.

MILLER,4.D., LAW, M-F. & VERMA, LM. (1985). Generation of helper-free amphotrophic retroviruses that transduce a dominant-acting methotrexate- resistant dihydroflolate reductase gene. Mol. Cell Biol. 5 : 431. MILLER, F. & KUSCHNER, M. (1969). Alpha-1-antitrypsin deficiency, emphysema, necrotizing angiitis and glomerulonephritis. Am. J. Med. 46 : 61 5.

MITCHELMORE, C., TRABONI, C. & CORTESE, R. (1991). lsolation of two cDNAs encoding zinc finger proteins which bind to the alpha-1-antitrypsin promoter and to the major histocompatibility complex class I enhancer. Nucl. Acids Res. 19 : 141 .

MIYAKE, K., SUZUKI, H., OKA, H., ODA, T. & HARADA, S. (1979). Distribution of c¡1-antitrypsin phenotypes in Japanese : description of PiM subtypes by isoelectric focusing. Jpn. J. Human. Genet. 24 :55.

MOLL, F.C., SINDEN, S.F. & BROWN, J.R. (1958). Partial purification of the serum trypsin inhibitor. J. Biol. Chem. 233 :121.

MONACI, P., NICOSIA, A. & CORTESE, R. (1988). Two different liver-specific factors stimulate in vitro transcription from the human ø1-antitrypsin promoter. EMBO. J.7 :2075.

MORNEX, J-F., CHYTIL-WEIR, 4., MARTINET, Y., COURTNEY, M., LECOCQ, J-P. & CRYSTAL, R.G. (1986). Expression of the alpha 1-antitrypsin gene in mononuclear phagocytes of normal alpha 1-antitrypsin-deficient individuals. J. Clin. lnvest. 77 :1952.

MOROZ, S.P., CUTZ, E., COX, D.W. & SASS-KORTSAK, A. (1976a). Liver disease associated with alpha-1-antitrypsin deficiency in childhood. J. Pediatr. 88 : 19.

MOROZ, S.P., CUTZ, E., BALFE, J.W. & SASS-KORTSAK, A. (1s76b). Membranoproliferative glomerulonephritis in childhood cirrhosis associated with alpha-1-antitrypsin deficiency. Pediatr. 57 :232. MORRISON, H.M. (1987). The proteinase-antiproteinase theory of emphysema : time for a reappraisal? Clin. Sci. 72 :151.

MORSE, J.O. (1978) Alpha 1-antitrypsin deficiency. N. Engl. J. Med. 299 : 1 045.

194 MORSE, J.O., LEBOWITZ, M.D., KNUDSON, R.J. & BURROWS, B. (1977). Relation of protease inhibitor phenotypes to obstructive lung diseases in a community. N. Engl. J. Med. 296 : 1190.

MULLEY, J.C. (1982). Distribution of alpha-1-antitrypsin (Pi) subtypes from newborne cord blood. Aust. J. Exp. Biol. Med. Sci. 60 :207.

MULLEY, J.C., COX, D.W.& SUTHERLAND, G.R. (1983). A new allele of o1- antitrypsin : PiN ADELA¡DE. Hum. Genet. 63 :73.

MULLIS, K.B. & FALOONA, F.A. (1987). Specific synthesis of DNA in vitrovia a polymerase-catalyzed chain reaction. Methods Enzymol. 155 : 335.

NEBB|A, G., HADCHOUEL, M. ODIÈVnr & ALAGILLE, D. (1983). Early assessment of evolution of liver disease associated with cr1-antitrypsin deficiency in childhood. J. Pediatr. 102 :661. NEVO, S. & CLEVE, H. (1991). Pi polymorphism in lsrael :report on six Jewish population groups. Am. J. Med. Genet. 39 :399. NEWTON, C.R., GRAHAM,4., HEPTINSTALL, 1.E., POWELL, S.J., SUMMERS, C., KALSHEKER, N., SMITH, J.C. & MARKHAM, A.F. (1989). Analysis of any point mutation in DNA. The amplification refractory mutation system (ARMS). Nucl. Acids Res. 17 :2503.

NOBLE-JAMIESON, G., BARNES, N.D., THIRD, S. & MOWAT, A.P. (1990). Severe hypertension after liver transplantation in alpha-1-ant¡trypsin deficiency. Arch. Dis. Child. 65 : 1217.

NOWAK, D. & RUTA, U. (1990). Nicotine inhibits alpha-1-proteinase inhibitor inactivation by oxidants derived from human polymorphonuclear leukocytes. Exp. Pathol. 38:249.

NUKIWA, T., BRANTLY, M., GARVER, R., PAUL, L., COURTNEY, M., LECOCQ, J-P. & CRYSTAL, R.G. (1986a). Evaluation of "at risk" alpha 1-antitrypsin genotype SZ with synthetic oligonucleotide gene probes. J. Clin. lnvest. 77:528. NUK¡WA, T., SATOH, K., BRANTLY, M.L., OGUSHI, F., FELLS, G.A., COURTNEY, M. & CRYSTAL, R.G. (1986b). ldentification of a second mutation in the protein-coding sequence of the Z lype alpha 1-antitrypsin. J. Biol. Chem. 261:15989.

NUKIWA, T., BRANTLY, M,, OGUSHI, F., FELLS, G., SATOH, K., STIER, L., COURTNEY, M. & CRYSTAL, R.G. (1987a). Characterization of the M1(Ala 213) type alpha 1-antitrypsin, a newly recognized common "normal" alpha 1-antitrypsin haplotype. Biochemislry 26 : 5259.

NUKIWA, T., TAKAHASHI, H., BRANTLY, M., COURTNEY, M. & CRYSTAL, R.G. (1987b). Alpha 1 antitrypsin Null eranite falls, â non-expressing alpha 1 antitrypsin gene associated with a frameshift to stop mutation in a coding exon. J. Biol. Chem. 262 :1 1999.

1 95. NUKIWA, T., BRANTLY, M.L., OGUSHI, F., FELLS, G.A. & CRYSTAL, R.G. (1988). Characterization of the gene and protein of the common alpha- lantitrypsin normal M2 allele. Am. J. Hum. Genet. 43:322.

ODIÈVRE, M., MARTIN, J-P., HADCHOUEL, M., ALAGILLE, D. & THALER, M. (1976). Alpha-1-antitrypsin deficiency and liver disease in children : phenotypes, manifestations, and prognosis. Pediatr. 57 :226.

OGUSHI, F., FELLS, G.A., HUBBARD, R.C., STRAUS, S.D. & CRYSTAL, R.G. (1987). Z-lype alpha 1 antitrypsin is less competent then M1-type alpha 1- anitrypsin as an inhibitor of neutrophil elastase. J. Clin. lnvest. 80 : 1366.

OGUSHI, F., HUBBARD, R.C., FELLS, G.A., CASOLARO, A., CURIEL, D.T., BRANTLY, M.L. & CRYSTAL, R.G. (1988). Evaluation of the S-type of alpha-1-antitrypsin as an in vivo and in vitro inhibitor of neutrophil elastase. Am. Rev. Respir. Dis. 137 :364.

OGUSHI, F., HUBBARD, R.C., VOGELMEIER, C., FELLS, G.A. & CRYSTAL, R.G. (1991). Risk factors for emphysema. Cigarette smoking is associated with a reduction in the association rate constant of lung alpha- 1-antitrypsin for neutrophil elastase, J. Clin. lnvest. 87 : 1060.

OHLSSON, K. (1971). Neutral leukocyte proteases and elastase inhibited by plasmaalpha 1-antitrypsin. Scand. J. Clin. Lab. lnvest. 28:251.

OKAYAMA, H., CURIEL, D.T., BRANTLY, M.1., HOLMES, M.D. & CRYSTAL, R.G. (1989a). Rapid non-radioactive detection of mutations in the human genome by alle-specific amplification. J. Lab. Clin. Med. 114 : 105. oKAYAMA, H., HOLMES, M.D., BRANTLY, M.L. & CRYSTAL, R.G. (1989b). Characterization of the coding sequence of the normal M4 ø1-antitrypsin gene. Bochem. Biophys. Res. Commun. 162 : 1560. oKAYAMA, H., BRANTLY, M., HOLMES, M. & CRYSTAL, R.G. (1991). Characterization of the molecular basis of the cr1-antitrypsin F allele. Am. J. Hum. Genet. 48 : 1154.

OLSEN, G.N., HARRIS, J.O., CASTLE, J.R., WALDMAN, R.H. & KARMGARD, H.J. (1975). Alpha 1-antitrypsin content in the serum, alveolar macrophages, and alveolar lavage fluid of smoking and nonsmoking normal subjects. J. Clin. lnvest. 55:427.

OPIE, E.L. (1905). Enzymes and anti-enzymes of inflammatory exudates. J. Exp.Med.7:316.

OSSANNA, P.J., TEST, S.T., METHESON, N.R., REGIANI, S. & WEISS, S.J. (1986). Oxidative regulation of neutrophil elastase-alpha-1-proteinase inhibitor interactions. J. Clin. lnvest. 77 :1939.

OWEN, M.C. & CARRELL, R.W. (1976) Alpha 1-antitrypsin : molecular abnormality of S variant. Br, Med. J. 1 : 130.

1 96. OWEN, M.C., CARRELL, R.W. & BRENNAN, S.D. (1976). The abnormality of the S variant of human a1-antitrypsin. Biochim. Biophys. Acta. 453 :257. owEN, M.C., BRENNAN, S.O., LEWIS, J.H. & CARRELL, R.W. (1983). Mutation to antithrombin:cr1-antitrypsin - Pittsburgh (358 Met "" Arg), a fatal bleeding disorder. N. Engl. J. Med. 309 :694.

PADRINES, M., RABAUD, M. & BIETH, J.G. (1992). Oxidized alpha-1- proteinase inhibitor : a fast acting inhibitor of human pancreatic elastase. Biochim. Biophys. Acta. 118 :174. PALMER, P.E., DE LELLIS, R.A. & WOLFE, H.J. (1974). lmmunohistochemistry of liver in alpha-1-antitrypsin deficiency. Am. J. Clin. Pathol. 62 :350.

PALMER, P.E. & WOLFE, H.J. (1976) Alpha 1-antitrypsin deposition in primary hepatic carcinomas. Arch. Pathol. Lab. Med. 1O0 :232.

PARHAM, D.M., PATERSON, J.R., GUNN, A. & GUTHRIE, W. (1989). Cholangiocarcinoma in two siblings with emphysema and alpha-1- antitrypsin deficiency. Q.J. Med. 71 :359.

PARIENTE, E-A., DEGTT, C., MARTIN, J-P., FELDMANN, G., POTET, F. & BENHAMOU, J-P. (1981). Hepatocytic PAS-positive diastase-resistant inclusions in the absence of alpha 1-antitrypsin deficiency-high prevalence in alcoholic cirrhosis. Am. J. Clin. Pathol. 76 :299.

PARKER, B.A. & STARK, G.R. (1979). Regulation of simian virus 40 transcription : sensitive analysis of the RNA species present early in infection by virus or viral DNA. J. Virol. 31 : 360.

PEARSON, S.J., TETRI, P., GEORGE, D.L. & FRANCKE, U. (1983). Activation of human alpha 1-antitrypsin gene in rat hepatoma X human fetal liver cell hybrids depends on the presence of human chromosome 14. Somatic Cell. Genet. 5 :567.

PELHAM, H.R.B. & JACKSON, R.J. (1976). An efficient mRNA-dependent translation system from reticulocyte lysates. Eur. J. Biochem. 67 :247.

PERL¡NO, E., CORTESE, R. & CILIBERTO, G. (1987). The human ø1 - antitrypsin gene is transcribed from two different promoters in macrophages and hepatocytes. EMBO. J. 6 : 2767.

PERLMUTTER, D.H. (1991). The cellular basis for liver lnjury in cr1-antitrypsin deficiency. Hepatol. 13 :172.

PERLMUTTER, D.H., COLE, F.S., GOLDBERGER, G. & COLTEN, H.R. (1984). Distinct primary translation products from human liver mRNA give rise to secreted and cell associated forms of complement protein C2. J. Biol. Chem. 259 :10380. 197. PERLMUTTER, D.H., COLE, F.S., KILBRIDGE, P. ROSSING, T.H. & COLTEN, H.R. (1985a). Expression of the alpha 1-proteinase inhibitor gene in human monocytes and macrophages. Proc. Natl. Acad. Sci. USA 82 : 795.

PERLMUTTER, D.H., KAY, R.M., COLE, F.S., ROSSING, T.H., van THIEL, D. & COLTEN, H.R. (1985b). The cellular defect in alpha 1-proteinase inhibitor (a1-Pl) deficiency expressed in human monocytes and Xenopus oocytes injected with human liver mRNA. Proc. Natl. Acad. Sci. USA 82 :6918.

PERLMUTTER, D.H., TRAVIS, J. & PUNSAL, P.l. (1988). Elastase regulates the synthesis of its inhibitor, o1-proteinase inhibitor, and exaggerates the defect in homozygous P\ZZal Pl deficiency. J.Clin. lnvest.81 :1774.

PERLMUTTER, D.H., DANIELS, J.D., AUERBACH, H.S., DE SCHRYVER- KECSKEMETI, K., WINTER, H.S. & ALPERS, D.H. (1989a). The o1 - antitrypsin gene is expressed in a human intestinal epithelial cell line. J. Biol. Chem. 264: 9485.

PERLMUTTER, D.H., SCHLESINGER, M.J., PIERCE, J.4., PUNSAL, P.I. & SCHWARTZ, A.L. (1989b). Synthesis of stress proteins is increased in individuals with homozygous P\ZZ ø1-antitrypsin deficiency and liver disease. J. Clin. lnvest.84 : 1555.

PERLMUTTER, D.H., GLOVER, G.I., RIVETNA, M., SCHASTEEN, C.S. & FALLON, R.J. (1990a). ldentification of a serpin-enzyme complex receptor on human hepatoma cells and human monocytes. Proc. Natl. Acad. Sci. USA. 87 :3753.

PERLMUTTER, D.H., JOSLIN, G., NELSON, P., SCHASTEEN, C., ADAMS, S.P. & FALLON, R.J. (1990b). Endozytosis and degradation of ø1- antitrypsin-protease complexes is mediated by the serpin-enzyme complex (SEC) receptor. J. Biol. Chem. 265 : 16713.

PETERSON, K.8., KØLVRAA, S., BOLUND, L., PETERSEN, G.8., KOCH, J. & GREGERSEN, N. (1988). Detection of alpha-1-antitrypsin genotypes by analysis of amplified DNA sequences. Nucl. Acids. Res. 16 : 352.

PFEFFER, S.R. & ROTHMAN, J.E. (1987). Biosynthetic protein transport and sorting by the endoplasmic reticulum and Golgi. Ann. Rev. Biochem. 56 : 829,

PIERCE, J.A. & ERADIO, B. (1981) MPsa¡nt touis I â new antitrypsin phenotype Hum. Hered. 31 35. ! PIERCE, J.4., ERADIO, B. & DEW, T.A. (1975). Antitrypsin phenotypes in St Louis. JAMA. 231 :609.

PILACIK,8., RUJNER, J., SOCHA, J., MADALINSKI, R., CIANCIARA, J. & MIAN, N. (1990). Alpha-1-antitrypsin phenotypes and their role in the development of early and late cirrhosis of the liver. Wiad-Lek. 43 : 553.

1 98. PLAZONNET, M.P., CONSTANS, J., MISSON, G. & GENTOU, C. (1980). Familial alpha-1-antitrypsin deficiency with Pi-Z and a new Pi-Gcler variant. Biomedicine 33 : 85.

POLLER, W., MEISEN, C. & OLEK, K. (1990a). DNA polymorphisms of the alpha-1-antitrypsin gene region in patients with chronic obstructive pulmonary disease. Eur. J. Clin. lnvest. 20 : 1.

POLLER, W., FABER, J.P. & OLEK, 1(. (1990b). Highly variable clinical course in severe alpha-1-antitrypsin deficiency - use of polymerase chain reaction for the detection of rare deficiency alleles. Klin. Wochenschr. 68 : 857.

POLLER, W., FABER, J.P., WEIDINGER, S. & OLEK, K. (1991). DNA polymorphisms associated with a new alpha-1-antitrypsin P|QO variant (PiQOr¡e¿enbursh). Hum. Genet. 86 :522.

PORTER, C.4., MOWAT,4.P., COOK, P.J.L., HAYNES, D.W.G., SHILKIN, K.B. & WILLIAMS, R. (1972). a1-antitrypsin deficiency and neonatal hepatitis. Br. Med. J. 3 :435.

PoTTAGE, J.C., TRENHOLME, G.M., ARONSON, LK. & HARRIS, A.A. (1983). Panniculitis associated with histoplasmosis and alpha-1-antitrypsin deficiency. Am. J. Med. 75 :150.

POWERS, J.C. (1983). Synthetic elastase inhibitors : prospects for use in the treatment of emphysema. Am. Rev. Respir. Dis. 127 :554.

POWERS, J.C., GUPTON, F., HARLEY, 4.D., NISHINO, N. & WHITLEY, R.J. (1977). Specificity of porcine pancreatic elastase, human leukocyte elastase and cathepsin G. Biochim. Biophys. Acta. 485 : 156.

PROCHOWNIK, E., BOCK, S.C. & ORKIN, S.H. (1985). lntron structure of the human antithrombin lll gene differs from that of other members of the serine protease inhibitor superfamily. J. Biol. Chem. 260 :9608.

PURRELLO, M., ALHADEFF, 8., WHITTINGTON, E., BUCKTON, K.E., DANIEL, A., ARNAUD, P., ROCCHI, M., ARCHIDIACOMO, N., FILIPPI, G. & SINISCALCO, M. (1987) Comparison of cytologic and genetic distances between long arm subtelomeric markers of human autosome 14 suggests uneven distribution of crossing-over Cytogenet. Cell Genet. 44 :32.

RABIN, M., WATSON, M., KIDD, V., WOO, S.L.C., BREG, W.R. & RUDDLE, F.H. (1986). Regional location of alpha 1-antichymotrypsin and alpta 1- antitrypsin genes on human chromosome 14. Somatic Cell Mol. Genet. 12:209.

RAGG, H. & PREIBISCH, G. (1988). Structure and expression of the gene coding for the human serpin hLS2 J. Biol. Chem. 263 :12129.

199 RANGAN, V.S. & DAS, G.C. (1990). Purification and biochemical characterization of hepatocyte nuclear factor 2 involved in liver-specific transcription of the human cr1-antitrypsin gene. J. Biol. Chem. 265:8874.

REID, C.1., WIENER, G.J., COX, D.W., RICHTER, J.E. & GEISINGER, K.R. (1987) Diffuse hepatocellular dysplasia and carcinoma associated with the M¡¿¡sn variant of alpha 1-antitrypsin. Gastroenterololgy 93 : 181.

REINTOFT, l. & HÄGERSTRAND, l.E. (1979). Does lhe Z gene variant of alpha-1-antitrypsin predispose to hepatic carcinoma? Hum. Pathol. 10 : 41 9.

RENNARD, S.I., BERG, R., MARTIN, G.R., FOIDART, J.M. & ROBEY, P.G. (1980). Enzyme linked immunoassay (ELISA) for connective tissue components. Anal. Biochem. 104 : 205.

RHEAUMET, C., LATIMER, J.J., BAUMANN, H. & BERGER, F.G. (1988). Tissue- and species-specific regulation of murine o1-antitrypsin gene transcription. J. Biol. Chem . 263 : 151 18.

RIGHETTI, P.G., GIANAZZA, E. & GELFI, C. (1988). lmmobilized pH gradients. Trends Biochem. 13 : 335. RITCHIE, R.F. (1967). A simple, direct, and sensitive technique for measurement of specific protein in dilute solution. J. Lab. Clin. Med.70 : 512.

ROBERTS, B.E. & PATERSON, B.M. (1973). Efficient translation of tobacco mosaic virus RNA and rabbit 95 globin RNA in a cell-free system from commercial wheat germ. Proc. Natl. Acad. Sci. USA 70 : 2330. ROBERTS, E.A., COX, D.W., MEDLINE, A. & WANLESS, I.R. (1984) Occurrence of alpha 1-antitrypsin deficiency in 155 patients with alcoholic liver disease. Am. J. Clin. Pathol. 82:424. ROBERTS, N.A. & SURGENOR, A.E. (1986). Comparison of peptide aldehydes with o1-antitrypsin as elastase inhibitors for use in emphysema. Biochem. Biophys. Res. Commun. 139 :896.

ROBINET-LEVY, M. & REUINER, M. (1972). Techniques d'identification des groupes Pi. Rev. Fr. Transfus. lmmunohematol. 15 :61.

ROGERS, J., KALSHEKER, N., WALLIS, S., SPEER, 4., COUTELLE, C.H., WOODS, D. & HUMPHRIES, S.E. (1983). The lsolation of a clone for human cr1-antitrypsin and the detection of a1-antitrypsin in mRNA from liverand leukocytes. Biochem. Biophys. Res. Commun. 116 :375'

ROSENBERG, S., BARR, P.J., NAJARIAN, R.C. & HALLEWELL, R.A. (1984). Synthesis in yeast of a functional oxidation-resistant mutant of human cr1- antitrypsin. Nature 312 :77. 200. K., ROSENFELD, M.4., SIEGFRIED, -W., Y.OSHIMURA, K., YONEYAMA, FUKAYAMA, M., STIER, 1.E., PÄÄKKÖ, P.K., GILARDI, P., STRATFORD- PERRICAUDET, 1.D., PERRICAUDET, M., JALLAT, S., PAVIRANI, 4., LECOCQ, J-P. & CRYSTAL, R.G. (1991). Adenovirus-mediated transfer of a recombinant cl1-antitrypsin gene to the lung epithelium in vivo. Science 252 : 431.

RUBINSTEIN, H.M., JAFFER, A.M., KUDRNA, J.C., LERTRATANAKUL, Y., CHANDRASEKHAR,4.J., SLATER, D. & SCHMID, F.R. (1977). Alpha-1- ant¡tryps¡n deficiency with severe panniculitis. Ann. lntern. Med. 86:742.

RUSSELL, D.W., LEHRMAN, M.4., SUDHOF, T.C., YAMAMOTO, T., DAVID, G.C., HOBBS, H.H., BROWN, M.S. & GOLDSTEIN, J.L. (1987). The LDL receptor in familial hypercholesterolemia : use of human mutations to dissect a membrane protein. Cold Spring Harbor Quant. Biol. 51 : 811. SAHA, N. (1990). Alpha-1-protease inhibitor (Pi) subtypes in seven populations of east Asia. Ann. Hum. Biol. 17 :229.

SAIKI, R.K., SCHARF, S., FALOONA, F., MULLIS, K.8., HORN, G.T., ERLICH, n.n. a ARNHEIM, N.(1985). Enzymatic amplification of ß-globin genomic sequences and restriction analysis for diagnosis of sickle cell anemia. Science 230 : 1350.

SAIKI, R.K., BUGAWAN, T.L., HORN, G.T., MULLIS, K.B. & ERLICH, H.A. (1986). Analysis of enzymatically amplified ß-globin and HLA-DQcr DNA with allele-specific oligonucleotide probes. Nature 324 :163.

SAIKI, R.K., GELFAND, D.H., STOFFEL, S., SCHARF, S.J., HIGUCHI, R., HORN, G.T., MULLIS, K.B. & ERLICH, H.A. (1988). Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239 : 487.

SAITO, A. & SINOHARA, H. (1991). Cloning and sequencing of cDNAcoding for rabbit alpha-1-antiproteinase F : amino acid sequence comparison of alpha-1-antiproteinases of six months. J. Biochem (Tokyo). 109 : 150.

SALTINI, C., HANCE,4.J., FERRANS, V.J., BASSET, F., BITTERMAN, P.B. & CRYSTAL, R.G. (1984). Accurate quantification of cells recovered by bronchoalveolar lavage. Am. Rev. Resp. Dis. 130 : 650'

SANGER, F., NICKLEN, S. & COULSON, A.R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74 : 5463.

SATOH, K., NUKIWA, T., BRANTLY, M., GARVER, R.I., HOFKER, M., COURTNEY, M. & CRYSTAL, R.G. (1988). Emphysema associated with complete absence of alpha 1 antitrypsin in serum and the homozygous inhei'itance of a stop codon in an alpha 1-antitrypsin-coding-exon. Am. J. Hum. Genet. 42:77.

201. SATOH, K., TAKAHASHI, H. & CRYSTAL, R.G. (1989). lnfluence of the number of carbohydrate side chains on the in vivo behaviour of cr1 - antitrypsin. Am. Rev. Respir. Dis. 139:4202.

SCHAUER, 1., EMR, S., GROSS, L. & SCHEKMAN, R. (1985). lnvertase signal and mature sequence substitutions that delay intercompatmental transpott of active enzyme. J. Cell Biol. 100 : 1164.

SCHLEISSNER, L.A. & COHEN, A.H. (1975). Alpha-1-antitrypsin deficiency and hepatic carcinoma. Am. Rev. Respir. Dis. 111 : 863.

SCHRoEDER, W.T., MILLER, M.F., WOO, S.L.C. & SAUNDERS, G.F. (1985). Chromosomal localization of the human alpha 1-antitrypsin gene (Pi) to 14q31-32. Am. J. Hum. Genet. 37 :868.

SCHULTZE, H.E. & HEREMANS, J.F. (1966). Synthesis of plasma proteins. ln : Molecular biology of human proteins. Elsevier, Amsterdam ,321. SCHULTZE, H.E., GOLLNER, I., HEIDE, K., SCHONENBERGER, M. & SCHWICK, G. (1955). Zur kenntnis der alpha globuline des menschilchen normalserums. Z. Naturforsch. 10 :463.

SCHULTZE, H.E., HEIDE, K. & HAUPT, H. (1962). Alpha 1-antitrypsin aus humanserum. Klin. Wochenschr. 40 : 427.

SCHWARTZ,M., PETERSEN, K.8., GREGERSEN, N., HINKEL, K. & NEWTON, C.R. (1989). Prenatal diagnosis of alpha-1-antitrypsin deficiency using polymerase chain reaction (PCR). Comparison of conventional RFLP methods with PCR used in combination with allele specif ic oligonucleotides or RFLP analysis. Clin. Genet. 36 :419.

SCHWARZENBERG, S.J., SHARP, H.1., MANTHEI, R.D. & SEELIG, S. (1986). Hepatic o1-antitrypsin mRNA content in cirrhosis with normal and abnormal protease inhibitor phenotypes. Hepatol. 6 :1252.

SCOTT, C.F., CARRELL, R.W., GLASER, C.8., KUEPPERS, F., LEWIS, J.H. & COLMAN, R.W. (1986). Alpha-1-antitrypsin - Pittsburgh. A potent inhibitor of human plasma factor Xla, Kallikrein and Factor Xll. J. Clin. lnvest. 77 : 631.

SEFTON, L., KELSEY, G., KEARNEY, P., POVEY, S. & WOLFE, J. ( 1eeo). A physical map of the human P/ and AACT genes. Genomics 7 : 382.

SENIOR, R.M., HUEBNER, P.F. & PIERCE, J.A. (1971). Measurement of elastase activity by elastin agar and its use in the detection of antitrypsin deficiency. J. Lab. Clin. Med. 77 :510.

SENIOR, R.M., TEGNER, H., KUHN, C., OHLSSON, K., STARCHER, B.C. & PIERCE, J.A. (1977). The induction of pulmonary emphysema with leukocyte elastase. Am. Rev. Respir. Dis. 1 16 : 469.

202. sERALlNl, G.E., BERUBE, 0., GAGNE, R. & HAMMOND, G.L. (1990). The human corticosteroid binding globulin gene is located on chromosome 14q31 -q32.1 neartwo other serine protease inhibitor genes. Hum. Genet. 86 :73.

SESBOÜÉ, R., VERCAIGNE, D., CHARLIONET, R., LEFEBVRE, F. & MARTIN, J-P. (1984) Human alpha-1-antitrypsin genetic polymorphism: Pi N subtypes. Hum. Hered. 34 : 105.

SEYAMA, K., NUKIWA,T., TAKABE, K., TAKAHASHI, H., MIYAKE, K. & KIRA, s. (1991). Siiy"r" (serine 53 (Tcc) to Phenylalanine 53 (TTC)). J. Biol. Chem. 266 : 12627.

SHARP, H.L. (1971). Alpha-1-antitrypsin deficiency. Hosp. Pract. 5 :831. SHARP, H.L. (1982). Alpha-1-antitrypsin : an ignored protein in understanding liver disease. Semin. Liver Dis. 2 :314.

SHARP, H.1., BRIDGES, R.4., KRlVlT, W. & FREIER, E.F. (1969) Cirrhosis associated with alpha 1-antitrypsin deficiency : a previously unrecognized inherited disorder. J. Lab. Clin. Med. 73 :934.

SHEN, R.F.,CLlFT, S.M., DeMAYO, J.L., SIFERS, R.N., FINEGOLD, M.J' & WOO, S.L. (1989). Tissue specific regulation of human alpha-1- antitrypsin gene expression in transgenic mice. DNA' I : 101.

SHIGEOKA, J.W., HALL, W.J., HYDE, R.W., SCHWARTZ,R.H., MUDHOLKAR, G.S., SPEERS, D.M. & LlN, C.C. (1976). The prevalence of alpha-1- antitrypsin heterozygotes (Pi MZ) in patients with obstructive pulmonary disease. Am. Rev. Respir. Dis. 114 :1077.

SHIMADA, T. & NEINHUIS, A.W. (1985). Only the promoter region of the constitutively expressed normal and amplified human dihydrofolate reductase géne is DNase t hypersensitive and undermethylated. J. Biol. Chem. 260 : 2468.

SIFERS, R.N., CARLSON, J.4., CLIFT, S.M., DeMAYO, F.J., BULLOCK, D'W. & WOO, S.L.C. (1987). Tissue specific expression of the human alpha-1- antitrypsin gene in transgenic mice. Nucl. Acids Res. 15 : 1459.

SIFERS, R.N., BRASHEARS-MACATEE, S., KIDD, V.J., MUENSCH, H. & WOO, S.L.C. (1988). A frameshift mutation results in a truncated alpla 1 antitrypsin that is reiained within the rough endoplasmic reticulum. J. Biol. Chem. 263:7330.

SIFERS, R.N., ROGERS, 8.8., HAWKINS, H.K., FINEGOLD, M.J. & WOO, S.L.C. (1989a). Elevated synthesis of human o1-antitrypsin hinders the secretion of murine a1-antitrypsin from hepatocytes of transgenic mice. J. Biol. Chem.264: 15696.

203 SIFERS, R.N., FINEGOLD, M.J. & WOO, S.L.C. (1989b) Alpha 1-antitrypsin deficiency : accumulation or degradation of mutant variants with the hepatic endoplasmic reticulum. Am. J. Respir. Cell Mol. Biol. 1 :341.

SIFERS, R.N., SHEN, R-F. & WOO, S.L.C. (1989c). Genetic control of human alpha 1-antitrypsin. Mol. Biol. Med.6 :127.

SIFERS, R.N., HARDICK, C.P. & WOO, S.L.C. (1989d) Disruption of the 290- 342 salt bridge is not responsible for the secretory defect of the PiZ alpha 1 antitrypsin variant. J. Biol. Chem. 264 2997.

SILVERMAN, E.K., PIERCE, J.4., PROVINCE, M.A., RAO, D.C. & CAMPBELL, E.J. (1989). Variability of pulmonary function in alpha-1-antitrypsin deficiency : clinical correlates. Ann. lntern. Med. 111 : 982.

SILVERMAN, E.K., PROVINCE, M.4., RAO, D.C., PIERCE, J.A. & CAMPBELL, E.J. (1990a). A family study of the variability of pulmonary function in alpha-1-antitrypsin deficiency. Quantitative phenotypes. Am. Rev. Respir. Dis. 142(5): 1015.

SILVERMAN, E.K., PROVINCE, M.4., CAMPBELL, E.J., PIERCE, J.A. & RAO, D.C. (1990b) Variability of pulmonary function in alpha-1-antitrypsin deficiency : residual family resemblance beyond the effect of the Pi locus. Hum. Hered. 40(6) :340.

SINHA, S., WATOREK, W., KARR, S., GILES, J., BODE, W. & TRAVIS, J. (1987). Primary structure of human neutrophil elastase. Proc. Natl. Acad. Sci. U.S.A. 84 :228.

SMITH, K.C., PITTELKOW, M.R. & SU, W.P. (1987). Panniculitis associated with severe alpha-1-antitrypsin deficiency. Treatment and review of the literature. Arch. Dermatol. 123 : 1655. sMlTH, K.C., SU, W.P., PITTELKOW, M.R. & WINKELMANN, R.K. (1989). Clinical and pathologic correlations in 96 patients with panniculitis, including 15 patients with deficient levels of alpha-1-antitrypsin. J. Am. Acad. Dermatol. 21 :1192. SMITH, K.F., HARRISON, R.A. & PERKINS, S.J. (1990). Structural comparisons of the native and reactive-centre-cleaved forms of alpha-1- antitrypsin by neutron- and x-ray-scattering in solution. Biochem. J. 267 : 203.

SNIDER, G.L., LUCEY, E.C., CHRISTENSEN, T.G., STONE, P.G., CALORE, J.0., CATANESE, A. & FRANZBLAU, C. (1984). Emphysema and bronchial secretory cell metaplasia induced in hamsters by human neutrophil products. Am. Rev. Respir. Dis. 129 : 155.

SOUTHERN, E.M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98 :503.

204 SPROULE,8.J., COX, D.W., HSU, K., SALKIE, M.L. & HERBERT, F.A. (1983). Pulmonary function associated with the M¡¿¡en deficient variant of alpha- 1-antitrypsin. Am. Rev. Respir. Dis. 127 :237.

STEAD, W,W., FRY, D.L. & EBERT, R.V. (1952). The elastic properties of the lung in normal men and in patients with chronic pulmonary emphysema. J. Lab. Clin. Med. 40 :674.

STEIN, P.E., LESLIE, A.G.W., FINCH, J.T., TURNELL, W.G., McLAUGHLIN, P.J. & CARRELL, R.W. (1990). Crystal structure of ovalbumin as a model for the reactive centre of serpins. Nature 347 :99.

STOCKLEY, R.A. & AFFORD, S.C. (1984). Qualitative studies of lung lavage cr,1-proteinase inhibitor. Hoppe-Seyler's Z. Physiol. Chem. 365 : 503. sTocKLEY, R.4., MORRISON, H.M., SMlrH, S. & TETLEY, T. (1984). Low molecular mass bronchial proteinase inhibitor and a1-proteinase inhibitor in sputum and bronchoalveolar lavage. Hoppe-Seyler's Z. Physiol. Chem. 365 :587.

STOCKLEY, R.A., SHAW, J., AFFORD, S.C., MORRISON, H.M. & BURNETT, D. (1990). Effect of alpha-1-proteinase inhibitor on neutrophil chemotaxis. Am. J. Respir. Cell Mol. Biol. 2 : 163.

STONE, P.J., CALORE, J.D., McGOWAN, S.E., BERNADO, J., SNIDER, G.L. & FRANZBLAU, C. (1979). Functional alpha-1-protease inhibitor in the lower respiratory tract of cigarette smokers is not decreased. Science 221 : 1187.

STRAUS, S.D., FELLS, G.4., WEWERS, M.D., COURTNEY, M., TESSIER, L- H., TOLSTOSHEV, P., LECOCQ, J-P. & CRYSTAL, R.G. (1985). Evaluation of recombinant DNA-directed E. Coli produced alpha 1- antitrypsin as an anti-neutrophil elastase for potential use as replacement therapy of alpha 1-antitrypsin def iciency. Biochem. Biophys. Res. Commun. 130 : 1177.

SVEGER, T. (1976). Liver disease in alpha-1-antitrypsin deficiency detected by screening of 200,000 infants. N. Engl. J. Med. 294 :1316.

SVEGER, T. (1978). Alpha-1-antitrypsin deficiency in early childhood Pediatrics 62 :22.

SVEGER, T. (1984). Prospective study of children with cr1-antitrypsin deficiency : eight-year-old follow-up. J. Pediatr. 104 : 91.

SVEGER, T, (1988). The natural history of liver disease in alpha-1-antitrypsin deficient children. Acta. Paediatr. Scand. 77 :847.

SVEGER, T. & THELIN, T. (1981). Four-year-old children with cr1-antitrypsin deficiency. Acta Pædiatr. Scand. 70 :171.

205 TAKAHASHI, H. & CRYSTAL, R.G. (1990). Alpha-1-antitrypsin Null¡ss¡¿diprocida : an alpha-1-antitrypsin deficiency allele caused by deletion of all alpha-1- antitrypsin coding exons. Am. J. Hum. Genet. 47 :403.

TAKAHASHI, H., NUKIWA, T., YOSHIMURA, K., QUICK, C.D., STATES, D.J., HOLMES, M.D., WHANG-PENG, J., KNUTSEN, T. & CRYSTAL, R.G. (1988a). Structure of the human neutrophil elastase gene. J. Biol. Chem. 263 :14739.

TAKAHASHI, H., NUKIWA, T., SATOH, K., OGUSHI, F., BRANTLY, M., FELLS, G., STIERS, 1., COURTNEY, M. & CRYSTAL, R.G. (1988b). Characterization of the gene and protein of the cr1-antitrypsin "deficiency" allele Mprocida. J. Biol. Chem. 263:15528.

TAKEMURA, S., ROSSING, T.H. & PERLMUTTER, D.H. ( 1986). A lymphokine regulates expression of alpha 1-proteinase inhibitor i n human monocytes and macrophages. J. Clin. lnvest. 77 :1207.

TAKESHITA, K., FORGET, 8.G., SCARPA, A. & BENZ, Jr. E.J. (1984). lntranuclear defect in ß-globin mRNA accumulation due to a premature translation termination codon. Blood 64 : 13.

TALAMO, R.C., LANGEY, C.E. & HYSLOP, N.E. (1972). A comparison of functional and immunochemical measurements of serum alpha-1- antitrypsin. ln : Mittman, C. (ed). Pulmonary emphysema and proteolysis. Academic Press, New York ,173.

TALAMO, R.C., LANGLEY, C.E., REED, C. & MAKINO, S. (1973). Alpha 1 antitrypsin deficiency : a variant with no detectable alpha 1 antitrypsin. Science 181 : 70.

TAYLOR, J.C., COLIN, M., INAMIZU, T. & MITTMAN, C. (1980). Familial temperature sensitive alpha-1-protease inhibitor (Mlanar'e¡m). Clin. Chim. Acta. 104 : 301 .

THYMANN, M. (1986). Distribution of alpha-1-antitrypsin (Pi) phenotypes in Denmark determined by separator isoelectric focusing in agarose gel. Hum. Hered. 36 : 19.

TOBIN, M.J., COOK, P.J.L. & HUTCHINSON, C.S. (1 983). Alpha-1 -antitrypsin deficiency : the clinical and physiologica I features of pulmonary emphysema in subjects homozygous for Pi lype Z Br. J. Dis. Chest 77 14.

TOWNLEY, R.G., SOUTHARD, J.G., RADFORD, P., HOPP, R.J., BEWTRA, A.K. & FORD, L. (1990). Association of MS Pi phenotype with airway hyperresponsiveness. Chest 98 : 594.

206 TRAPNELL, 8.C., NAGAOKA, 1., CHYTIL, A. & CRYSTAL, R.G. (1989). Surface activation-induced upregulation of a1-antitrypsin gene expression in mononuclear phagocytes is associated with selective 5' non-coding exon usage. Clin. Res. 37 :4824.

TRAVIS, J. & SALVESEN, G.S. (1983). Human plasma proteinase inhibitors. Ann. Rev. Biochem.52 :655. TRAVIS, J., OWEN, M., GEORGE, P., CARRELL, R., ROSENBERG, S., HALLENELL, R.A. & BARR, P.J. (1985). lsolation and properties of recombinant DNA produced variants of human alpha-1-proteinase inhibitor. J. Biol. Chem. 260 : 4384.

TRAVIS, J., MATHESON, N.R., GOERGE, P.M. & CARRELL, R.W. (1986). Kinetic studies on the interaction of a1-proteinase inhibitor (Pittsburgh) with trypsin-like serine proteinases. Biol. Chem. Hoppe Seyler 376 : 853.

TRAVIS, J., SHIEH, B.H. & POTEMPA, J. (1988). The functional role of acute phase plasma proteinase inhibitors. Tokai, J. Exp. Clin. Med. 13 :313.

TRIPODI, M., ABBOTT, C., VIVIAN, N., CORTESE, R. & LOVELL-BADGE, R. (1991). Disruption of the LF-A1 and LF-81 binding sites in the human alpha-1-antitrypsin gene has a differential effect during development in transgenic mice. Embo. J. 10 :3177.

TURLEAU, C., DE GROUCHY, J., CHAVIN-COLIN, F., DORE, F., SEGER, J., DAUTZENBERG, M.D., ARTHUIS, M. & JEANSON, C. (1984). Two patients with interstitial del (14q), one with features of Holt-Oram syndrome. Exclusion mapping of Pi (alpha-1-antitrypsin). Ann. Genel. 27 :237.

UDALL, J.N., BLOCH, K.J. & WALKER, W.A. (1982) Transport of proteases across neonatal intestine and development of liver disease in infants with alpha 1-antitrypsin deficiency. Lancet i : 1441.

UDALL, J.N., DIXON, M., NEWMAN, 4.P., WRIGHT, J.A., JAMES, B. & BLOCH, K.J. (1985). Liver disease in a1-antitrypsin deficifncy. A retrospective analysis of the influence of early breast - vs. bottle Feeding. JAMA 253 : 2679.

UNDERHILL, D.A. & HAMMOND, G.L. ('1989). Organisation of the human corticosteroid binding globulin gene and analysis of its S'-flanking region. Mol. Endocinol. 3 :1448.

URLAUB, G., MITCHELL, P.J., CUIDAD, C.J. & CHASIN, L.A. (1989). Nonsense mutations in the dihydrofolate reductase gene affect RNA processing. Mol. Cell Biol. 9 :2868.

VAN FURTH, R., KRAMPS, J.A. & DIESSELHOF-DEN DULK, M M.(1e83). Synthesis of alpha-1-antitrypsin by human monocytes. Clin. Exp. lmmunol. 51 : 551. 207. VAN FURTH, R., KRAMPS, J.4., VAN DER FUTTEN, 4.8., KROM, R.A. & GIPS, C.H. (1986). Change in alpha-1-antitrypsin phenotype after orthotopic liver transplant. Clin. Exp. lmmunol. 66 : 669.

VAN OMMEN, G-J.B., ARNBERG,4., BAAS, F., BROCAS, H., STERK,4., TEGELAERS, W.H.H., VASSART, G. & de VIJLDER, J.J.M. (1983) The human thyroglobulin gene contains two 15-17 Kb introns near its 3' end. Nucl. Acids Res. 1 1 :2273.

VANDEVILLE, 0., MARTIN, J-P. & ROPARTZ, C. (1974). cr1-antitrypsin polymorphism of a Bantu population : description of a new allele PiL. Humangenetik 21 : 33.

VARPELA, E., KOISTINEN, J., KORHOLA, O. & KESKINEN, H. (1978). Deficiency of alpha-1-antitrypsin and bronchiectasis. Ann. Clin. Res. 10 : 79.

VAUGHAN, 1., LORIER, M.A. & CARRELL, R.W. (1982) o1-antitrypsin microheterogeneity : isolation and physiological significance of isoforms. Biochim et Biophys Acta 701 : 339.

VERBANAC, K.M. & HEATH, E.C. (1986) Biosynthesis, processing and secretion of M and Z varianl human alpha 1 antitrypsin. J. Biol. Chem. 261:9979.

VERLINSKY, Y., GINSBERG, N., LIFCHEZ, A., VALLE, J., MOISE, J. & STROM, C.M. (1990). Analysis of the first polar body : preconception genetic diagnosis. Hum. Reprod. 5 : 826. VOGELMEIER, C., BUHL, R., HOYT, R.F., WILSON, E., FELLS, G.4., HUBBARD, R.C., SCHNEBLI, H.P., THOMPSON, R.C. & CRYSTAL, R.G. (1 990). Aerosolization of recombinant secretory leukoprotease inhibitor as a strategy to augment the anti-neutroph¡l elastase protective screen of the pulmonary epithelial surface. Am. Rev. Respir. Dis. 141 :447. WAKEFIELD, D., BREIT, S.N., CLARK, P. & PENNY, R. (1982). lmmunogenetic factors in inflammatory eye disease. Arth. & Rheumat. 25 : 1 431.

WALLMARK, 4., ALM, R. & ERIKSSON, S. (1984). Monoclonal antibody specific for the mutant PiZ alpha-1-antitrypsin and its application in an ELISA procedure for identification ol PiZ gene carriers. Proc. Natl. Acad. Sci. USA. 81 :5690.

WEBER, W. & WEIDINGER, S. (1988). Pl Scologne : a new variant in the alpha-1-antitrypsin system. Hum. Genet. 80 : 102.

WEIDINGER, S. & CLEVE, H. (1984). High resolution of alpha-1-antitrypsin Pl M subtypes by isoelectric focusing with a modified immobilized pH gradient. Electrophoresis 5 :223.

208 WEIDINGER,' S. & CLEVE, H. (1986). Hybrid isoelectric focusing. for classification of alpha 1-antitrypsin variants. ln :Peeters, H. (ed). Protides of the biological fluids. Pergamon Press, Oxford. 34:863. WEIDINGER, S., CLEVE, H. & PATATSCHNICK, W. (1982). Alpha-1- antitrypsin : evidence for a fourth PiM allele. Distribution of the PiM subtypes in Southern Germany. Z.Rechtsmed 88 : 203.

WEIDINGER, S., JAHN, W., CUJNIK, F. & SCHWARZFISCHER, F. (1985). Alpha 1-antitrypsin : evidence for a fifth PiM subtype and a new deficiency allele Pi Zaugsburg. Hum. Genet. 71 :27.

WELCH, S.G., McGREGOR, l.A. & WILLIAMS, K. (1980). Alpha-1-antitrypsin (Pi) phenotypes in a village population from The Gambia, West Africa. Evidence of a new variant occurring at a polymorphic frequency. Hum. Genet. 53 :233.

WEWERS, M.D., GADEK, J.E., KEOGH, 8.4., FELLS, G.A. & CRYSTAL, R.G. (1986). Evaluation of Danazol therapy for patients with P\ZZ alpha-1- antitrypsin deficiency. Am. Rev. Respir. Dis. 134:476. WEWERS, M.D., CASALARO, M.A., SELLERS, S.E., SWAYZE, S.C., McPHAUL, K.M., WITTES, J.T. & CRYSTAL, R.G. (1987a). Replacement therapy for alpha 1-antitrypsin deficiency associated with emphysema. N. Eng. J. Med.316 :1055.

WEWERS, M.D., CASOLARO, M.A. & CRYSTAL, R.G. (1987b). Comparison of alpha-1-antitrypsin levels and antineutrophil elastase capacity of blood and lung in a patient with the alpha-1-antitrypsin phenotype Null-Null before and during alpha-1-antitrypsin augmentation therapy. Am. Rev. Respir. Dis. 135 : 539.

WEWERS, M.D., BRANTLY, M.L., CASALARO, M.A. & CRYSTAL, R.G. (1987c) Evaluation of tamoxifen aS a therapy to augment alpha 1-antitrypsin concentrations in Z homozygous alpha 1-antitrypsin deficient subjects. Am. Rev. Respir Dis 135 : 401.

WHITE, B.A. & BANCROFT, F.C. (1982). Cytoplasmic dot hybridization. Simple analysis of relative mRNA levels in multiple small cell or tissue samples. J. Biol. Chem. 257 :8569.

WHITEHOUSE, D.8., ABBOTT, C.M., LOVEGROVE, J.U., McINTOSH, 1., McMAHON, C.J., MIELI-VERGANI, G., MOWAT, A.P. & HOPKINSON, D.A' (1989a). Genetic studies on a new deficiency gene (Pl.Ztun) at the Pl locus. J. Med. Genet. 26:744. wHtrEHousE, 0.8., LOVEGROVE, J.U. & HOPKINSON, D.A. (1989b). Variation in alpha-1 -antitrypsin phenotypes associated with penicillamine therapy. Clin. Chim. Acta. 179 : 109.

209 WILSON, J.M., STOUT, J,T., PALELLA, T.D., DAVIDSON, 8.L., KELLEY, N.N. & CASKEY, C.T. (1986). A molecular survey of hypoxanthine-guanine phosphoribosyltransferase deficiency in man. J. Clin. lnvest. 77 :188.

WU, M.C. & ERIKSSON, S. (1988). Lung function, smoking and survival in severe alpha-1-antitrypsin deficiency P\ZZ. J. Clin. Epidemiol.4l :1157.

WU, Y. & FOREMAN, R.C. (1990). The effect of amino acid substitutions at position 342 on the secretion of human cl1-antitrypsin from Xenopus oocytes. FEBS. Lett. 268 : 21. WYSS, A., VIRCA, G.D. & SCHNEBLI, H.P. (1984)' Cigarette smoke components are not very effective in directly inactivating a1-proteinase inhibitor. Hoppe-Seyler's Z. Physiol. Chem. 365 : 511.

YAMAMOTO, Y., SAWA, R., OKAMOTO, N., MATSUI,4., YANAGISAWA, M. & IKEMOTO, S. (1986). Deletion 14q(q24.3 to q32.1) syndrome : significance of peculiar facial appearance in its diagnosis, and deletion mapping of Pi (alpha-1-antitrypsin). Hum. Genet. 74:190.

YING, Q.1., LIANG, C.C. & ZHANG, M.L. (1984). Pl.LBEl and PI.JHOU : two new alpha-1-antitrypsin alleles. Hum. Genet. 68 : 48.

YING, Q-L., ZHANG, M-L., LIANG, C-C., CHEN, L-C., CHEN, L-F., HUANG, Y- W., WANG, R.X., ZHANG, N-J., LI, H-J., LIU, S-S. & GAO, E-X. (1985b). Alpha-1-antitrypsin types in five Chinese national minorities. Hum. Genet. 71:225.

YING, Q-L., ZHANG, M-L.,.LIANG, C-C., LIU, X-P., HUANG, Y-W., WANG, R-X., ZHANG, N-J., CHEN, L.C., CHEN, L-F., YU, N-C. & MUO, X-P. (1985) Geographic variability of alpha 1 - antitrypsin alleles in China - a Study of six Chinese populations. Hum. Genet. 69 : 184.

YOSHIDA, 4., LIEBERMAN, J., GAIDUHS, L. & EWING, C. (1976). Molecular abnormality of human alpha 1-antitrypsin variant (Pi-ZZ) associated with plasma activity deficiency. Proc. Natl. Acad. Sci. USA 73 :1324.

YOSHIDA, A., CHILLAR, R. & TAYLOR, J.C. (1979). An cr1-antitrypsin variant, Pi B Rlrramura (LyS " AsP, Glu " Asp), with rapid anodal electrophoretic mobility. Am. J. Hum. Genet. 31 :555.

YU, S.D. & GAN, J.C. (1977). The role of sialic acid and galactose residues in determining the survival of human plasma o1-antitrypsin in the blood circulation. Arch. Biochem. Biophys. 179 : 477.

YUASA, I., SUENAGA, K., GOTOH, Y., ITO, K., YOKOYAMA, N. & OKADA, K. (1984) Pl (cr1-antitrypsin) polymorphism in the Japanese : confirmation of PlM4 and description of new Pl variants. Hum' Genet. 67 :2O9.

210. YUASA, l. & OKADA, K. (1985). A new cx,1-antitrypsin allele Pi Poki : lsoelectric focusing with immobilized pH gradients as a tool for identification of Pi variants. Hum. Genet. 70 :333.

211. APPENDICES

A. PUBLICATIONS ARISING FROM THIS TH ESIS

PU B LICATIONS

1 CURIEL,0.T., HOLMES, M.D., OKAYAMA, H., BRANTLY, M.L., VOGELMEIER, C., TRAVIS, W.D., STIER, 1., PERKS, W.H. & CRYSTAL, R.G. (198S) Molecular basis of the lung and liver disease associated w¡th the ø1-antitrypsin deficiency allele Mmafton. J. Biol. Chem. 264 : 13938.

2 HOLMES, M., CURIEL, D., BRANTLY, M. & CRYSTAL, R.G. (1989) Characterization of the intracellular mechanism causing the ct1-antitrypsin Nullsranite 1"¡¡s deficiency state. Am. Rev. Respir. Dis. 1 40 :1662.

3 HOLMES, M.D., BRANTLY, M.1., CURIEL, D.T., WEIDINGER, S. & CRYSTAL, R.G. (1990) Characterization of the normal ø1-antitrypsin allele Vmunich : A variant associated with a unique protein isoelectric focusing pattern. Am. J. Hum. Genet. 46 :810.

4. HOLMES, M.D., BRANTLY, M.L. & CRYSTAL, R.G. (1990) Molecular analysis of the heterogeneity among the P-family of a1-antitryps¡n variants. Am. Rev. Respir. Dis. 142:1185.

5. HOLMES, M.D., BRANTLY, M.1., FELLS, G.A. & CRYSTAL, R.G. (1990) Analysis of the molecular basis of the cr1-antitrypsin deficiency variant Wbethesda. Biochem. Biophys. Res. Commun. 170 : 1013.

ABSTRACTS

1 CURIEL, D., HOLMES, M., OKAYAMA, H., TRAVIS, W., STIER, L., BRANTLY, M. & CRYSTAL, R.G. (1989) Molecular basis of liver disease assoc¡ated with cr1-antitrypsin deficiency. Clin. Res. 37 (2) : 3664.

212. 2 HOLMES, M., CURIEL, D., BRANTLY, M. & CRYSTAL, R.G. (1989) Characterization of the intracellular mechanism causing the o1-antitrypsin Nullsranite rals deficiency state. Am. Rev. Respir. Dis. 139 : 4202.

3. HOLMES, M., BRANTLY, M., WURTS, L. & CRYSTAL, R.G. (1989) Characterization of the sequence differences among the P-family of a1- antitrypsin alleles. Am. Rev. Respir. Dis. 139 : 4369.

4 HOLMES, M., BRANTLY, M., FELLS, G. & CRYSTAL, R.G. (1990) Analysis of the molecular basis of the ø1-antitrypsin deficiency variant Wberhesda. Aust. N.Z. J. Med.20 :523.

213 B. CURRICULUM VITAE

PERSONAL DETAILS

NAME Mark Derek HOLMES

DATE OF BIRTH I June, 1957.

PLACE OF BIRTH Adelaide, South Australia.

CITIZENSHIP Australian.

MARITAL STATUS Married

CHILDREN Two

HOME ADDRESS

WORK ADDRESS Department of Thoracic Medicine, Royal Adelaide Hospital, North Terrace, ADELAIDE. South Australia. 5000.

BASIC MEDICAL QUALIFICATIONS

1 980 Bachelor of Medicine, Bachelor of Surgery (M.B.B.S.), UniversitY of Adelaide.

214 POSTGRADUATE MEDICAL QUALIFICATIONS

1 988 Fellowship, Royal Australian College of Physicians (F.R.A.C.P.).

POSTGRADUATE MEDICAL EXPERIENCE

1 981 lntern, The Queen Elizabeth Hospital, Woodvi le, South Australia.

1 982 Resident Medical Officer, The Queen Elizabeth Hospital, Woodville, South Australia.

1 983 Resident Medical Officer / General Medical Registrar, The Queen Elizabeth Hospital, Woodville, South Australia.

1 984 General Medical Registrar, The Queen Elizabeth Hospital, Woodville, South Australia.

1 985 Registrar in Thoracic Medicine, The Queen Elizabeth Hospital, Woodville, South Australia.

1 986 Registrar in Thoracic Medicine, Royal Adelaide Hospital, Adelaide. South Australia.

1987 - 1989 Fogarly lnternational Visiting Associate, Pulmonary Branch, National Heart, Lung and Blood lnstitute, National lnstitutes of Health, Bethesda, Maryland. USA.

1990 - (January - May) Locum Staff Physician, Department of Thoracic Medicine, The Queen Elizabeth Hospital, Woodville. South Australia.

215 1990 - Consultant Physician, Department of Thoracic Medicine, Royal Adelaide Hospital, Adelaide. South Australia.

SOCIETY / ASSOCIATION MEMBERSHIP

Royal Australian College of Physicians - Fellow.

Thoracic Society of Australia and New Zealand - Member.

South Australian Salaried Medical Officers Association - Member

Australian Salaried Medical Officers Association - Member.

American Thoracic Society - Member.

216 PU B LI CATI O NS

1 TAKAHASHI, H., NUKIWA, T., YOSHIMURA, K., QUICK, C.D., STATES, D.J., HOLMES, M.0., WHANG-PENG, J., KNUTSEN, T. & CRYSTAL, R.G. (1988) Structure of the human neutrophil elastase gene. J. Biol. Chem. 263 :14739.

2 CRYSTAL, R.G., BRANTLY, M.1., HUBBARD, R.C., CURIEL, D.T., STATES, D.J. & HOLMES, M.D. (1988) The a1-antitrypsin Gene and its Mutations : Clinical consequences and strategies for therapy. Chest 95 : 196.

3 OKAYAMA, H., CURIEL, T., BRANTLY, M.L., HOLMES, M.D. & CRYSTAL, R.G. (1989) Rapid, non-radioactive detection of mutations in the human genome by allele specific amplification. J. Lab. Clin. Med. 144: 105.

4 CURIEL, D.T., HOLMES, M.0., OKAYAMA, H., BRANTLY, M.L., VOGELMEIER, C., TRAVIS, W.D., STIER, L., PERKS, W.H. & CRYSTAL, R.G. (1 989) Molecular basis of the lung and liver disease associated with the cx,1-antitrypsin deficiency allele Mmatton. J. Biol. Chem.,264:13938.

5 HOLMES, M., CURIEL, D., BRANTLY, M. & CRYSTAL, R.G. (1989) Characterization of the intracellular mechanism causing the a1-antitrypsin Nullgran¡e 1¿¡¡s deficiency State. Am. Rev. Respir. Dis. 1 40 :1662.

6. ABE, T., TAKAHASHI, H., HOLMES, M.D., CURIEL, D.T. & CRYSTAL, R.G. (1989) Ribonuclease a cleavage combined with the polymerase chain reaction for detection of lhe Z mutation of the ø1-antitrypsin gene. Am. J. Respir. Cell Mol. Biol. 1 :329.

7 OKAYAMA, H., HOLMES, M.D., BRANTLY, M.L. & CRYSTAL, R.G. (1989) Characterization of the coding sequence of the normal M4 cr1-antitrypsin gene. Biochem. Biophys. Res. Commun. 162 : 1560.

8 HOLMES, M.D., BRANTLY, M.L., CURIEL, D.T., WEIDINGER, S. & CRYSTAL, R.G. (1990) Characterization of the Normal cr1-antitrypsin Allele Vmunich : A variant associated with a unique protein isoelectric focusing pattern. Am. J. Hum. Genet., 46 : 810.

9 HOLMES, M.D., BRANTLY, M.L. & CRYSTAL, R.G. (1990) Molecular analysis of the heterogeneity among the P-family of cr1-antitrypsin variants. Am. Rev. Respir. Dis. 142,1185-1192.

217 10. HOLMES, M.D., BRANTLY, M.1., FELLS, G.A. & CRYSTAL, R.G. (1990) cr1-antitrypsin Wsslhesda : Molecular basis of an unusual cr1-antitrypsin deficiency variant. Biochem. Biophys. Res. Commun. 170 : 1013.

11 OKAYAMA, H., BRANTLY, M., HOLMES, M. 7 CRYSTAL, R.G. (1991) Characterization of the molecular basis of the cr1-antitrypsin F allele. Am. J. Hum. Genet. 48 :1154.

12 CHIA, M.M., HOLMES, M.D. & McLENNAN, G. (1991) The molecular biology of lung cancer. Med. J. Aust. 154 : 501.

ABSTRACTS

1 HOLMES, M.D., DREW, M.J.R., ANTIC, 4., RODER, D., ESTERMAN, 4., GILL, P.G. & McLENNAN, G. (1987) Comparison of two chemotherapy regimes for small cell carcinoma. Medical & Paediatric Oncology 15 : 1344.

2 TAKAHASHI, H., NUKIWA, T., YOSHIMURA, K., QUICK, C.D., STATES, D.J., HOLMES, M.D., BASSET, P., WHANG-PENG, J., KNUTSEN, T. & CRYSTAL, R.G. (1988) Structure and expression of the human neutrophil elastase gene. Clin. Res. 36(3) :4204.

3 CURIEL, D., HOLMES, M., OKAYAMA, H., TRAVIS, W., STIER, L., BRANTLY, M. & CRYSTAL, R.G. (1989) Molecular basis of liver disease associated with a1-antitrypsin deficiency. Clin. Res. 37(2):3664.

4" OKAYAMA, H., CURIEL, D., BRANTLY, M., HOLMES, M. & CRYSTAL, R.G. (1989) Rapid, non-radioactive analysis of mutations in the cr1-antitrypsin gene by allele specific amplification. Am. Rev. Respir. Dis. 139 :4201.

5 HOLMES, M., CURIEL, D., BRANTLY, M. & CRYSTAL, R.G. (1989) Characterization of the intracellular mechanism causing the o1-antitrypsin Nullsran¡e ralls deficiency state. Am. Rev. Respir. Dis. 139 : 4202.

6 LAUBACH, V., CURIEL, D., BRANTLY, M., HOLMES, M. & CRYSTAL, R.G. (1989) Characterization of the gene sequence of the common ø1 - antitrypsin normal M3 allele. Am. Rev. Respir. Dis. 139 :4368.

7 ABE, T., TAKAHASHI, H., HOLMES, M., CURIEL, D.& CRYSTAL, R.G. (1989) Ribonuclease cleavage combined with the polymerase chain reaction for detection of lhe Z mutation of the a1-antitrypsin gene. Am. Rev. Respir. Dis. 139 : 4202.

218. 8. HOLMES, M., BRANTLY, M., WURTS, L. & CRYSTAL, R.G. (1989) Characterization of the sequence differences among the P-Family of cr1- antitrypsin alleles. Am. Rev. Respir. Dis. 139 : 4369.

9 HOLMES, M., BRANTLY, M., FELLS, G. & CRYSTAL, R.G. (1990) Analysis of the molecular basis of the cr1-antitrypsin deficiency variant Wberhesda. Aust. N.Z. J. Med. 20 :523.

10. DILLON, T.J., WALSH, R.1., SCHICCHITANO, R., HOLMES, M., BETTS, H. & McLENNAN, G. (1992). Elastin derived peptides modulate superoxide production by neutrophils in response to FMLP. Proceedings of the Annual Scientific Meeting of the Thoracic Society of Australia and New Zealand, Canberra.

219.