Introduction

More than thirty years have passed since the association between chronic lung disease and an obscure serum component was first noted. Alpha one-Antitrypsin (A1AT) deficiency is now one of the best understood hereditary anomalies involving the lungs and a key to our current understanding of the patogenesis of emphysema. Though only 1-2% of the more than two million Americans with emphysema are thought to have A1AT associated disease, among patients younger than 50 years old nearly one-half have this hereditary version and it is therefore important for the clinician to ask:

• What have simple clinico-pathological observations taught us?
• What remains to be understood?
• Do we know enough to be therapeutically effective?

Background: Coming in Out of the Cold

In 1963, Carl-Bertil Laurell, the Head of the Clinical Chemistry Department in Malmo, Sweden, performed electrophoretic analyses of serum proteins from patients at a chest hospital and noted that some patients had a missing alpha one band. Enlisting a young medical resident, Sten Eriksson, he reviewed five new patients in whom the deficiency was noted.(1)

Three of the five patients suffered from obstructive lung disease, suggesting the hypothesis that this deficiency was somehow linked to the pathogenesis of chronic airway obstruction. It was not unexpected that this observation was first described in Sweden, a country long interested in the characterization of serum proteins. The names of Svedberg, Tiselius and Waldenstrom had already been intimately associated with the protein composition of plasma.

The subsequent identification of severe airway obstruction in three members of a single family afflicted with A1AT deficiency(2) led to the conclusion that this deficiency was hereditary. Further work revealed that a single gene was associated with the disorder, at least in its most common form. By 1965, based on 33 deficient homozygous individuals and their families, Eriksson in his PhD thesis postulated that this hereditary disorder was associated with premature, or programmed, emphysema.

Eriksson realized that not only was the protein band which carried the bulk of the serum's anti-trypsin activity markedly reduced, but what protein was found in the band was markedly abnormal.

The healthy state was characterized by a double dose of the normal M allele (MM), while the most common pathologic variant was found in both a heterozygous MZ state and a homozygous ZZ state. Only the homozygous state was clearly associated with the clinical disease and very low levels of A1AT.

Clinical, Biochemical and Hereditary Features of A1AT Lung Diseas

The clinical features of A1AT deficiency are now well defined. A majority of the homozygous affected individuals are predisposed to premature (before the fifth decade) emphysema. The lung destruction involves the lower lung fields and a bronchitic component is usually absent. This tends to distinguish A1AT from the typical patient with chronic obstruction in whom bronchitic and emphysematous findings co-exist. Hyperinflation and bullae are seen on chest X-ray and diffusion capacity is characteristically reduced.

Measurements of the serum's trypsin inhibitory capacity (STIC) in the relatives of these patients is found to be either normal, intermediate (about 60% of normal) or markedly reduced (approximately 10% of normal) in the affected individuals. Current technology allows direct measurement of alpha-1 levels by an immunologic assay.

The original suggestion that the genetic pattern of inheritance was recessive was not quite correct. In fact, the alleles M and Z are autosomal co-dominant, each coding for their own protein.(4) The nomenclature adopted for describing the genetics of this disease is the Pi (Proteinase inhibitor) system in which the designation Pi is followed by the allelic representation of the protein (e.g., M, Z, S, etc.). The normal phenotype is expressed by the M allele with the normal homozygote represented by PiMM.

The principal cells that express A1AT are liver hepatocytes and mononuclear phagocytes. The A1AT genes are upregulated by inflammatory processes (such as infection). In blood, the half life of normal A1AT is four to five days.

At least 75 variants of the M allele have been described, but only a minority are associated with A1AT deficiency.(5)

* Levels in mg/dl;
+ Grouping by elastase function
** Frequency data from U.S.; rare= <.001
++ Assumed homozygous state

The Z variant of A1AT differs from the M protein by a single amino acid substitution. In addition to being found in lower concentrations, the Z variant has a reduced ability to inhibit neutrophil elastase (NE), the major proteolytic enzyme found in the lower airway. The Z variant of A1AT has difficulty exiting hepatocytes and is present in abnormal globules within these cells. This property explains the low serum level and may account for the severe liver dysfunction in some patients.

The deficiency is found in one out of every 3,000 to 4,000 Euro-Americans but is very rare in African-Americans and Asian-Americans.

Pathogenesis of Emphysema

Emphysema in patients with A1AT is presumed to be due to unrestrained proteolysis that occurs in the absence of sufficient or functional inhibitor. This hypothesis is suggested by the fact that installation of excess proteolytic enzyme in the airways of animals (e.g., guinea pigs) results in pathology indistinguishable from garden variety emphysema. Inflammatory situations like infections or cigarette smoking are assumed to accelerate the process.

The neutrophil associated enzyme neutrophil elastase (NE) is capable of cleaving a wide variety of components of the lung's elastic matrix, thus reducing lung recoil. In the alveolar space A1AT is the major protection against NE. Based on these findings the neutrophil elastase hypothesis of A1AT associated emphysema is formulated as an unequal balance between NE proteolysis and anti-NE defense.

How does this mechanism relate to the bulk of patients with emphysema who do not have a primary A1AT deficiency? One possible association that has been extensively investigated is that cigarette smokers have reduced A1AT and NE inhibitory function.(6)

Cigarette smoking is strongly associated with clinical disease.(7),(8) In a group of 69 individuals with A1AT deficiency only the smokers developed early onset airway obstruction. Moreover, there was a direct relationship between the amount smoked by these individuals and the degree of airway obstruction. By contrast children with A1AT deficiency followed to age 16 have normal lung function.

Smoking accelerates mortality in A1AT. Fifty percent survival is expected in A1AT deficient persons with FEV1 less than 15% of predicted. Such individuals are prime candidates for lung transplantation.(8)

Therapeutic Options

The treatment of established emphysema in patients with A1AT deficiency involves the same options as standard obstructive lung disease:

1. oxygen therapy
2. treatment of cigarette addiction
3. prophylaxis and treatment of respiratory infection
4. lung reduction surgery
5. lung transplantation

Lung transplantation for this disease requires special consideration since the A1AT deficiency persists in the transplanted individual.(9) Prolastin® supplementation is recommended for these patients.

A number of novel therapeutic approaches based on the neutrophil elastase hypothesis have been advanced. These include:

1. Increase production and/or secretion of A1AT by:
   • Liver transplantation
   • Raising liver production and/or secretion (danazol, tamoxifen)
2. Administration of substances with anti-NE properties:
   • IV administration of Prolastin®
   • Aerosol administration of Prolastin®
3. Gene therapy
   • Plasmids
   • Adenovirus and retrovirus vectors

Liver transplantation is a direct approach to restoring normal A1AT levels and function. Both adults and children with A1AT deficiency and associated end stage liver disease have been transplanted with normalization of serum A1AT levels. At present, for the vast majority of A1AT deficient individuals with normal liver function this therapy cannot be considered appropriate because liver transplantation is a major procedure with high mortality.

Attempts to increase hepatocyte production of A1AT by drugs such as danazol (Danocrine®) and tamoxifen (Nolvadex®) have not proven successful in the most common PiZZ variant.

Currently, a prophylactic approach involves the administration of A1AT obtained by purification of pooled human donor plasma. This can be administered once weekly by intravenous infusion in a dose of 60 mg/kg. This strategy is effective in raising lung anti-NE levels but it is as yet unclear whether such administration enhances clinical outcome (morbidity or mortality).

Aerosol administration of Prolastin®, the plasma derived factor, is feasible and potentially as effective. Use of bioengineered alpha 1-proteinase inhibitors is currently under investigation(11) but is not yet approved. A major advantage of such a therapeutic agent would be freedom from viral hepatitis contamination, a potential hazard with Prolastin®.

Another agent currently being evaluated for its anti-NE activity is secretory leukoprotease inhibitor (SLPI) which can be produced in its original state using E. coli to synthesize recombinant SPLI.(12) The A1AT protein which is produced by genetic engineering differs significantly from the native variety.

Gene Therapy

A1AT deficiency is potentially amenable to gene therapy and has been the focus of intense investigation. A1AT gene transfer to the deficient host requires the appropriate vector. Three potential delivery mechanisms have been considered:(13),(14),(15)

• Plasmids (circular double-stranded DNA) are relatively inefficient and must be pre-packed in liposomes or bonding molecules which can target specific cells (e.g., ligands of Immunoglobulin A or apoproteins of surfactant).
• Retroviruses can produce effective gene transfer. These are particularly well suited for ex-vivo transfection of cells which are removed from the body and then replaced. This might be suitable for hepatic cells which can divide. It is unsuitable for airway epithelium via aerosol because these are terminally differentiated cells.
• Adenoviruses are double stranded DNA viruses naturally attracted to respiratory tract cells. Genetic engineering allows the A1AT genome to be inserted along with an appropriate promoter and the virus' replicating machinery to be eliminated. This is the most promising of the three vectors but safety and efficacy problems have surfaced.

Summary

With an estimated 20,000 affected individuals in the U.S., and many more worldwide, the search for effective prophylaxis and treatment will continue. Lessons learned in the study of this disease have already shed light on the pathogenesis of non-hereditary emphysema and other obstructive lung diseases as well as therapeutic options for other hereditary pulmonary disease. For example, gene therapy is being investigated for cystic fibrosis and genetically engineered IgE is being tested in asthma.