**1. Introduction**

178 Chronic Obstructive Pulmonary Disease – Current Concepts and Practice

Yohannes, A. (2005) Depression and COPD in older people: a review and discussion. *British Journal of Community Nursing,* Vol. 10, No. 1, pp.42-6, ISSN 1462-4753. Yohannes, A., Baldwin, R. And Connolly, M. (2000) Depression and anxiety in elderly

Young, P., Dewse, M. & J., F. K. (1999) Respiratory rehabilitation in chronic obstructive

Zigmond, A., & Snaith, R. (1983) The hospital anxiety and depression scale. *Acta Psychiatrica* 

*Geriatric Psychiatry,* Vol. 15, No. 12, pp.1090-96, ISSN 1462-4753.

*Scandinavica,* Vol. 67, No. 6, pp.361-370, ISSN 0001-690X.

13, No. 4, pp.855-9, ISSN 0903-1936.

outpatients with Chronic Obstructive Pulmonary Disease: Prevalence, and the validation of the B.A.S.D.E.C. screening questionnaire. *International Journal of* 

pulmonary disease; predictors of non-adherence. *European Respiratory Journal,* Vol.

Alpha-1 antitrypsin deficiency (AATD) is a hereditary disorder characterised by low circulating levels of the key antiprotease alpha-1 antitrypsin (AAT) and is associated with the development of chronic obstructive pulmonary disease (COPD), often by the 3rd or 4th decade, and liver disease. The two most common SERPINA1 mutations associated with AATD are the Z and S mutations, and the vast majority of AATD individuals diagnosed with COPD are ZZ homozygotes. AATD is an under-diagnosed condition with the majority of cases misdiagnosed as COPD. The World Health Organisation (WHO) and the American Thoracic Society/European Respiratory Society (ATS/ERS) advocate a targeted screening approach for the detection of AATD in patients with COPD, non-responsive asthma, cryptogenic liver disease and first degree relatives of known AATD patients (Alpha 1 antitrypsin deficiency: memorandum from a WHO meeting 1997; American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. 2003). It is our contention that a diagnosis of AATD gives the clinician a vital and unique opportunity for early medical intervention and the possible prevention of COPD in both the affected individual and first-degree relatives. Unfortunately, despite huge strides in awareness and understanding of this condition, this opportunity is too often missed.

#### **2. Alpha-1 antitrypsin deficiency (AATD)**

#### **2.1 Alpha-1 antitrypsin (AAT)**

AAT is a secretory glycoprotein produced by the liver and is the most abundant serum antiprotease in circulation (Kueppers 1971). While the majority of AAT in the body is hepatocyte-derived, it is actively transcribed and secreted by other cell types including monocytes (Carroll et al. 2010), macrophages (Mornex et al. 1986), neutrophils (Bergin et al. 2010), intestinal epithelial cells (Perlmutter et al. 1989), and various epithelial cells in the lung (Hu and Perlmutter 2002; Venembre et al. 1994; Cichy, Potempa, and Travis 1997), albeit in smaller quantities. In keeping with its role as an acute phase reactant, the

Alpha-1 Antitrypsin Deficiency – A Genetic Risk Factor for COPD 181

Other prominent serpins include alpha-1 antichymotrypsin, alpha-2 antiplasmin, antithrombin, plasminogen activator inhibitor I, and C1 inhibitor (Law et al. 2006). The serpins act like molecular snares or mousetraps, existing in a metastable state, until they entice the unsuspecting cognate protease into their trap, and their energy is released by cleavage of their reactive centre loop (Carrell and Lomas 2002). AAT irreversibly inhibits serine proteases by presenting its inhibitory amino acid sequence as a mobile, exposed reactive centre loop with a methionine residue at its centre (Elliott et al. 1996). This reactive centre loop is a pseudosubstrate for target proteases, such as NE, and upon docking of the protease AAT undergoes a conformational change that results in the irreversible inhibition and destruction of the protease (Huntington, Read, and Carrell 2000). Although AAT can regulate the activity of many other serine proteases including trypsin, chymotrypsin, plasmin, thrombin, plasminogen, cathepsin G and proteinase 3 (Silverman et al. 2001; Travis and Salvesen 1983), it exhibits the highest association constant for NE which is 25 times greater than for any other protease (Beatty, Bieth, and Travis 1980). In addition, NE is the most relevant protease in the lung setting as it is released in large amounts by activated

AATD is a hereditary monogenic disorder caused by mutations within the SERPINA1 gene, previously known as the "protease inhibitor" or "PI" gene. The condition was first reported in the early 1960s by Carl-Bertil Laurell and Sten Eriksson at the University of Lund in Sweden when they noticed the absence of the alpha-1 globulin band on routine serum protein electrophoresis of five patients (Laurell and Eriksson 1963). It was already known that over 90% of the alpha-1 band comprised a single protein that was capable of inhibiting the proteolytic enzyme trypsin (Jacobsson 1955; Schultze, Heide, and Haupt 1962). Following a review of the five patients, three were found to have developed "a degenerative pulmonary disease" at a young age and the condition we now know as AATD was born (Eriksson 1964). Interestingly, the oldest reported case is thought to have been a young girl who had remained undiscovered in the Alaskan permafrost for over 800 years (Zimmerman, Jensen, and Sheehan 2000). More recently there has been speculation that the composer Frédéric Chopin suffered from AATD with his physician failing to confirm pulmonary tuberculosis instead discovering "a disease not previously encountered" on autopsy

AATD is associated with a substantially increased risk for the development of COPD, often by the third or fourth decade, and is also associated with risks for development of liver disease (Sharp et al. 1969; Lieberman, Mittman, and Gordon 1972; Sveger 1976), cutaneous panniculitis (Edmonds, Hodge, and Rietschel 1991), bronchiectasis (King et al. 1996), vasculitis (Lewis et al. 1985), and Wegener's granulomatosis (Barnett, Sekosan, and Khurshid 1999). AATD is characterised by misfolding of the AAT protein and belongs to a class of genetic diseases underpinned by aberrant protein folding collectively termed conformational disorders (Gooptu and Lomas 2009; Greene et al. 2008). In addition, AATD is a paradigm of the subclass of conformational diseases known as the serpinopathies (Carrell and Lomas 1997). Serpins have a highly conserved structure of three β–sheets, nine α–helices, and an exposed reactive centre loop (Gettins 2002). Serpinopathies are

neutrophils (Gadek 1992).

**2.2 Alpha-1 antitrypsin deficiency (AATD)** 

(Kuzemko 1994; Kubba and Young 1998).

**2.2.1 History and clinical features** 

hepatocyte expresses approximately 200 times more AAT mRNA than other cells (Rogers et al. 1983) and serum levels can rapidly increase by between two- and five-fold during infection, trauma, surgery and burns (Kossmann et al. 1995; Voulgari et al. 1982; Sandford et al. 1999; Jeschke, Barrow, and Herndon 2004). The AAT released by inflammatory cells is more relevant to the local inflammatory milieu, serving to limit local degradation of the extracellular matrix by proteases released from migrating immune cells (Knoell et al. 1998). In a similar fashion, the range of cells producing AAT locally in the lung underlines its key role in protecting the fragile lung parenchyma against proteolytic damage. AAT was originally named because of its ability to inhibit pancreatic trypsin (Schultze, Heide, and Haupt 1962) and while its principal function as an antiprotease has been well established, AAT has significant anti-inflammatory properties affecting numerous cell types, and has been implicated in areas as diverse as apoptosis (Petrache, Fijalkowska, Medler, et al. 2006), chemotaxis (Bergin et al. 2010), tissue repair (Dabbagh et al. 2001), innate immunity (Bergin et al. 2010), and cell signalling (Koulmanda et al. 2008).

The gene encoding AAT is the SERPINA1 gene, found on the long arm of chromosome 14 at q32.1 and is composed of four coding exons (II, III, IV, V), three untranslated exons (Ia, Ib, and Ic) in the 5' region and six introns. The hepatocyte transcription start site resides within exon Ic, while the mononuclear phagocyte transcription start sites reside within exons Ia and Ib. The SERPINA1 gene encodes a 418 amino acid protein, which includes a 24 amino acid signal peptide. The mature form of the protein is a 394 amino acid, 52 kDa glycoprotein with three asparagine-linked carbohydrate side chains (Carrell et al. 1982). Serum concentrations of AAT are genetically determined by the two alleles of the SERPINA1 coding gene, expressed co-dominantly, and normal serum concentrations of AAT are in the range of 1.04 g/L to 2.76 g/L (20 – 53 µM) (Brantly et al. 1991). Over 30 mg/kg body weight of AAT is secreted into the circulation daily, and the glycosylated protein has a serum halflife of 4 – 5 days (Jeppsson et al. 1978; Jones et al. 1978).

Fig. 1. SERPINA1 gene encoding the alpha-1 antitrypsin protein. The protein coding region is within exons II-V, with the S and Z mutations arising in exon III and exon V respectively.

Although AAT diffuses though all tissues of the body its main site of action is the lung, where it protects the fragile alveolar, connective, and epithelial tissues from proteolytic attack by the omnivorous serine protease neutrophil elastase (NE) (Bieth 1986). In this regard, AAT is the archetype of the serine protease inhibitor or serpin super-family, whose primary function is to inhibit a family of proteolytic enzymes with serine at the active site.

hepatocyte expresses approximately 200 times more AAT mRNA than other cells (Rogers et al. 1983) and serum levels can rapidly increase by between two- and five-fold during infection, trauma, surgery and burns (Kossmann et al. 1995; Voulgari et al. 1982; Sandford et al. 1999; Jeschke, Barrow, and Herndon 2004). The AAT released by inflammatory cells is more relevant to the local inflammatory milieu, serving to limit local degradation of the extracellular matrix by proteases released from migrating immune cells (Knoell et al. 1998). In a similar fashion, the range of cells producing AAT locally in the lung underlines its key role in protecting the fragile lung parenchyma against proteolytic damage. AAT was originally named because of its ability to inhibit pancreatic trypsin (Schultze, Heide, and Haupt 1962) and while its principal function as an antiprotease has been well established, AAT has significant anti-inflammatory properties affecting numerous cell types, and has been implicated in areas as diverse as apoptosis (Petrache, Fijalkowska, Medler, et al. 2006), chemotaxis (Bergin et al. 2010), tissue repair (Dabbagh et al. 2001), innate immunity (Bergin

The gene encoding AAT is the SERPINA1 gene, found on the long arm of chromosome 14 at q32.1 and is composed of four coding exons (II, III, IV, V), three untranslated exons (Ia, Ib, and Ic) in the 5' region and six introns. The hepatocyte transcription start site resides within exon Ic, while the mononuclear phagocyte transcription start sites reside within exons Ia and Ib. The SERPINA1 gene encodes a 418 amino acid protein, which includes a 24 amino acid signal peptide. The mature form of the protein is a 394 amino acid, 52 kDa glycoprotein with three asparagine-linked carbohydrate side chains (Carrell et al. 1982). Serum concentrations of AAT are genetically determined by the two alleles of the SERPINA1 coding gene, expressed co-dominantly, and normal serum concentrations of AAT are in the range of 1.04 g/L to 2.76 g/L (20 – 53 µM) (Brantly et al. 1991). Over 30 mg/kg body weight of AAT is secreted into the circulation daily, and the glycosylated protein has a serum half-

Fig. 1. SERPINA1 gene encoding the alpha-1 antitrypsin protein. The protein coding region is within exons II-V, with the S and Z mutations arising in exon III and exon V respectively.

Although AAT diffuses though all tissues of the body its main site of action is the lung, where it protects the fragile alveolar, connective, and epithelial tissues from proteolytic attack by the omnivorous serine protease neutrophil elastase (NE) (Bieth 1986). In this regard, AAT is the archetype of the serine protease inhibitor or serpin super-family, whose primary function is to inhibit a family of proteolytic enzymes with serine at the active site.

et al. 2010), and cell signalling (Koulmanda et al. 2008).

life of 4 – 5 days (Jeppsson et al. 1978; Jones et al. 1978).

Other prominent serpins include alpha-1 antichymotrypsin, alpha-2 antiplasmin, antithrombin, plasminogen activator inhibitor I, and C1 inhibitor (Law et al. 2006). The serpins act like molecular snares or mousetraps, existing in a metastable state, until they entice the unsuspecting cognate protease into their trap, and their energy is released by cleavage of their reactive centre loop (Carrell and Lomas 2002). AAT irreversibly inhibits serine proteases by presenting its inhibitory amino acid sequence as a mobile, exposed reactive centre loop with a methionine residue at its centre (Elliott et al. 1996). This reactive centre loop is a pseudosubstrate for target proteases, such as NE, and upon docking of the protease AAT undergoes a conformational change that results in the irreversible inhibition and destruction of the protease (Huntington, Read, and Carrell 2000). Although AAT can regulate the activity of many other serine proteases including trypsin, chymotrypsin, plasmin, thrombin, plasminogen, cathepsin G and proteinase 3 (Silverman et al. 2001; Travis and Salvesen 1983), it exhibits the highest association constant for NE which is 25 times greater than for any other protease (Beatty, Bieth, and Travis 1980). In addition, NE is the most relevant protease in the lung setting as it is released in large amounts by activated neutrophils (Gadek 1992).

### **2.2 Alpha-1 antitrypsin deficiency (AATD)**

#### **2.2.1 History and clinical features**

AATD is a hereditary monogenic disorder caused by mutations within the SERPINA1 gene, previously known as the "protease inhibitor" or "PI" gene. The condition was first reported in the early 1960s by Carl-Bertil Laurell and Sten Eriksson at the University of Lund in Sweden when they noticed the absence of the alpha-1 globulin band on routine serum protein electrophoresis of five patients (Laurell and Eriksson 1963). It was already known that over 90% of the alpha-1 band comprised a single protein that was capable of inhibiting the proteolytic enzyme trypsin (Jacobsson 1955; Schultze, Heide, and Haupt 1962). Following a review of the five patients, three were found to have developed "a degenerative pulmonary disease" at a young age and the condition we now know as AATD was born (Eriksson 1964). Interestingly, the oldest reported case is thought to have been a young girl who had remained undiscovered in the Alaskan permafrost for over 800 years (Zimmerman, Jensen, and Sheehan 2000). More recently there has been speculation that the composer Frédéric Chopin suffered from AATD with his physician failing to confirm pulmonary tuberculosis instead discovering "a disease not previously encountered" on autopsy (Kuzemko 1994; Kubba and Young 1998).

AATD is associated with a substantially increased risk for the development of COPD, often by the third or fourth decade, and is also associated with risks for development of liver disease (Sharp et al. 1969; Lieberman, Mittman, and Gordon 1972; Sveger 1976), cutaneous panniculitis (Edmonds, Hodge, and Rietschel 1991), bronchiectasis (King et al. 1996), vasculitis (Lewis et al. 1985), and Wegener's granulomatosis (Barnett, Sekosan, and Khurshid 1999). AATD is characterised by misfolding of the AAT protein and belongs to a class of genetic diseases underpinned by aberrant protein folding collectively termed conformational disorders (Gooptu and Lomas 2009; Greene et al. 2008). In addition, AATD is a paradigm of the subclass of conformational diseases known as the serpinopathies (Carrell and Lomas 1997). Serpins have a highly conserved structure of three β–sheets, nine α–helices, and an exposed reactive centre loop (Gettins 2002). Serpinopathies are

Alpha-1 Antitrypsin Deficiency – A Genetic Risk Factor for COPD 183

Mutations in the SERPINA1 gene that confer an increased risk for the development of COPD and/or liver disease are those in which deficiency or Null alleles are combined in homozygous or heterozygous states, and encode AAT plasma levels below a putative protective threshold of 11 µM or 0.57 g/L (Brantly et al. 1991). The two most common mutations associated with disease in populations of European descent are the Z (Glu342Lys) and S (Glu264Val) mutations, and both are caused by a single amino acid replacement of glutamic acid at positions 342 and 264 of the mature protein, respectively. In general, AAT alleles can be classified according to plasma levels and function and divided among four

Normal alleles are most commonly M subtypes which account for 95% of gene variants and are characterized by normal plasma levels in homozygotes (greater than 1.04 g/L or 20 µM).

The Z allele is the most common deficiency variant, with plasma levels of ZZ homozygotes in the range of 0.10 – 0.52 g/L (2 – 10 µM). The S variant is also common but is only clinically significant if inherited with Z or other severe mutations (Mmalton, Null etc.). For example, SZ individuals have AAT plasma levels in the range of 0.33 – 0.98 g/L (6 - 20 µM).

The class of SERPINA1 mutations termed silent or "Null" cause a complete abrogation of AAT production (Lee and Brantly 2000) and while ultra rare, confer a particularly high risk of emphysema (Fregonese et al. 2008). As these mutations do not cause the AAT protein to polymerise they pose no risk of liver disease. Most frequent among this class are those mutations that introduce a premature stop codon, for example Q0cairo (Zorzetto et al. 2005).

Like Null variants, dysfunctional alleles are extremely rare. For example, the single amino acid change caused by the Pittsburgh mutation (Met358Arg) at the active site of the AAT molecule converts it from an elastase inhibitor to a thrombin inhibitor and was discovered

The Z mutation (Glu342Lys) in AAT is an excellent example of the destabilizing effect of a mutation near the critical reactive centre loop of the protein. The proximity of the Z mutation to the loop and its location at the head of strand 5 of β–sheet A causes the AAT molecule to adopt a more open and promiscuous conformation (Gooptu et al. 2000). Thus, the β–sheet A of one molecule can readily accept the reactive loop of another AAT molecule to form a dimer. Elongation of this dimer leads to the formation of loop-sheet polymers in which the reactive centre loop of one AAT molecule sequentially inserts into the accessible β–sheet of another (Lomas et al. 1992; Sivasothy et al. 2000; Ekeowa et al. 2010). This mechanism of polymerization forms the basis for the liver disease observed in ZZ individuals, and the plasma deficiency arising from the intracellular accumulation of entangled polymers of AAT in the liver forms the basis for the lung disease observed in ZZ individuals. The Siiyama (Ser53Phe) and Mmalton (deletion of 52Phe) alleles predispose to polymerisation of the AAT molecule through a similar mechanism of loop-sheet formation.

in a child who died from an episodic bleeding disorder (Owen et al. 1983).

Other normal variants include V, T, and Lfrankfurt (Faber et al. 1994).

broad groups: **a. Normal** 

**b. Deficient** 

**c. Null** 

**d. Dysfunctional** 

Fig. 2. Serum protein electrophoresis comparison between MM and ZZ individuals. Note the complete absence of the alpha-1 globulin band in ZZ individual as first described by Laurell and Eriksson in 1963.

characterized by mutations that subvert this serpin primary structure to permit the formation of intracellular polymers and this can have pronounced deleterious consequences. In the rare condition familial encephalopathy with neuroserpin inclusion bodies (FENIB) polymerisation of the neuron-specific serpin neuroserpin, caused by point mutations in the shutter domain region of the protein, favours the accumulation of intraneuronal aggregates and eventually leads to early onset dementia (Davis et al. 1999; Belorgey et al. 2002). In contrast to this gain of (pathological) function causing dementia, a loss of (physiological) function, specifically a lack of circulating serpin in individuals with C1 inhibitor, antithrombin, and alpha-1 antichymotrypsin deficiency facilitates unchecked activity of cognate proteases and the development of angioedema, thrombosis, and COPD, respectively (Lomas and Mahadeva 2002).

#### **2.2.2 Molecular pathophysiology**

The SERPINA1 gene is highly pleiomorphic with over 120 allelic variants, including 80 deficiency alleles, identified to date (DeMeo and Silverman 2004). The majority of individuals carry two copies of the normal AAT allele, termed M, and are designated M homozygous (Fagerhol and Laurell 1970). The technique of starch gel electrophoresis originally used to separate AAT variants is responsible for the nomenclature used to identify the earliest described variants. These variants were originally designated according to how fast they migrated on starch gel, for example M (medium), S (slow), and F (fast) (Fagerhol and Laurell 1970). As detection techniques advanced and proteins began to be separated on the basis of their isoelectric point (for AAT between pH 4 – 5), the nomenclature system was revised so AAT variants were designated with earlier letters of the alphabet if exhibiting anodal migration and later letters of the alphabet if exhibiting cathodal migration. Furthermore, as the letters of the alphabet were exhausted, place of origin names are used in addition to the letter of the closest anodal allele (Cox, Johnson, and Fagerhol 1980). More precisely the birthplace of the first described individual to carry a novel allele is used, for example Q0cairo was used to describe a novel Null mutation found in the first recognised case whose birthplace was Cairo (Zorzetto et al. 2005).

Mutations in the SERPINA1 gene that confer an increased risk for the development of COPD and/or liver disease are those in which deficiency or Null alleles are combined in homozygous or heterozygous states, and encode AAT plasma levels below a putative protective threshold of 11 µM or 0.57 g/L (Brantly et al. 1991). The two most common mutations associated with disease in populations of European descent are the Z (Glu342Lys) and S (Glu264Val) mutations, and both are caused by a single amino acid replacement of glutamic acid at positions 342 and 264 of the mature protein, respectively. In general, AAT alleles can be classified according to plasma levels and function and divided among four broad groups:

#### **a. Normal**

182 Chronic Obstructive Pulmonary Disease – Current Concepts and Practice

Fig. 2. Serum protein electrophoresis comparison between MM and ZZ individuals. Note the complete absence of the alpha-1 globulin band in ZZ individual as first described by

characterized by mutations that subvert this serpin primary structure to permit the formation of intracellular polymers and this can have pronounced deleterious consequences. In the rare condition familial encephalopathy with neuroserpin inclusion bodies (FENIB) polymerisation of the neuron-specific serpin neuroserpin, caused by point mutations in the shutter domain region of the protein, favours the accumulation of intraneuronal aggregates and eventually leads to early onset dementia (Davis et al. 1999; Belorgey et al. 2002). In contrast to this gain of (pathological) function causing dementia, a loss of (physiological) function, specifically a lack of circulating serpin in individuals with C1 inhibitor, antithrombin, and alpha-1 antichymotrypsin deficiency facilitates unchecked activity of cognate proteases and the development of angioedema, thrombosis, and COPD,

The SERPINA1 gene is highly pleiomorphic with over 120 allelic variants, including 80 deficiency alleles, identified to date (DeMeo and Silverman 2004). The majority of individuals carry two copies of the normal AAT allele, termed M, and are designated M homozygous (Fagerhol and Laurell 1970). The technique of starch gel electrophoresis originally used to separate AAT variants is responsible for the nomenclature used to identify the earliest described variants. These variants were originally designated according to how fast they migrated on starch gel, for example M (medium), S (slow), and F (fast) (Fagerhol and Laurell 1970). As detection techniques advanced and proteins began to be separated on the basis of their isoelectric point (for AAT between pH 4 – 5), the nomenclature system was revised so AAT variants were designated with earlier letters of the alphabet if exhibiting anodal migration and later letters of the alphabet if exhibiting cathodal migration. Furthermore, as the letters of the alphabet were exhausted, place of origin names are used in addition to the letter of the closest anodal allele (Cox, Johnson, and Fagerhol 1980). More precisely the birthplace of the first described individual to carry a novel allele is used, for example Q0cairo was used to describe a novel Null mutation found

in the first recognised case whose birthplace was Cairo (Zorzetto et al. 2005).

Laurell and Eriksson in 1963.

respectively (Lomas and Mahadeva 2002).

**2.2.2 Molecular pathophysiology** 

Normal alleles are most commonly M subtypes which account for 95% of gene variants and are characterized by normal plasma levels in homozygotes (greater than 1.04 g/L or 20 µM). Other normal variants include V, T, and Lfrankfurt (Faber et al. 1994).

#### **b. Deficient**

The Z allele is the most common deficiency variant, with plasma levels of ZZ homozygotes in the range of 0.10 – 0.52 g/L (2 – 10 µM). The S variant is also common but is only clinically significant if inherited with Z or other severe mutations (Mmalton, Null etc.). For example, SZ individuals have AAT plasma levels in the range of 0.33 – 0.98 g/L (6 - 20 µM).

#### **c. Null**

The class of SERPINA1 mutations termed silent or "Null" cause a complete abrogation of AAT production (Lee and Brantly 2000) and while ultra rare, confer a particularly high risk of emphysema (Fregonese et al. 2008). As these mutations do not cause the AAT protein to polymerise they pose no risk of liver disease. Most frequent among this class are those mutations that introduce a premature stop codon, for example Q0cairo (Zorzetto et al. 2005).

#### **d. Dysfunctional**

Like Null variants, dysfunctional alleles are extremely rare. For example, the single amino acid change caused by the Pittsburgh mutation (Met358Arg) at the active site of the AAT molecule converts it from an elastase inhibitor to a thrombin inhibitor and was discovered in a child who died from an episodic bleeding disorder (Owen et al. 1983).

The Z mutation (Glu342Lys) in AAT is an excellent example of the destabilizing effect of a mutation near the critical reactive centre loop of the protein. The proximity of the Z mutation to the loop and its location at the head of strand 5 of β–sheet A causes the AAT molecule to adopt a more open and promiscuous conformation (Gooptu et al. 2000). Thus, the β–sheet A of one molecule can readily accept the reactive loop of another AAT molecule to form a dimer. Elongation of this dimer leads to the formation of loop-sheet polymers in which the reactive centre loop of one AAT molecule sequentially inserts into the accessible β–sheet of another (Lomas et al. 1992; Sivasothy et al. 2000; Ekeowa et al. 2010). This mechanism of polymerization forms the basis for the liver disease observed in ZZ individuals, and the plasma deficiency arising from the intracellular accumulation of entangled polymers of AAT in the liver forms the basis for the lung disease observed in ZZ individuals. The Siiyama (Ser53Phe) and Mmalton (deletion of 52Phe) alleles predispose to polymerisation of the AAT molecule through a similar mechanism of loop-sheet formation.

Alpha-1 Antitrypsin Deficiency – A Genetic Risk Factor for COPD 185

The Siiyama allele is the most common cause of AATD in Japan (Seyama et al. 1995), while the Mmalton variant (also known as Mcagliari and Mnichinan) is the most commonly found

The formation of rogue AAT polymers can also be caused by the S (Glu264Val) and I (Arg39Cys) alleles, which are associated with milder plasma deficiencies and less risk of disease (Curiel et al. 1989; Graham et al. 1989). The point mutations underlying these variants cause less disruption to β–sheet A when compared to the Z mutation. Thus, the rate of polymer production for each variant is much slower than Z AAT, leading to less retention of protein within liver cells, milder plasma deficiency, and a negligible risk of disease in heterozygotes. However, there is a risk of disease in compound heterozygotes. If a mild, slowly polymerising I or S variant of AAT is inherited with a rapidly polymerising Z (or potentially Mmalton) variant, the two variants when co-expressed can interact to form heteropolymers within hepatocytes, leading to cirrhosis and plasma deficiency (Mahadeva et al. 1999). In addition to the pathophysiological effects of polymer formation and plasma deficiency, the Z AAT protein that does make it out of the liver and is available to defend the lungs is a poor inhibitor of NE. It has been shown that Z AAT is five times less effective at inhibiting NE when compared to normal M AAT (Ogushi et al. 1987). The S AAT protein is also a slightly less efficient inhibitor of NE when compared to M AAT, but the reduction is

The final deficient mutation that warrants mention is the F (Arg223Cys) allele. Like the Z, S, and I variants, the anodal F variant was first described by starch gel electrophoresis (Fagerhol and Braend 1965) but the molecular basis for the allele was not identified until much later (Okayama et al. 1991). The point mutation in this variant introduces a cysteine instead of an arginine, and the same amino acid substitution occurs in the I variant. The normal AAT molecule possesses a single cysteine residue and the introduction of a second cysteine potentially favours the formation of disulphide bonds intramolecularly and intermolecularly with other AAT molecules. Interestingly, and possibly a reflection of the extra cysteine residue, the major F bands run as doublets on IEF gels. In the disease context, the inhibitory activity of the F AAT protein against NE is reduced, with the *in vivo* inhibition time for FZ three times longer than for MZ (Cook et al. 1996). This would suggest that individuals who co-inherit the F allele with another severe deficiency allele such as Z or Null would have a high risk for the development of COPD. The rate of polymerisation of the F variant has not been investigated but it may well exhibit a higher rate of polymerisation. A case report described finding hepatomegaly and globules positive for AAT in a liver biopsy from an FZ individual with emphysema (Kelly et al. 1989). Unfortunately, there have been no reports published to date describing F homozygotes (or I homozygotes) as these might shed some light on the polymerigenicity of the F protein and associated risk of disease.

Since the first description of AATD as a distinct clinical entity by Laurell and Eriksson almost 50 years ago, a wealth of data pertaining to the epidemiology of AATD has been accumulated. However, although a large number of groups have been investigated, the majority of studies have been biased to varying degrees as they investigated cohorts of symptomatic individuals, blood donors, soldiers, seamen, hospital staff, pregnant women, rheumatic disorders or other diseases, newborn infants and work-based employees (Blanco,

allele in Sardinia (Ferrarotti et al. 2005).

marginal in contrast to Z AAT (Ogushi et al. 1988).

**2.2.3 Epidemiology** 


Table 1. Molecular basis of the most common SERPINA1 variants associated with disease.

Fig. 3. Phenotype analysis of the most common deficient AAT variants by agarose gel isoelectric focusing followed by immunofixation (Sebia).

Polymerisation, impaired secretion and Lung & liver

Lung & liver

Lung & liver

Lung & liver (in compound heterozygotes e.g. SZ)

Lung & liver (case reports in compound heterozygotes

Lung (case reports in compound heterozygotes

e.g. IZ)

e.g. FZ)

severe plasma deficiency

severe plasma deficiency

severe plasma deficiency

impaired secretion and

impaired secretion and

Impaired secretion and mild plasma deficiency

Impaired secretion and mild plasma deficiency

Defective neutrophil elastase inhibition

Table 1. Molecular basis of the most common SERPINA1 variants associated with disease.

Fig. 3. Phenotype analysis of the most common deficient AAT variants by agarose gel

isoelectric focusing followed by immunofixation (Sebia).

No AAT production Lung

Variant Mutation Effect Disease Risk

**Z** GAG – AAG,

Glu342Lys

**Null** Mutations causing

**S** GAA – GTA,

**I** CGC – TGC,

**F** CGT – TGT,

gene deletion, premature stop codon or mRNA degradation

Glu264Val

Arg39Cys

Arg223Cys

**Siiyama** TCC – TTC, Ser53Phe Polymerisation,

**Mmalton** TTC, Phe52 Polymerisation,

The Siiyama allele is the most common cause of AATD in Japan (Seyama et al. 1995), while the Mmalton variant (also known as Mcagliari and Mnichinan) is the most commonly found allele in Sardinia (Ferrarotti et al. 2005).

The formation of rogue AAT polymers can also be caused by the S (Glu264Val) and I (Arg39Cys) alleles, which are associated with milder plasma deficiencies and less risk of disease (Curiel et al. 1989; Graham et al. 1989). The point mutations underlying these variants cause less disruption to β–sheet A when compared to the Z mutation. Thus, the rate of polymer production for each variant is much slower than Z AAT, leading to less retention of protein within liver cells, milder plasma deficiency, and a negligible risk of disease in heterozygotes. However, there is a risk of disease in compound heterozygotes. If a mild, slowly polymerising I or S variant of AAT is inherited with a rapidly polymerising Z (or potentially Mmalton) variant, the two variants when co-expressed can interact to form heteropolymers within hepatocytes, leading to cirrhosis and plasma deficiency (Mahadeva et al. 1999). In addition to the pathophysiological effects of polymer formation and plasma deficiency, the Z AAT protein that does make it out of the liver and is available to defend the lungs is a poor inhibitor of NE. It has been shown that Z AAT is five times less effective at inhibiting NE when compared to normal M AAT (Ogushi et al. 1987). The S AAT protein is also a slightly less efficient inhibitor of NE when compared to M AAT, but the reduction is marginal in contrast to Z AAT (Ogushi et al. 1988).

The final deficient mutation that warrants mention is the F (Arg223Cys) allele. Like the Z, S, and I variants, the anodal F variant was first described by starch gel electrophoresis (Fagerhol and Braend 1965) but the molecular basis for the allele was not identified until much later (Okayama et al. 1991). The point mutation in this variant introduces a cysteine instead of an arginine, and the same amino acid substitution occurs in the I variant. The normal AAT molecule possesses a single cysteine residue and the introduction of a second cysteine potentially favours the formation of disulphide bonds intramolecularly and intermolecularly with other AAT molecules. Interestingly, and possibly a reflection of the extra cysteine residue, the major F bands run as doublets on IEF gels. In the disease context, the inhibitory activity of the F AAT protein against NE is reduced, with the *in vivo* inhibition time for FZ three times longer than for MZ (Cook et al. 1996). This would suggest that individuals who co-inherit the F allele with another severe deficiency allele such as Z or Null would have a high risk for the development of COPD. The rate of polymerisation of the F variant has not been investigated but it may well exhibit a higher rate of polymerisation. A case report described finding hepatomegaly and globules positive for AAT in a liver biopsy from an FZ individual with emphysema (Kelly et al. 1989). Unfortunately, there have been no reports published to date describing F homozygotes (or I homozygotes) as these might shed some light on the polymerigenicity of the F protein and associated risk of disease.

#### **2.2.3 Epidemiology**

Since the first description of AATD as a distinct clinical entity by Laurell and Eriksson almost 50 years ago, a wealth of data pertaining to the epidemiology of AATD has been accumulated. However, although a large number of groups have been investigated, the majority of studies have been biased to varying degrees as they investigated cohorts of symptomatic individuals, blood donors, soldiers, seamen, hospital staff, pregnant women, rheumatic disorders or other diseases, newborn infants and work-based employees (Blanco,

Alpha-1 Antitrypsin Deficiency – A Genetic Risk Factor for COPD 187

Moreover, as current genotyping methods only identify Z and S alleles it is worth considering that allele frequencies described in population studies using genotyping methods will potentially underestimate AATD as the rarer SERPINA1 mutations such as I,

Across Europe the frequency of the Z and S mutations varies widely between countries, geographic regions, and ethnic groupings. It is estimated that approximately 3 - 4% of northern Europeans carry the Z allele and 6% carry the S allele (Blanco, de Serres et al. 2006). The frequency of the Z variant is highest in northern and western European countries with a mean gene frequency of 0.014, peaking in southern Scandinavia with a gene frequency of greater than 0.02, and in general declines from west to east and north to south throughout the continent (Luisetti and Seersholm 2004). The highest frequency for the Z allele recorded to date of 0.045 was found in western Latvia in a region that has seen centuries of immigration from mainland Sweden (Beckman et al. 1999). Haplotype analysis has estimated that the Z mutation first arose from a single origin 66 generations or 2,000 years ago (Cox, Woo, and Mansfield 1985; Byth, Billingsley, and Cox 1994) and its high frequency in southern Scandinavia suggests that the mutation arose in this area and was subsequently dispersed by migration patterns, for example the Viking colonisation of north-western

The region with the highest frequency of the S allele described to date is the Iberian Peninsula with a mean gene frequency of 0.0564. The highest reported frequencies of the S allele in Europe are in the region of Galicia in north-western Spain with 149 alleles found per 1000 (0.149) (Carracedo and Concheiro 1983) and in a Portuguese study with 152 alleles found per 1000 (0.152) (Santos Rosa and Robalo Cordeiro 1986). The allele frequency for S appears to decrease along south to north and west to east gradients, and is extremely low in Serbia and Russia (Blanco, Fernandez, and Bustillo 2001). In general high S frequencies are found all along the western Atlantic seaboard but accepting the premise that the higher the allele frequency in a particular country, the more likely it is to have arisen there first, it would appear the S allele arose in the north-western corner of the Iberian peninsula between

The only other AAT alleles whose origins can be reliably predicted are the Siiyama mutation, which appears to have arisen exclusively in Japan, and the Mmalton mutation, which is remarkably common in central Italy and on the island of Sardinia (Ferrarotti et al. 2005). Interestingly, the same mutation underlying Mmalton appears to have also arisen independently in the Japanese population (Matsunaga et al. 1990). Undoubtedly, more surveys relating to the epidemiology of AATD are necessary as there is a remarkable absence of AAT allele frequency data in many regions of Europe, particularly in eastern countries, as well as the lack of data pertaining to the frequencies and origins of rarer AAT

Upon reviewing the European allele frequencies for the Z and S mutations one is compelled to ask the obvious question: why are these two mutations so common? Could there possibly be an advantage to having one of these mutations? This would then provide the selective pressure to retain these mutations. An intriguing theory put forward by David Lomas in an effort to explain the high prevalence of AATD in European populations surmised that the Z and S mutations conferred a survival advantage on heterozygotes in the pre-antibiotic era

F, and Mmalton are missed by genotyping methods.

Europe between 800 and 1200 AD (Hutchison 1998).

10,000 and 15,000 years ago (Seixas et al. 2001).

alleles such as the I and F mutations.

Fig. 4. Phenotype analysis of rare deficient AAT variants (I, F, Mmalton, Zbristol, and Q0) by isoelectric focusing followed by immunofixation. Reference MM, MS, and ZZ standards are included for clarity. The Mmalton variant depicted is from an individual homozygous for this mutation.

Fernandez, and Bustillo 2001). Allele frequencies collected from patients with pulmonary disease cannot be included as they would yield disproportionately high allele frequencies. Findings from several investigations of healthy control cohorts are also unreliable as these by their definition exclude symptomatic ZZ, SZ, and MZ individuals, and this leads to underestimation of the allele frequencies in a general population. Studies of blood bank donors also risk underestimating AATD as these donors tend to be healthy individuals. To our knowledge, a survey carried out in Asturias (Spain) is the only truly representative investigation of AAT allele frequencies in a general population as the subjects were of all ages and randomly selected from a municipal population register (Blanco et al. 1999).

The small sample size and the less accurate isoelectrofocusing methods employed in many of these population studies represent significant limitations. To illustrate this point, a small study was performed in Ireland in 1992 to determine if an association existed between AATD and coeliac disease (Bourke et al. 1993). AAT phenotypes in 111 coeliacs were compared to those in 250 blood donor controls including some hospital staff. Allele frequencies for Z and S in the blood donor group were determined to be 0.008 and 0.04, respectively, using a polyacrylamide gel (pH 4 – 5) isoelectric focusing method. We subsequently carried out a larger analysis of 1,100 individuals randomly selected from the national electoral register and genotyped this group for Z and S mutations. In contrast to the 1983 data, we found allele frequencies for Z and S to be 0.022 and 0.054, respectively (Carroll et al. 2011). The reason for this disparity is most probably due to the small sample size of the original cohort. Methodological limitations may also be a factor. We have previously employed the same phenotyping method (Pharmacia) as was used in the 1983 study and found the MZ phenotype was often difficult to correctly identify and was uncomfortably similar to the M2 subtype, compared to the more accurate and specific Sebia method used in the more recent study. Similar to the revised findings on the prevalence of AATD in Ireland, many of the studies in other countries may have significantly underestimated the frequency of the Z and S mutations due to small sample size and/or methodological limitations.

Fig. 4. Phenotype analysis of rare deficient AAT variants (I, F, Mmalton, Zbristol, and Q0) by isoelectric focusing followed by immunofixation. Reference MM, MS, and ZZ standards are included for clarity. The Mmalton variant depicted is from an individual homozygous for

Fernandez, and Bustillo 2001). Allele frequencies collected from patients with pulmonary disease cannot be included as they would yield disproportionately high allele frequencies. Findings from several investigations of healthy control cohorts are also unreliable as these by their definition exclude symptomatic ZZ, SZ, and MZ individuals, and this leads to underestimation of the allele frequencies in a general population. Studies of blood bank donors also risk underestimating AATD as these donors tend to be healthy individuals. To our knowledge, a survey carried out in Asturias (Spain) is the only truly representative investigation of AAT allele frequencies in a general population as the subjects were of all ages and randomly selected from a municipal population register (Blanco et al. 1999).

The small sample size and the less accurate isoelectrofocusing methods employed in many of these population studies represent significant limitations. To illustrate this point, a small study was performed in Ireland in 1992 to determine if an association existed between AATD and coeliac disease (Bourke et al. 1993). AAT phenotypes in 111 coeliacs were compared to those in 250 blood donor controls including some hospital staff. Allele frequencies for Z and S in the blood donor group were determined to be 0.008 and 0.04, respectively, using a polyacrylamide gel (pH 4 – 5) isoelectric focusing method. We subsequently carried out a larger analysis of 1,100 individuals randomly selected from the national electoral register and genotyped this group for Z and S mutations. In contrast to the 1983 data, we found allele frequencies for Z and S to be 0.022 and 0.054, respectively (Carroll et al. 2011). The reason for this disparity is most probably due to the small sample size of the original cohort. Methodological limitations may also be a factor. We have previously employed the same phenotyping method (Pharmacia) as was used in the 1983 study and found the MZ phenotype was often difficult to correctly identify and was uncomfortably similar to the M2 subtype, compared to the more accurate and specific Sebia method used in the more recent study. Similar to the revised findings on the prevalence of AATD in Ireland, many of the studies in other countries may have significantly underestimated the frequency of the Z and S mutations due to small sample size and/or methodological limitations.

this mutation.

Moreover, as current genotyping methods only identify Z and S alleles it is worth considering that allele frequencies described in population studies using genotyping methods will potentially underestimate AATD as the rarer SERPINA1 mutations such as I, F, and Mmalton are missed by genotyping methods.

Across Europe the frequency of the Z and S mutations varies widely between countries, geographic regions, and ethnic groupings. It is estimated that approximately 3 - 4% of northern Europeans carry the Z allele and 6% carry the S allele (Blanco, de Serres et al. 2006). The frequency of the Z variant is highest in northern and western European countries with a mean gene frequency of 0.014, peaking in southern Scandinavia with a gene frequency of greater than 0.02, and in general declines from west to east and north to south throughout the continent (Luisetti and Seersholm 2004). The highest frequency for the Z allele recorded to date of 0.045 was found in western Latvia in a region that has seen centuries of immigration from mainland Sweden (Beckman et al. 1999). Haplotype analysis has estimated that the Z mutation first arose from a single origin 66 generations or 2,000 years ago (Cox, Woo, and Mansfield 1985; Byth, Billingsley, and Cox 1994) and its high frequency in southern Scandinavia suggests that the mutation arose in this area and was subsequently dispersed by migration patterns, for example the Viking colonisation of north-western Europe between 800 and 1200 AD (Hutchison 1998).

The region with the highest frequency of the S allele described to date is the Iberian Peninsula with a mean gene frequency of 0.0564. The highest reported frequencies of the S allele in Europe are in the region of Galicia in north-western Spain with 149 alleles found per 1000 (0.149) (Carracedo and Concheiro 1983) and in a Portuguese study with 152 alleles found per 1000 (0.152) (Santos Rosa and Robalo Cordeiro 1986). The allele frequency for S appears to decrease along south to north and west to east gradients, and is extremely low in Serbia and Russia (Blanco, Fernandez, and Bustillo 2001). In general high S frequencies are found all along the western Atlantic seaboard but accepting the premise that the higher the allele frequency in a particular country, the more likely it is to have arisen there first, it would appear the S allele arose in the north-western corner of the Iberian peninsula between 10,000 and 15,000 years ago (Seixas et al. 2001).

The only other AAT alleles whose origins can be reliably predicted are the Siiyama mutation, which appears to have arisen exclusively in Japan, and the Mmalton mutation, which is remarkably common in central Italy and on the island of Sardinia (Ferrarotti et al. 2005). Interestingly, the same mutation underlying Mmalton appears to have also arisen independently in the Japanese population (Matsunaga et al. 1990). Undoubtedly, more surveys relating to the epidemiology of AATD are necessary as there is a remarkable absence of AAT allele frequency data in many regions of Europe, particularly in eastern countries, as well as the lack of data pertaining to the frequencies and origins of rarer AAT alleles such as the I and F mutations.

Upon reviewing the European allele frequencies for the Z and S mutations one is compelled to ask the obvious question: why are these two mutations so common? Could there possibly be an advantage to having one of these mutations? This would then provide the selective pressure to retain these mutations. An intriguing theory put forward by David Lomas in an effort to explain the high prevalence of AATD in European populations surmised that the Z and S mutations conferred a survival advantage on heterozygotes in the pre-antibiotic era

Alpha-1 Antitrypsin Deficiency – A Genetic Risk Factor for COPD 189

**<sup>100</sup> \*\*\***

**Total (n=73) TDP (n=53) FHx (n=20)**

Fig. 5. Mean FEV1 (% predicted) in ZZ individuals diagnosed by symptomatic screening in a targeted detection programme (TDP) versus those diagnosed via family screening (FHx) enrolled in Irish National AATD Registry (\*\*\*p < 0.0001, t-test). The mean age of the TDP cohort was 46.1 +/- 1.4 years compared to 41.2 +/- 2.8 years in the family screened cohort, while 77% of the TDP cohort was ever-smoking compared to 65% of the family screened

supported by a study from the Danish registry which found that non-index ZZ individuals had longer estimated life expectancies when compared to index (symptomatic) cases (Seersholm, Kok-Jensen, and Dirksen 1994). More recently, data from the Swedish registry showed that ZZ individuals identified by family screening had longer median survival times compared to ZZ individuals detected by symptomatic (respiratory and non-

Many of the early guidelines for AATD screening advocated testing younger COPD patients and this is to the detriment of the larger COPD population of all ages. The age at which manifestations of airway obstruction, pulmonary emphysema, or chronic bronchitis appear in ZZ individuals is highly variable (Survival and FEV1 decline in individuals with severe deficiency of alpha1-antitrypsin. The Alpha-1-Antitrypsin Deficiency Registry Study Group 1998). While a common presentation of AATD is indeed early onset COPD, a subset of ZZ patients do not develop symptoms until much later in life, particularly if non-smokers (Campos, Alazemi, Zhang, Salathe, et al. 2009). In fact, among ZZ never-smokers the risk of liver disease increases with age (Tanash et al. 2008; Willson, Seow, and Zimmerman 2004). Numerous case reports have described AATD in elderly individuals with COPD who were lifelong never smokers (Jack and Evans 1991). Taken together, it is clear that screening for AATD should be performed in all patients with COPD regardless of advanced age or smoking history, especially as failure to do so has clinical repercussions for undiagnosed

While risk factors such as male gender and asthma can contribute to lung disease risk (Demeo et al. 2007), cigarette smoke is by far the single most important risk factor for the development of COPD in AATD patients (Janoff and Carp 1977; Mayer et al. 2006; Seersholm and Kok-Jensen 1995). It has been shown that smoking can reduce the life expectancy of a ZZ patient by up to 25 years (Survival and FEV1 decline in individuals with severe deficiency of alpha1-antitrypsin. The Alpha-1-Antitrypsin Deficiency Registry Study Group 1998). All AATD patients, including ZZ, SZ, and MZ phenotypes, need to be

**0**

**25**

**50**

**FEV1 (% predicted)**

respiratory) screening (Tanash et al. 2010).

**2.2.4.1 Smoking cessation & occupational exposures** 

cohort.

family members.

**75**

(Lomas 2006). The hypothesis describes how Z and S mutations could favour the generation of polymers at sites of inflammation. Polymers of Z and S could then help focus and amplify the host inflammatory response and help eradicate invading infectious organisms. Heat is known to induce Z and S polymer formation and AAT can be produced locally in the gut and the lung, the most common portals for invading pathogens. Febrile episodes in the infected individual could generate large quantities of polymer at the site of infection and recruit protective neutrophils. Thus, the pro-inflammatory response driven by Z polymers was probably hugely advantageous to a population living in the preantibiotic era. However, in the modern era increased life span and cigarette smoking have transformed a previously protective allele into a harmful one. In support of this theory polymers of Z AAT protein isolated from lung lavage have been shown to act as neutrophil chemoattractants and were found localised with neutrophils (Mulgrew et al. 2004; Parmar et al. 2002; Mahadeva et al. 2005). Furthermore, increased neutrophil numbers and IL-8 concentrations have been demonstrated in sputum from MZ heterozygotes (Malerba et al. 2006). It is also possible that the intermediate levels of AAT in MZ and MS individuals are associated with more exuberant protease activity and this allows for faster and more potent protease-dependent killing of harmful pathogens. Comprehensive studies in MZ and MS heterozygotes are required to provide additional evidence for this attractive theory.

#### **2.2.4 Treatment and management**

Although there is still no cure for AATD the accurate and early diagnosis of the condition remains imperative for improved health outcomes in AATD individuals. As soon as a diagnosis is established, AATD individuals can benefit from behavioural changes, careful management, and more targeted therapies. Effective management is guided by smoking cessation, vaccine administration and aggressive management of respiratory exacerbations (American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. 2003). Augmentation therapy, while prescribed in many countries, remains to be conclusively proven to benefit AATD patients (McCarthy and Dimitrov 2010). Specialised referral centres for AATD are urgently needed to ensure the best care and management of AATD patients. The national referral centre for AATD in Ireland has a rapid access weekly AATD clinic for newly-diagnosed ZZ, SZ, and MZ individuals. The clinic is coordinated by a dedicated clinical research nurse and AATD individuals are seen by a multidisciplinary team of doctors, nurses, and physiotherapists with international best practice standards of care followed. Similar models are being followed or implemented in several European countries but not all.

To examine the importance of early diagnosis and mode of ascertainment in AATD we collected spirometry data from 73 ZZ individuals enrolled in the Irish National AATD Registry. After stratification for mode of ascertainment, the mean % predicted FEV1 (forced expiratory volume in one second) was significantly higher in ZZ subjects diagnosed through family screening, compared to ZZ subjects who were diagnosed because of symptoms (Fig. 5). While impaired lung function would be expected in the symptomatic group, the preserved lung function in the family screened cohort, despite similar age and smoking history, highlights the importance of family screening as a tool for early detection and possible prevention of COPD in ZZ individuals. The findings from the Irish registry are

(Lomas 2006). The hypothesis describes how Z and S mutations could favour the generation of polymers at sites of inflammation. Polymers of Z and S could then help focus and amplify the host inflammatory response and help eradicate invading infectious organisms. Heat is known to induce Z and S polymer formation and AAT can be produced locally in the gut and the lung, the most common portals for invading pathogens. Febrile episodes in the infected individual could generate large quantities of polymer at the site of infection and recruit protective neutrophils. Thus, the pro-inflammatory response driven by Z polymers was probably hugely advantageous to a population living in the preantibiotic era. However, in the modern era increased life span and cigarette smoking have transformed a previously protective allele into a harmful one. In support of this theory polymers of Z AAT protein isolated from lung lavage have been shown to act as neutrophil chemoattractants and were found localised with neutrophils (Mulgrew et al. 2004; Parmar et al. 2002; Mahadeva et al. 2005). Furthermore, increased neutrophil numbers and IL-8 concentrations have been demonstrated in sputum from MZ heterozygotes (Malerba et al. 2006). It is also possible that the intermediate levels of AAT in MZ and MS individuals are associated with more exuberant protease activity and this allows for faster and more potent protease-dependent killing of harmful pathogens. Comprehensive studies in MZ and MS heterozygotes are

Although there is still no cure for AATD the accurate and early diagnosis of the condition remains imperative for improved health outcomes in AATD individuals. As soon as a diagnosis is established, AATD individuals can benefit from behavioural changes, careful management, and more targeted therapies. Effective management is guided by smoking cessation, vaccine administration and aggressive management of respiratory exacerbations (American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. 2003). Augmentation therapy, while prescribed in many countries, remains to be conclusively proven to benefit AATD patients (McCarthy and Dimitrov 2010). Specialised referral centres for AATD are urgently needed to ensure the best care and management of AATD patients. The national referral centre for AATD in Ireland has a rapid access weekly AATD clinic for newly-diagnosed ZZ, SZ, and MZ individuals. The clinic is coordinated by a dedicated clinical research nurse and AATD individuals are seen by a multidisciplinary team of doctors, nurses, and physiotherapists with international best practice standards of care followed. Similar models are being followed or implemented in several European countries

To examine the importance of early diagnosis and mode of ascertainment in AATD we collected spirometry data from 73 ZZ individuals enrolled in the Irish National AATD Registry. After stratification for mode of ascertainment, the mean % predicted FEV1 (forced expiratory volume in one second) was significantly higher in ZZ subjects diagnosed through family screening, compared to ZZ subjects who were diagnosed because of symptoms (Fig. 5). While impaired lung function would be expected in the symptomatic group, the preserved lung function in the family screened cohort, despite similar age and smoking history, highlights the importance of family screening as a tool for early detection and possible prevention of COPD in ZZ individuals. The findings from the Irish registry are

required to provide additional evidence for this attractive theory.

**2.2.4 Treatment and management** 

but not all.

Fig. 5. Mean FEV1 (% predicted) in ZZ individuals diagnosed by symptomatic screening in a targeted detection programme (TDP) versus those diagnosed via family screening (FHx) enrolled in Irish National AATD Registry (\*\*\*p < 0.0001, t-test). The mean age of the TDP cohort was 46.1 +/- 1.4 years compared to 41.2 +/- 2.8 years in the family screened cohort, while 77% of the TDP cohort was ever-smoking compared to 65% of the family screened cohort.

supported by a study from the Danish registry which found that non-index ZZ individuals had longer estimated life expectancies when compared to index (symptomatic) cases (Seersholm, Kok-Jensen, and Dirksen 1994). More recently, data from the Swedish registry showed that ZZ individuals identified by family screening had longer median survival times compared to ZZ individuals detected by symptomatic (respiratory and nonrespiratory) screening (Tanash et al. 2010).

Many of the early guidelines for AATD screening advocated testing younger COPD patients and this is to the detriment of the larger COPD population of all ages. The age at which manifestations of airway obstruction, pulmonary emphysema, or chronic bronchitis appear in ZZ individuals is highly variable (Survival and FEV1 decline in individuals with severe deficiency of alpha1-antitrypsin. The Alpha-1-Antitrypsin Deficiency Registry Study Group 1998). While a common presentation of AATD is indeed early onset COPD, a subset of ZZ patients do not develop symptoms until much later in life, particularly if non-smokers (Campos, Alazemi, Zhang, Salathe, et al. 2009). In fact, among ZZ never-smokers the risk of liver disease increases with age (Tanash et al. 2008; Willson, Seow, and Zimmerman 2004). Numerous case reports have described AATD in elderly individuals with COPD who were lifelong never smokers (Jack and Evans 1991). Taken together, it is clear that screening for AATD should be performed in all patients with COPD regardless of advanced age or smoking history, especially as failure to do so has clinical repercussions for undiagnosed family members.

#### **2.2.4.1 Smoking cessation & occupational exposures**

While risk factors such as male gender and asthma can contribute to lung disease risk (Demeo et al. 2007), cigarette smoke is by far the single most important risk factor for the development of COPD in AATD patients (Janoff and Carp 1977; Mayer et al. 2006; Seersholm and Kok-Jensen 1995). It has been shown that smoking can reduce the life expectancy of a ZZ patient by up to 25 years (Survival and FEV1 decline in individuals with severe deficiency of alpha1-antitrypsin. The Alpha-1-Antitrypsin Deficiency Registry Study Group 1998). All AATD patients, including ZZ, SZ, and MZ phenotypes, need to be

Alpha-1 Antitrypsin Deficiency – A Genetic Risk Factor for COPD 191

occupations reported in our ZZ COPD cohort include welders, chemical factory workers,

In the Irish health system there are dedicated smoking cessation officers both in the hospital setting and in the community, as well as another grade of health promotion officers, whose remit also includes smoking cessation. Exposure to second-hand cigarette smoke is also harmful to the lungs of AATD patients and the most important development in the area of lung health in Europe since the introduction of smokeless fuels has been the adoption in many countries of the ban on smoking in the workplace. After many decades of antismoking legislation and campaigning the Irish government implemented a law banning smoking in the workplace in 2004 (McElvaney 2004). This was the first law of its kind in Europe and paved the way for a host of other European countries to introduce similar laws (Rada 2010). The positive impact on Europe's lung health may not be significantly visible for several years, but already cigarette consumption in Ireland has dropped to 23.6% (Office of Tobacco Control, www.otc.ie) and this can only be a good thing for AATD individuals.

Influenza and pneumococcal vaccinations are recommended for all individuals with AATD (American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. 2003). Campos et al., (2007) investigated the practice of vaccinations and respiratory outcomes in AATD individuals in the United States and showed over 80% of AATD individuals had received adequate influenza and pneumococcal vaccinations during the influenza season (Campos et al. 2008). However, there was no significant difference in severity or rate of exacerbations between vaccinated and unvaccinated individuals but the authors concluded that the vaccinated group may represent 'sicker' AATD individuals. Influenza and pneumococcal vaccinations in COPD patients are documented as being effective and beneficial for patients (Halpin 2004; Fromer and Cooper 2008) and therefore vaccinations remain highly recommended for AATD individuals (American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. 2003). Further studies are necessary to ascertain the benefits of vaccinations for AATD individuals, especially whether protection is afforded

Exposure to bacterial and viral infections can potentially result in a respiratory exacerbation. Symptoms typically include increased dyspnoea, cough, and production of sputum (Hoogendoorn et al. 2010). Aggressive treatment of infections is recommended with antibiotics and specific treatments as symptomatically required as per ATS/ERS guidelines. This is particularly important as exacerbations adversely affect COPD patients and frequent exacerbations have been shown to be related to worsening health-related quality of life (HRQoL). Predictors for frequent exacerbations in COPD patients include symptoms of wheeze, cough and sputum production (Seemungal et al. 1998). Needham and Stockley (2005) investigated health status in AATD individuals over 12 months and recorded exacerbations, lung function and HRQoL. The authors concluded exacerbations occur commonly in AATD patients and correlate to worse health status. Exacerbations were associated with a decline in the gas transfer of the lung for carbon monoxide over time

painters, and firemen.

**2.2.4.2 Vaccinations** 

during the influenza season.

**2.2.4.3 Exacerbation management** 

educated about the harmful effects of cigarette smoke. Smoking cessation and the avoidance of occupational and environmental exposures (for example particulate matter, chemical vapours, and agricultural dusts) is paramount in AATD patient education (ATS/ERS guidelines, 2003). AATD patients without apparent lung disease should also be encouraged to quit smoking as this cohort offers the most realistic chance of delaying or in some cases preventing the development of COPD. Another important benefit in diagnosing a COPD patient with AATD is that he/she is twice as likely to attempt to quit smoking compared to an AAT-replete, smoking-related COPD patient (Carpenter et al. 2007). Carpenter *et al.* demonstrated that knowledge of AATD motivates smokers towards cessation when compared with COPD patients.

To illustrate the effect of smoking on lung function in AATD patients we collected spirometry data from 70 ZZ patients enrolled in the Irish National AATD Registry and stratified the group according to smoking history. The contribution of occupational exposures was not taken into account; nevertheless, the mean FEV1 (% predicted) was significantly higher in ZZ subjects who never smoked compared to ZZ subjects who ever smoked, with no difference in mean age between groups (Fig. 6). This clearly demonstrates the major role of tobacco consumption in the pathogenesis of COPD in ZZ individuals.

Fig. 6. Mean FEV1 (% predicted) in ever-smoking versus never-smoking ZZ individuals enrolled in Irish National AATD Registry (\*\*\*p < 0.0001, t-test). Mean age of ever-smoking cohort (50.9 +/- 1.4 years), mean age of never-smoking cohort (50.0 +/- 2.9 years).

However, even in the absence of significant smoking history there exists a significant risk for COPD. The first study to investigate a non-smoking ZZ cohort observed marked variability in both clinical course and lung function decline (Black and Kueppers 1978). Another US study showed that exposure to second-hand tobacco smoke in childhood can accelerate the onset of symptoms in ZZ AATD individuals (Mayer et al. 2006). A study from the Swedish registry demonstrated that, while non-smoking ZZ individuals may not develop COPD until later in life, this cohort still displays a decline in lung function (FEV1) with age, especially after the age of 50 (Piitulainen, Tornling, and Eriksson 1997). A follow up study by the same group found that an agricultural occupation was associated with decreased lung function in non-smoking ZZ individuals (Piitulainen, Tornling, and Eriksson 1998). Passive smoking was associated with an increased frequency of chronic bronchitis, but not with impaired lung function. Anecdotally, several of the ZZ individuals with COPD attending our clinic are farmers, specifically poultry and grain. A number of the other occupations reported in our ZZ COPD cohort include welders, chemical factory workers, painters, and firemen.

In the Irish health system there are dedicated smoking cessation officers both in the hospital setting and in the community, as well as another grade of health promotion officers, whose remit also includes smoking cessation. Exposure to second-hand cigarette smoke is also harmful to the lungs of AATD patients and the most important development in the area of lung health in Europe since the introduction of smokeless fuels has been the adoption in many countries of the ban on smoking in the workplace. After many decades of antismoking legislation and campaigning the Irish government implemented a law banning smoking in the workplace in 2004 (McElvaney 2004). This was the first law of its kind in Europe and paved the way for a host of other European countries to introduce similar laws (Rada 2010). The positive impact on Europe's lung health may not be significantly visible for several years, but already cigarette consumption in Ireland has dropped to 23.6% (Office of Tobacco Control, www.otc.ie) and this can only be a good thing for AATD individuals.

### **2.2.4.2 Vaccinations**

190 Chronic Obstructive Pulmonary Disease – Current Concepts and Practice

educated about the harmful effects of cigarette smoke. Smoking cessation and the avoidance of occupational and environmental exposures (for example particulate matter, chemical vapours, and agricultural dusts) is paramount in AATD patient education (ATS/ERS guidelines, 2003). AATD patients without apparent lung disease should also be encouraged to quit smoking as this cohort offers the most realistic chance of delaying or in some cases preventing the development of COPD. Another important benefit in diagnosing a COPD patient with AATD is that he/she is twice as likely to attempt to quit smoking compared to an AAT-replete, smoking-related COPD patient (Carpenter et al. 2007). Carpenter *et al.* demonstrated that knowledge of AATD motivates smokers towards cessation when

To illustrate the effect of smoking on lung function in AATD patients we collected spirometry data from 70 ZZ patients enrolled in the Irish National AATD Registry and stratified the group according to smoking history. The contribution of occupational exposures was not taken into account; nevertheless, the mean FEV1 (% predicted) was significantly higher in ZZ subjects who never smoked compared to ZZ subjects who ever smoked, with no difference in mean age between groups (Fig. 6). This clearly demonstrates the major role of tobacco consumption in the pathogenesis of COPD in ZZ individuals.

**<sup>100</sup> \*\*\***

**Total (n=70) Smoking (n=50) Never (n=20)**

Fig. 6. Mean FEV1 (% predicted) in ever-smoking versus never-smoking ZZ individuals enrolled in Irish National AATD Registry (\*\*\*p < 0.0001, t-test). Mean age of ever-smoking

However, even in the absence of significant smoking history there exists a significant risk for COPD. The first study to investigate a non-smoking ZZ cohort observed marked variability in both clinical course and lung function decline (Black and Kueppers 1978). Another US study showed that exposure to second-hand tobacco smoke in childhood can accelerate the onset of symptoms in ZZ AATD individuals (Mayer et al. 2006). A study from the Swedish registry demonstrated that, while non-smoking ZZ individuals may not develop COPD until later in life, this cohort still displays a decline in lung function (FEV1) with age, especially after the age of 50 (Piitulainen, Tornling, and Eriksson 1997). A follow up study by the same group found that an agricultural occupation was associated with decreased lung function in non-smoking ZZ individuals (Piitulainen, Tornling, and Eriksson 1998). Passive smoking was associated with an increased frequency of chronic bronchitis, but not with impaired lung function. Anecdotally, several of the ZZ individuals with COPD attending our clinic are farmers, specifically poultry and grain. A number of the other

cohort (50.9 +/- 1.4 years), mean age of never-smoking cohort (50.0 +/- 2.9 years).

compared with COPD patients.

**0**

**25**

**50**

**FEV1 (% predicted)**

**75**

Influenza and pneumococcal vaccinations are recommended for all individuals with AATD (American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. 2003). Campos et al., (2007) investigated the practice of vaccinations and respiratory outcomes in AATD individuals in the United States and showed over 80% of AATD individuals had received adequate influenza and pneumococcal vaccinations during the influenza season (Campos et al. 2008). However, there was no significant difference in severity or rate of exacerbations between vaccinated and unvaccinated individuals but the authors concluded that the vaccinated group may represent 'sicker' AATD individuals. Influenza and pneumococcal vaccinations in COPD patients are documented as being effective and beneficial for patients (Halpin 2004; Fromer and Cooper 2008) and therefore vaccinations remain highly recommended for AATD individuals (American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. 2003). Further studies are necessary to ascertain the benefits of vaccinations for AATD individuals, especially whether protection is afforded during the influenza season.

#### **2.2.4.3 Exacerbation management**

Exposure to bacterial and viral infections can potentially result in a respiratory exacerbation. Symptoms typically include increased dyspnoea, cough, and production of sputum (Hoogendoorn et al. 2010). Aggressive treatment of infections is recommended with antibiotics and specific treatments as symptomatically required as per ATS/ERS guidelines. This is particularly important as exacerbations adversely affect COPD patients and frequent exacerbations have been shown to be related to worsening health-related quality of life (HRQoL). Predictors for frequent exacerbations in COPD patients include symptoms of wheeze, cough and sputum production (Seemungal et al. 1998). Needham and Stockley (2005) investigated health status in AATD individuals over 12 months and recorded exacerbations, lung function and HRQoL. The authors concluded exacerbations occur commonly in AATD patients and correlate to worse health status. Exacerbations were associated with a decline in the gas transfer of the lung for carbon monoxide over time

Alpha-1 Antitrypsin Deficiency – A Genetic Risk Factor for COPD 193

leading cause of death in the United States (Eisner et al. 2010), COPD often is a silent and unrecognised disease, particularly in its early phases (Mannino et al. 2000). Globally, COPD is a growing cause of mortality and will become the third biggest killer by 2020 according to

In the mid-1960s there were two major discoveries that led to an exponential increase in our knowledge of COPD and the development of the proteolytic hypothesis of lung disease. These were the discovery of AATD and its association with COPD, and the induction of emphysema by intratracheal instillation of a protease (Gross et al. 1965). While numerous different elastases were subsequently shown to cause emphysema in the same model, none of the elastases known at that time had access to the human lung. This changed in 1968 when a potent elastase was discovered within human neutrophilic leukocytes, the primary acute inflammatory white blood cell in the body (Janoff and Scherer 1968). Called neutrophil elastase or NE, it was found to be exquisitely sensitive to inhibition by AAT. It was then discovered that certain products of cigarette smoke were able to destroy the anti-elastase properties of AAT (Johnson and Travis 1979). While AAT is an excellent inhibitor of NE, the Met358 amino acid at its active site is easily oxidised by cigarette smoke and oxidants released by immune cells (Carp et al. 1982; Hubbard et al. 1987). On foot of these revelations, it was proposed in the 1970s that all COPD could be due, at least in part, to a deficiency of AAT (Gadek, Fells, and Crystal 1979). In the majority of individuals who develop COPD due to smoking, the deficiency is functional and due to the inactivation of AAT by cigarette smoke and the influx of inflammatory cells. In

While smoking-related COPD is acquired, AATD is a form of inherited COPD and is responsible for approximately 1 – 3 % of COPD cases. A US study in 1986 investigated 965 COPD patients and found 1.9% were ZZ and over 8% were MZ (Lieberman, Winter, and Sastre 1986). Another US study investigated 969 patients with diagnosed with emphysema, asthma, or chronic bronchitis and found 1 ZZ case in every 31 samples, which is a case detection rate of over 3% (Brantly M 2003). Moreover, the contribution of SERPINA1 heterozygosity to COPD, while controversial, may account for over 10% of COPD cases if one includes ZZ, SZ and MZ phenotypes (Carroll et al. 2011). The classic pulmonary presentation of lung disease in AATD is severe, early onset panacinar emphysema with a basilar predominance in adults (Gishen et al. 1982). Evaluation of the lungs in ZZ individuals often shows diffuse destruction of the alveoli, first in the lower lung zones, and eventually throughout the entire lungs (Parr et al. 2004). This contrasts with the classic pattern of emphysema observed in smoking-related COPD, which is centrilobular (centriacinar) (Kim et al. 1991). However, emphysema in ZZ individuals may also occur in a diffuse distribution or predominantly in the upper lobes (Parr et al. 2004). Bronchiectasis, with or without accompanying emphysema, is less frequent (Parr et al. 2007). The most prominent early symptom is dyspnea, particularly upon exercise (McElvaney et al. 1997).

COPD is characterized by neutrophil-dominated airway inflammation and elevated protease levels in the lung (Abboud and Vimalanathan 2008). The main physiological role of AAT is to protect fragile alveolar lung tissue from attack by proteases, in particular

the WHO (WHO 2004).

**3.2 What we have learned about COPD from AATD** 

individuals with AATD, the deficiency is genetic.

**3.3 The ZZ phenotype as a genetic risk factor for COPD** 

(DLCO), but not FEV1 (Needham and Stockley 2005). Interestingly, a study investigated exacerbation frequency in AATD patients with COPD who were receiving augmentation therapy and found subjects with frequent exacerbations had the worst baseline HRQoL scores, as well as more physician visits and hospitalizations. Unfortunately, AATD patients not receiving augmentation therapy were not included for comparison (Campos, Alazemi, Zhang, Wanner, Salathe, et al. 2009).

A recent longitudinal study by Campos et al., (2009) undertaken in the United States, evaluated the effectiveness of a disease management and prevention program for AATD individuals, involving 905 individuals, over a 2 year period. The program comprised of written educational material for self-study and individualised treatment plans for exacerbations. This study illustrated improved patient compliance in the use of bronchodilators, oxygen therapy, and steroids during exacerbations. The management program significantly reduced medical visits and showed a considerably slower deterioration of HRQoL during an exacerbation (Campos, Alazemi, Zhang, Wanner, and Sandhaus 2009). A follow-up study would be beneficial by providing additional evidence to evaluate the long-term benefits of an AATD disease management program.

#### **2.2.4.4 Replacement therapy**

Replacement therapy is a specific therapy for AATD, and the therapy comprises of intravenous administration of AAT derived from human plasma (Stoller and Aboussouan 2004). At present, this treatment is available in a number of European countries and the United States (Chapman et al. 2009). Some AATD individuals may be candidates for AAT replacement therapy; however, the efficacy of this treatment remains controversial (McCarthy and Dimitrov 2010). Uncertainty persists concerning the therapy's effectiveness and ongoing randomised clinical trials are being performed to definitively assess the efficacy of the treatment. Previous trials have been under-powered and have mostly shown only biochemical efficacy with AAT levels restored to above the putative threshold in the blood and lung. There is some evidence that augmentation therapy can slow lung function decline in patients with AAT deficiency, however, patients with moderate obstruction are most likely to benefit (Modrykamien and Stoller 2009). The therapy comprises of weekly or fortnightly intravenous infusion of an AAT preparation that augments existing levels of circulating AAT in the blood.
