Role of alpha-1 antitrypsin in human health and disease


  • F. de Serres,

    Corresponding author
    1. Center for the Evaluation of Risks to Human Reproduction, National Toxicology Program, National Institute of Environmental Health Sciences, Research Triangle Park, NC, USA
    • Correspondence: Frederick de Serres, PhD, 632 Rock Creek Road, Chapel Hill, NC 27514-6716, USA.

      (fax: 1-919-967-8681; e-mail:


      Ignacio Blanco, MD, Board of Directors of the Alpha1-Antitrypsin Deficiency Spanish Registry, Lung Foundation Breathe, Spanish Society of Pneumology (SEPAR), Provenza, 108 bajo. 08029, Barcelona, Spain.

      (fax: 34-985-652006; e-mail:

    Search for more papers by this author
  • I. Blanco

    Corresponding author
    1. Board of Directors of the Alpha1-Antitrypsin Deficiency Spanish Registry, Lung Foundation Breathe, Spanish Society of Pneumology (SEPAR), Provenza, Barcelona, Spain
    • Correspondence: Frederick de Serres, PhD, 632 Rock Creek Road, Chapel Hill, NC 27514-6716, USA.

      (fax: 1-919-967-8681; e-mail:


      Ignacio Blanco, MD, Board of Directors of the Alpha1-Antitrypsin Deficiency Spanish Registry, Lung Foundation Breathe, Spanish Society of Pneumology (SEPAR), Provenza, 108 bajo. 08029, Barcelona, Spain.

      (fax: 34-985-652006; e-mail:

    Search for more papers by this author


Alpha-1 antitrypsin (AAT) deficiency is an under-recognized hereditary disorder associated with the premature onset of chronic obstructive pulmonary disease, liver cirrhosis in children and adults, and less frequently, relapsing panniculitis, systemic vasculitis and other inflammatory, autoimmune and neoplastic diseases. Severe AAT deficiency mainly affects Caucasian individuals and has its highest prevalence (1 : 2000–1 : 5000 individuals) in Northern, Western and Central Europe. In the USA and Canada, the prevalence is 1: 5000–10 000. Prevalence is five times lower in Latin American countries and is rare or nonexistent in African and Asian individuals. The key to successful diagnosis is by measuring serum AAT, followed by the determination of the phenotype or genotype if low concentrations are found. Case detection allows implementation of genetic counselling and, in selected cases, the application of augmentation therapy. Over the past decade, it has been demonstrated that AAT is a broad-spectrum anti-inflammatory, immunomodulatory, anti-infective and tissue-repair molecule. These new capacities are promoting an increasing number of clinical studies, new pharmacological formulations, new patent applications and the search for alternative sources of AAT (including transgenic and recombinant AAT) to meet the expected demand for treating a large number of diseases, inside and outside the context of AAT deficiency.


Alpha-1 antitrypsin (AAT) deficiency is a hereditary disorder, discovered 50 years ago, that has generated collaboration amongst patients and their families and clinical and basic scientists, nurses, therapists, public health professionals and pharmaceutical companies. As a result of this cooperation, substantial progress has been made in pivotal areas. On the 50th anniversary of the discovery of AAT deficiency, this review updates some basic, clinical and therapeutic advances in AAT deficiency, with the goal of increasing interest and generating more support for research in this area.

Methodology used for the current review

We performed a literature search of articles published between 1965 and 2013 via MEDLINE, EMBASE and Cochrane Library database. The following search terms were used: ‘Alpha-1 antitrypsin deficiency’, ‘Alpha-1 antitrypsin deficiency review’, ‘Alpha-1 antitrypsin deficiency treatment’, ‘Replacement therapy’ (a term progressively replaced by ‘augmentation’) and ‘Augmentation therapy’. Meta-analyses and systematic reviews based on scientific evidence, as well as a few classical articles from the 1960s cited in previously selected papers, were also included for review. With the aforementioned search terms, a total of 3872, 1221, 3100, 143 and 160 articles, respectively, were obtained. Titles and abstracts were assessed, and when they were not sufficiently explicit, the full text of these papers was analysed. Many repeated articles and those with redundant content were eliminated. Finally, a total of 141 papers were selected [1-141]. Where deemed appropriate, evidence-based recommendations of the American College of Chest Physicians Task Force [142] modified by the Canadian Thoracic Society COPD Clinical Assembly Alpha-1 Antitrypsin Deficiency Expert Working Group [143] were provided (Table 1).

Table 1. Strength of evidence and grading of recommendations
Quality of evidence
  1. Adapted from Marciniuk et al. (Ref. 75).

Grade AWell-designed randomized controlled trials with consistent and directly applicable results
Grade BRandomized trials with limitations including inconsistent results or major methodological weaknesses
Grade CObservational studies, and from generalization from randomized trials in one group of patients to a different group of patients
Strength of recommendation
Grade 1Strong recommendation, with desirable effects clearly outweighing undesirable effects (or vice versa)
Grade 2Weak recommendation, with desirable effects closely balanced with undesirable effects

Historic perspective

The human plasma inhibitory capacity of proteases was first discovered by Fermi and Pernossi at the end of the 19th century [1]. However, due to the insufficient development of laboratory techniques at that time, it was not possible to identify the responsible agent until 1955, when Schultze isolated the protein responsible for the antiprotease activity from the blood, naming it α1-antitrypsin by its location in the α1-globulins band and its ability to inhibit pancreatic trypsin [2].

In 1952, Laurell introduced the technique for electrophoresis of the plasma proteins in his laboratory, and in 1963, he observed the lack of the α band in the serum proteinograms from five patients. Laurell's young medical resident, Sten Eriksson, found that three of these patients had emphysema and an extensive history of respiratory diseases. In addition, one of the patients belonged to a family with an excess number of cases of emphysema. The two remaining patients had no notable pathology. At that time, it was thought that the α1-globulin band represented the protein inhibitor of trypsin, and this new ‘dysproteinemia’ was named AAT deficiency [3]. In 1964, Eriksson detected two homozygous siblings with severe emphysema, whose children had partial AAT deficiency [4]. In 1965, Laurell and Eriksson [5] collected 33 cases with similar electrophoretic patterns, some of them with emphysema. Taking all these data into account, they suspected that they had discovered a new hereditary disorder, transmitted from parents to their children by simple Mendelian inheritance.

In 1967 and 1968, Fagerhol and Laurell proposed the name protease inhibitor (Pi) for the set of electrophoretic variants of the gene [6]. Initially, they designated these variants with letters of the alphabet, to indicate their electrophoretic migration velocity: PiM (medium velocity), PiS (slow velocity), PiF (fast velocity) and PiZ (very slow velocity) [7].

The association of AAT deficiency with liver cirrhosis was documented by Sharp et al. [8] in 10 children in 1969. The same authors first detected diastase-resistant periodic acid-Schiff-positive inclusions in cirrhotic liver of patients with AAT deficiency. In the same year, Turino et al. [9, 10] related lung emphysema development to neutrophil elastase. In 1972, Berg and Eriksson [11] reported the association of AAT deficiency with liver cirrhosis in adults.

With the passage of time, diagnostic techniques were gradually improved, and since 1974, isoelectric focusing (IEF) has become the standard method for classifying phenotypes [12]. It was soon discovered that the cause of the different migration velocities was due to the replacement of one amino acid for another in the protein chain. In 1976, Carrell's team demonstrated that the PiS variant was produced by substitution of glutamic acid with valine at position 264 in the protein chain [13]. In 1976, Jeppson et al. [14] reported that the PiZ variant was produced by substitution of glutamic acid by lysine at position 342 and that this mutation did not affect protein synthesis, but caused severe conformational changes in the protein structure, causing faulty discharge from the liver into the blood.

In 1976, Sveger [15] reported the results of neonatal screening in 200 000 newborn infants in Sweden, amongst whom 127 PiZ cases were detected. Long-term follow-up of this cohort continues providing important data about the natural history of severe AAT deficiency [16, 17].

In 1977 and 1978, Larsson demonstrated a close relationship between lung emphysema and tobacco smoking in patients with AAT deficiency [18, 19]. At the same time, Cox proposed a new nomenclature for the Pi system, designating slow-migrating variants as L–Z and rapidly migrating variants as A–M [20, 21]. At present, IEF can be used to identify up to 30 AAT variants.

After these initial discoveries, advances in clinical and basic science have been rapid and highly relevant to the diseases associated with this genetic defect. In the 1980s, human AAT was synthesized, its gene was sequenced and cloned, and it was located on the long arm of chromosome 14 [22, 23].

In 1987, the US Food and Drug Administration (FDA), supported by the work of Gadek et al. [24], approved the marketing of Prolastin (purified AAT obtained from human serum by Bayer Laboratories) [25] for replacement therapy in adults with emphysema and severe AAT deficiency [26, 27]. Subsequently, Baxter, Grifols, CSL Behring and Kamada introduced their plasma derivative products Aralast (2002), Zemaira (2003), Trypsone (2004) and Glassia (2010) to the market.

In the 1990s, with the spread of sophisticated techniques of DNA amplification and gene sequencing (polymerase chain reaction), the number of variants was increased to 125, including more than 40 rare and null alleles [28, 29]. Rare and null variants were named with a suffix with the name of the city, region or country from the first patient detected. When referring to specific genotypes, the standard nomenclature of Pi* followed by the two allele names has recently been proposed [28-32].

In December 1991, Lomas et al. [33] discovered that the mutant Z AAT polymerizes, forming intrahepatic stable polymers. These accumulate in hepatocytes and cause liver damage, as well as a circulatory deficiency of AAT, which results in neutrophil elastase lung destruction and emphysema [34]. Later, relapsing panniculitis and systemic vasculitis were added to the increasing list of diseases associated with AAT deficiency [28, 31].

In the mid-1990s, the National Heart, Lung and Blood Institute Registry of Patients with Severe Deficiency of Alpha-1-Antitrypsin [35, 36] was created, along with the Alpha One Foundation and AlphaNet, an exemplary health management system that fully embraces affected individuals as well as their families [37, 38]. In 1997, the Alpha One International Registry (AIR) was created in Malmo, Sweden [39, 40]. In addition, effective guidelines for diagnosis and management of AAT deficiency have been published in many countries [28, 41]. In more recent years, to measure lung density, high-resolution computed tomography (CT) was incorporated as a highly sensitive technique into the more classical methods of assessment of emphysema [40].

In the past 15 years, there has been significant progress in genetic epidemiology [42-50], natural history [28, 31, 32], mechanisms of disease [33, 51, 53] and novel properties of AAT [54-59].

AAT has evolved from a simple acute-phase reactant protein to a multifunctional anti-inflammatory, immunomodulatory, anti-infective and tissue-repair molecule [51-53]. These new activities of AAT have promoted the development of an increasing number of clinical studies and new pharmacological formulations, as well as the search for alternative sources of AAT (including transgenic and recombinant AAT) to meet increasing demand for this protein, and new patent applications for treating a large number of diseases, inside and outside the context of AAT deficiency [60-62]. In this context, various formulations of recombinant AAT have been developed [63], and stem cell therapy [64] and gene therapy [65] are in their first stages. The autophagy-enhancing drugs rapamycin (also called sirolimus) and carbamazepine decreased the hepatic load of Z-AAT and improved hepatic fibrosis in a mouse model of AAT-deficiency-associated liver disease. These new studies provide strong support for currently progressing clinical trials to assess the efficacy of some proautophagic drugs, including sirolimus, carbamazepine, lithium and valproic acid [66-68].

Significant progress is being made on potential new applications of AAT. For example, it appears that aerosolized AAT (nebulized or dry powder inhaled) could become clinically available for alpha and nonalpha patients with chronic obstructive pulmonary disease (COPD), cystic fibrosis, bronchiectasis or bronchial asthma [69]. In addition, intravenous AAT has demonstrated efficacy in selected cases of autoimmune diabetes type 1, Crohn's disease, graft-versus-host disease, transplant rejection, multiple sclerosis, systemic vasculitis, fibromyalgia, chronic fatigue syndrome, rheumatoid arthritis, human immunodeficiency virus (HIV), influenza virus and other types of infection, relapsing panniculitis, myocardial infarction and possibly other inflammatory, immunological and infective disorders [60-62].

Physiological role of AAT in humans

Human AAT, also named α1 proteinase inhibitor (α1-Pi) and SERPINA1 (serine protease inhibitor, group A, member 1), is a water-soluble and tissue-diffusible, medium-sized (6.7 × 3.2 nm) circulating glycoprotein, with a molecular weight of 52 kDa and a blood half-life of 4–5 days (Fig. 1). Over 80% of AAT is synthesized and secreted by hepatocytes, and in additional quantities by monocytes, macrophages, pancreas, lung alveolar cells, enterocytes, endothelium and some cancers. Humans produce ~34 mg kg−1 day−1, resulting in high plasma concentrations of 1–2 g L−1. During acute-phase responses, AAT levels increase up to fourfold. From plasma, 80% diffuses to interstitial tissues, and 0.5–10% reaches biological fluids, including alveolar fluid (where local concentrations reach 0.1–0.3 g L−1), saliva, tears, milk, semen, bile, urine and cerebrospinal fluid [28, 31, 32, 56].

Figure 1.

Molecular structure of AAT (alpha-1 proteinase inhibitor, α-1 Pi, SERPINA1). Left: surface representation of AAT. Red spheres represent the amino acid chain, surrounded by other green, blue and yellow spheres representing carbohydrates. Methionine and serine active site protrudes from the globular contour of the molecule. Right: 3D structure of AAT with the mobile active site loop in an exposed position, supported on a frame of helices and sheets.

The specific substrate of AAT is the serine proteinase elastase, with which it reacts with one of the highest association constants known in biology (k = 6.5 × 10M−1 s−1). In addition to inhibiting the excess of free elastase from neutrophils, pancreas or bacteria, AAT neutralizes proteinase-3, myeloperoxidase, cathepsin G and α-defensins from neutrophils; chymase and tryptase from mast cells; trypsin from pancreas; granzyme-B from T lymphocytes; circulating kallikreins 7 and 14; and the coagulation cascade serine proteinases plasmin, thrombin, urokinase and factor Xa. AAT provides >90% of the antiproteinase activity in human serum, with the remaining 10% belonging to α-2 macroglobulin [54, 56, 59, 70].

Over the past decade, an increasing amount of evidence has indicated that AAT possesses multiple anti-inflammatory and tissue-protective properties independent of its broad-spectrum antiproteinase activity. For example, AAT reduces the expression of leukotriene B4 (a potent neutrophil chemoattractant), NO and the proinflammatory cytokines tumour necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, IL-8, IL-32 and monocyte chemoattractant protein-1, without interfering with the release of anti-inflammatory cytokines IL-10 and IL-1 receptor antagonist [56, 58, 59]. Besides, AAT inhibits caspases 1 and 3, protecting lung alveolar and endothelial cells, pancreas β cells, cardiomyocytes and skin fibroblasts from apoptosis [53, 72]. AAT prevents microorganisms binding to host cells, inhibiting their replication and infectivity [61, 73]. However, under inflammatory conditions, AAT can lose some of its effectiveness, because oxidants [74] and metalloproteinases [75] inactivate it by oxidation or excision of its reactive site loop, respectively.

The AAT gene locus is located in chromosome 14 (Fig. 2), and it is activated by by-products generated during inflammation and infection, such as lipopolysaccharide, β-interferon, IL-1, IL-6 and TNF-α, oxidative stress, free proteinases and serpine: proteinase complexes [28].

Figure 2.

Gene locus and schematic structure of the AAT (SERPINA1) gene. The top of the figure represents the gene locus in chromosome 14. The bottom shows a schematic representation of the gene, with several sites where various activities are located (pointed to with arrows).

The AAT gene has two alleles, which are transmitted from parents to their children by autosomal co-dominant Mendelian inheritance. Normal alleles, present in 85–90% of individuals, are designated M, and therefore, a normal individual shows an MM genotype. The most prevalent deficiency alleles are designated S and Z, and their prevalence in Caucasian populations ranges from 5% to 10% and 1% to 3%, respectively. Consequently, the vast majority of genotypes result from combinations of M, S and Z, that is, MM (normal genotype present in about of 85–95% people, expressing 100% of AAT), MS, SS, MZ, SZ and ZZ (five deficiency genotypes present in 5–15% of remaining people, expressing grosso modo 80%, 60%, 55%, 40% and 15% of AAT, respectively) [29, 31] (Table 1).

Genetic epidemiology

The highest prevalence of Pi*ZZ genotypes is found in the coastal regions and island communities of Europe on the Atlantic Ocean and its seas, with a gradual decline in prevalence in the east, until it disappears almost entirely in Asia. Specifically, the highest prevalences (1 : 1500–1 : 2000 individuals) are found in the Baltic Republics, south of the Scandinavian Peninsula and Denmark. It declines, but still remains high (1 : 2500–1 : 4000), in Belarus, Ukraine, Poland, Germany, the Netherlands, France, British Isles and the Iberian Peninsula. Prevalence gradually declines (1 : 10 000–1 : 90 000) in the outermost regions of Northern, Southern and Western Europe and almost disappears in Lapland, Southern Italy and the Balkan Peninsula [42-47] (Fig. 3). In Anglo-Saxon populations from New Zealand and Australia, Pi*ZZ values are similar to those reported in the UK (1 : 2500 and 1 : 4000). The Pi*ZZ genotype is almost absent in most regions of Asia and Africa. Its prevalence is moderate in Caucasian populations from Canada and the US (1 : 5000–1 : 6000) and is low in Mexico, the Caribbean and Central and South America [43, 46, 47].

Figure 3.

Map of AAT Pi*Z allelic frequency in Europe estimated by an inverse distance weighting (IDW) multivariate interpolation method. To display the frequency distribution of AAT deficiency in Europe, we used maps constructed with ArcMap for Windows, which uses a method of multivariate interpolation (IDW), which generates new data from known, using a logarithm based on geographical distance between the points. Black spots indicate the places where the in situ epidemiological studies were conducted. Coloured scale: shades of red and orange tones represent maximal values (21–40 per 1000), shades of yellow and green represent intermediate values (8–20 per 1000), and shades of blue represent minimal values (0–8 per 1000). The maximum frequency of Pi*Z (20–40 per 1000) was in Southern Scandinavia, Denmark and the Baltic Republics. High values, although somewhat lower than previously (15–20 per 1000) appeared in the Netherlands, Southern England, Eastern Ireland, Western France and Northwestern regions of the Iberian Peninsula. Pi*Z values ​​decrease gradually from west to east and in the most remote regions of the north and south of the continent.

Amongst an estimated world population of 5.264 billion, there might be at least 181 894 Pi*ZZ people at high risk for development of AAT-deficiency-related diseases, most of them Caucasians living in Western and Central Europe (74 000, 41% of the total), 20% in the US (34 000) and 16% in South America (Figs 4-6). In addition, there might be 1 269 054 Pi*SZ, 4 017 900 Pi*SS and 42 564 136 Pi*MZ individuals, with a potential risk of developing AAT-deficiency-related diseases, with 615 193 Pi*SZ living in Europe (74%, 458 074 Spanish, Portuguese, French and British), 252 599 in North and Central America (60% Americans), 202 575 in South America (55% Brazilians) and 29 687 in Australia and New Zealand [47]. It is particularly interesting to note that there could be regional differences within any given country for the prevalence of the Pi*Z and Pi*S alleles, as demonstrated by a study in Italy [48], which found striking differences amongst 18 of its 20 regions.

Figure 4.

Estimated numbers and distribution of Pi*ZZ genotypes in Europe. It is estimated that in Europe, there are nearly 80 000 ZZ individuals, most of them in the Iberian Peninsula, British Isles, France, Germany, Denmark, Latvia, Italy, Belgium, Romania, Netherlands, Sweden and European Russia.

Figure 5.

Estimated numbers and distribution of Pi*ZZ genotypes in North and Central America, Caribbean, South America, Australia and New Zealand. In North America, Central America, South America and the Caribbean Islands, there are 61 157 Pi*ZZ genotypes, mostly in the US with 34 395 (58%), 92% of whom are Caucasian, 7.5% Hispanic, 0.5% African American and 0% Asian. Rounding the figures, there would be 5500 Pi*ZZ in Canada, 4000 in Mexico and 6500 in Brazil. In Australia and New Zealand, there are ~6000 ZZ, mostly descendants of Anglo-Saxon people.

Figure 6.

Estimated numbers and distribution of Pi*ZZ genotypes in Africa and Asia. It is estimated that there are only about 5000 ZZ amongst >1 billion people in Africa, distributed throughout Nigeria, Mali, Morocco and Somalia. In Asia, with >4 billion people, there are 40 818 Pi*ZZ individuals, mostly in Pakistan, Thailand and Saudi Arabia.

The prevalence of AAT deficiency in COPD patients is not well established, but it has been reported by the American Thoracic Society/European Respiratory Society that it might be in the range of 1–3% [28]. However, the prevalence of COPD can differ amongst carriers of each of the five phenotypic classes of Pi*S and Pi*Z. Specifically, the odds ratios for each of the phenotypic classes amongst white COPD patients from the US demonstrate highly significant decreases in the normal phenotype Pi*MM, no significant change in Pi*MS and Pi*SS deficiency phenotypes, but highly significant increases in the prevalence of Pi*MZ, Pi*SZ and Pi*ZZ deficiency phenotypes. The database from this study supports the concept of targeted screening of AAT deficiency in countries with large populations of Caucasian COPD patients [49]. In contrast, within any given country, there may be a different prevalence of Pi*S and PI*Z in different racial subgroups. For example, an analysis of the prevalence of Pi*S and Pi*Z amongst the five major ethnic subgroups in the US has demonstrated that the highest risk for AAT deficiency is found in Caucasians, followed by Hispanics and Blacks, with the lowest prevalence amongst Mexican Americans and no risk amongst Asians [50].

AAT deficiency is probably one of the most common hereditary under-recognized conditions worldwide, with only a small minority (<2%) of individuals currently detected. Diagnostic delay ranges from 7 to 10 years, with patients usually needing several consultations with pulmonary medicine specialists to reach a correct diagnosis. Underdiagnosis and no notification constitute a major problem worldwide, attributable to: lack of awareness amongst some professionals dealing with AAT deficiency patients; clinical variability of AAT deficiency presentation, with up to one-third of Pi*ZZ patients and two-thirds of Pi*SZ patients not expressing clinical symptoms; medical misdiagnosis, with many patients classified as having COPD or hepatic cirrhosis related to smoking or alcohol; and the ‘therapeutic nihilism’ of some physicians regarding AAT augmentation therapy effectiveness, availability and costs [76].

Clinical expression of AAT deficiency

Severe AAT deficiency, defined by AAT serum levels <35% of the mean expected value (50 mg dL−1 measured by nephelometry, 11 μmol L−1, or 80 mg dL−1 measured by radial inmunodiffusion), is usually associated with Pi*ZZ genotypes and less frequently with combinations of Pi*Z, Pi*S or rare and null alleles (Table 2). In clinical practice, 96% of AAT-deficiency-related pathologies occur in Pi*ZZ homozygotes [28, 31, 32].

Table 2. Serum concentrations of Alpha-1 antitrypsin (AAT) expressed in mg dL−1 (measured by nephelometry), percentage (%) and micromols (μmol L−1), liver accumulation of polymers and risk of liver cirrhosis and pulmonary emphysema development for the different Pi* AAT genotypes
AAT Pi* genotypesAAT serum levels: mg dL−1 (%) [μmol L−1]Amount of liver polymers accumulationRisk for liver cirrhosisRisk for pulmonary emphysema
  1. Values in mg dL−1 can be expressed in micromolar units (μmol L−1) by multiplying its value by 0.1923. Conversion of μmol L−1 to mg dL−1 can be done by multiplying its value by a conversión factor of 5.2.

  2. a

    It is necessary the presence of other environmental or genetic factors (i.e. tabaquism, inhalation of toxic fumes, and/or different favouring genes not well identified so far).

  3. b

    It is necessary the presence of other enviromental or genetic factors (i.e. B and/or C hepatitis, nonsteroidal anti-inflammatory drugs (NSAIDs) and/or different favouring genes not well identified so far).

MM100–200 (80–120) [20–48]NoNormalNormal
MS100–180 (75–85) [19–35]Very slightNot increasedNormal
SS70–105 (45–70) [15–36]SlightNot increasedPossible, but not establisheda
MZ66–120 (50–70) [12–35]ModerateIncreased, but low (~3%)bUncertain, but possible in ~10% of subjectsa
SZ45–80 (30–45) [8–19]LargeSlightly increasedbSlightly increaseda



10–40 (10–20) [2.5–7]Very largeVery increased (~2.5% children; ~30% adults)Very increased (~60%)

Not detectable


NoNoGreatest risk (practically100%)



~10–15 (5–10) [2–3]ModerateLowVery increased

Severe AAT deficiency predisposes patients to the development of different types of diseases throughout their lifetime, especially lung emphysema and different types of liver disease in children and adults (Fig. 7). There is also evidence of a relationship with systemic vasculitis [typically Wegener antinuclear cytoplasmic antibody (ANCA) positive], necrotizing panniculitis and possibly, but not yet confirmed, with a myriad of different inflammatory and neoplastic diseases [28, 57, 77].

Figure 7.

Clinical consequences of AAT liver and lung polymerization. The consequences of AAT polymerization are accumulation of polymers and inclusion bodies in hepatocytes, which favour the development of liver disease in children and adults, as well as a decrease in serum and tissue concentrations of AAT, which are insufficient to protect the lungs and other organs from the proteolytic damage caused by elastase and other free proteases. AAT deficiency is not a disease by itself but a hereditary condition that predisposes to the development of several types of diseases throughout life, especially if they occur in the same individual simultaneously with other adverse environmental factors and/or other facilitator genes that interact with each other to cause disease.

AAT deficiency and lung disease

Although it is likely that AAT deficiency can promote the development of bronchial asthma [78] and bronchiectasis [79], there is no reliable evidence as to whether AAT deficiency influences the frequency or gravity of these diseases [80]. AAT-deficiency-related lung emphysema is characterized by early onset at 35–45 years, with chronic, progressive fatigue and other nonspecific respiratory symptoms. It is usually associated with Pi*ZZ genotypes and rarely with Pi*SZ, rare or null genotypes. Pi*ZZ penetrance (i.e. percentage of Pi*ZZ individuals who develop emphysema) is ~60%. In these patients, chest X-rays and CT scans show basal bilateral panacinar emphysema. In addition, pulmonary function tests show fixed bronchial obstruction with air trapping and diminished lung diffusion capacity. In the US, Canada, Germany, Austria, Spain and Italy, augmentation therapy is available for selected cases, according to established criteria contained in different national and international guidelines [28, 41, 81].

AAT deficiency and liver disease

Ninety-nine of the published cases of AAT-deficiency-related liver disease (typically cirrhosis) were Pi*ZZ. In 10% of children, the condition may become evident by the appearance of intrahepatic cholestasis (long-term obstructive jaundice). Amongst these children, 2.5% develop childhood–adolescent cirrhosis that usually requires liver transplantation. In adults (mostly men), the condition may become evident by the appearance of chronic hepatitis or liver cirrhosis, whose incidence increases in parallel with age: 10% in those aged <50 years and 20–40% in those over this age. Hepatocarcinoma appears in 2–3% and can occur in both cirrhotic and noncirrhotic livers. Augmentation therapy is not applicable for patients with liver disease [15, 16, 28, 82].

AAT deficiency and panniculitis

There are about 40 published cases of necrotizing panniculitis associated with AAT deficiency, most of them Pi*ZZ. Its estimated penetrance is 0.1%, but it could be higher if AAT deficiency were systematically tested in panniculitis. This disease is characterized by relapsing, painful, nodular, inflammatory lesions (with predominance of neutrophils and free elastase), in subcutaneous fatty tissues from the trunk, pelvic region and a segment proximal to the lower extremities. These lesions tend to fistula and scar formation. In spite of the fact that augmentation therapy has not been approved for panniculitis so far, several studies showed spectacular results when it was given as a last resort, for compassionate reasons, in patients who did not respond to conventional therapies [28, 31, 60].

AAT deficiency and systemic vasculitis

AAT-deficiency-related systemic vasculitis has been described in Pi*ZZ, Pi*SZ and Pi*MZ patients, most of them showing ANCAs. Vasculitis affects mostly middle-aged and older adults, and although it can affect nearly every organ of the body, it has a preference for the sinuses, lungs and kidneys. Its penetrance is about 2%, but it could be higher if AAT deficiency were systematically tested in vasculitis. Proteinase-3 seems to play an important role in its pathogenesis [30]. Augmentation therapy is not approved for vasculitis; nevertheless, some isolated studies have shown spectacular results when it is given as a last resort in isolated cases in patients not responding to conventional therapies [60, 83].

Variability of clinical manifestations of AAT deficiency

An intriguing characteristic of AAT deficiency is the marked variability of its clinical manifestations. Although a high percentage of AAT-deficient individuals develop COPD or liver cirrhosis, a minority suffers from panniculitis or vasculitis, some from two diseases simultaneously, and more than one-third may only have banal symptoms or even remain asymptomatic throughout life. This variability indicates that AAT deficiency is not an illness itself, but a complex monogenic disorder that predisposes to the development of different pathologies when other factors (both environmental and/or genetic) are also present (Fig. 7). For example, in the case of liver cirrhosis, hepatitis B and C viruses, chronic or recurring inflammatory processes, nonsteroidal anti-inflammatory drugs, alcohol abuse or other hepatotoxic substances can increase the polymers load in hepatocytes, with pathological accumulations of misfolded proteins and formation of inclusion bodies, secondary cellular stress and organ damage. Mutations of mannosidase I and human hemochromatosis protein genes are also associated with liver damage in Pi*ZZ individuals [28, 82, 84]. In the case of COPD, tobacco smoking, workplace and other environmental pollution, and respiratory infections are strong contributing factors [28, 85]. In addition, some genes, such as NO synthase, glutathione S-transferase P1, plasminogen activator inhibitor serpin E2 and IL-10, have been proposed as possible candidate genes [86, 87].

Clinical risk related to AAT deficiency genotypes other than Pi*ZZ

On the basis of the currently available data, carriers of Pi*MS phenotypes are not considered at risk for disease [28]. Although most studies have not demonstrated an increased prevalence of COPD in Pi*MZ nonsmokers, some have found a more rapid FEV1 decline than that found in PI*MM individuals and an increased prevalence of Pi*MZ amongst COPD patients. Available data suggest that a subgroup of around 10% of Pi*MZ individuals appear to have an increased risk for COPD [88-90]. However, the relative risk for cirrhosis development is increased when compared to that in Pi*MM individuals, but such a risk is small (3%) when compared to that of Pi*ZZ (30%) individuals [28]. In contrast to what happens with Pi*ZZ individuals, alcohol and hepatitis B and C are additional factors necessary for developing liver cirrhosis [82] in persons with the Pi*MZ genotype.

The Pi*SS genotype is infrequent (<1%) in Europe, and for this reason, there are insufficient studies with evidence of good quality to reveal whether it contributes or not to COPD or liver disease development. However, it seems likely that if the Pi*SS phenotype is associated with other factors such as tobacco smoking, it may increase susceptibility to COPD [82, 91, 92].

Heterozygotes for Pi*SZ have an increased risk for COPD, especially in smokers [89, 90], and isolated cases with liver cirrhosis have been reported [82].

Rare genotypes, such as Pi*Mmalton and Pi*Mduarte, are at least 300 times less frequent than the Pi*ZZ type, and the consequent paucity of data makes it difficult to acquire adequate knowledge of its significance. Most reported cases presented with pulmonary emphysema, and although the relative risk for liver disease was not established, their tendency to form polymeric intrahepatic inclusions suggests a similarity to Pi*ZZ [93]. Most of the 24 reported null genotypes presented with emphysema in early adult life, but because their livers do not synthesize AAT, they do not develop liver disease [30].

Under what circumstances is a test for AAT deficiency recommended?

The World Health Organization and the medical societies of Europe, Canada and the US recommend AAT should be measured at least once in the blood of all patients with COPD, regardless of whether they smoke. It is also recommended if early-onset COPD occurs in nonsmokers or if there is a family history of COPD or AAT deficiency (consistent recommendation with a high quality of evidence, Table 3) [28, 81, 94]. These same organizations recommend, at least once in a lifetime, measuring AAT in the blood of all patients with chronic liver disease of unknown aetiology and especially if there are familial cases of liver disease and/or AAT deficiency (consistent recommendation with high quality of evidence, Table 3) [28, 81].

Table 3. Clinical situations in which measurement of alpha-1 antitrypsin serum levels is recommended, and if serum values are low, it is recommended to determine the patient's phenotype (with isoelectrofocusing) or its genotype (with PCR techniques)
 Strength of recommendation
  1. ANCA, antineutrophil cytoplasmic antibodies.


 At least once on life, specially if:

  • COPD with symptoms starting in early adult life or no-smokers
  • Family history of COPD
  • Family history of AAT deficiency with or without COPD
Difficult control severe asthmaModerate

 Cirrhosis/chronic hepatitis in children and in adults without evident cause, specially if:

  • Family history of liver cirrhosis of no well proven cause
  • Family history of AAT deficiency even without hepatopathy

 Systemic vasculitis, especially if ANCA+

 Relapsing panniculitis

 Blood relatives of subjects with severe AAT deficiency

Pheno and genotype determination candidates


  All the above mentioned situations with subjects showing low AAT serum levels



  Lack of correlation between low AAT serum levels and apparently normal phenotypes (‘M-like’)

  Not characterized deficient phenotypes, different from the usual S and Z


Severe adult asthma, ANCA and systemic vasculitis, and relapsing panniculitis are other conditions for which screening for AAT deficiency is recommended (consistent recommendation with moderate quality of evidence, Table 3) [28].

Intravenous AAT augmentation therapy

The only specific treatment available for emphysema associated with AAT deficiency is intravenous AAT augmentation therapy. However, despite large observational studies and limited interventional studies, controversy remains about the efficacy of this treatment. Reservations are due primarily to the impracticality of conducting adequately powered studies to evaluate the rate of decline in lung function, because of the low prevalence and slow progression of the disease [81, 95-105] (Table 4).

Table 4. Summary of selected studies on augmentation therapy
AuthorsDoses of AATType of studyOutcome measuresResultsEvidence level
  1. FEV1, forced expiratory volume in the first second; DLCO, carbon monoxide diffusion capacity; TC, computerized tomography.

No randomized studies
Seersholm et al. [92]60 mg kg−1 per 7 daysObservational with control group (= 295)FEV1 decline

Reduction in FEV1 decline in treated group (56 vs. 75 mL per año; P = 0.02)

Greater benefit in patients with FEV1 31–65%

American AAT Deficiency Registry Study Group [93]33% of cases 60 mg kg−1 per 7 days; 43%, 120 mg kg−1 per 14 days; and 24% 180 mg kg−1 per monthlyObservational with control group (= 1129)

FEV1 decline


Significant decrease in mortality (P = 0.02)

Smaller decline of FEV1 amongst patients in augmentation therapy and FEV1 35%–49% (66 vs. 93 mL year−1; = 0.03)

Wencker et al. [94]60 mg kg−1 per 7 díasObservational without control group (= 96)FEV1 decline (pre-post)Smaller decline of FEV1 during the therapy period (49 vs. 34 mL year−1, P = 0.019). Smaller decline in patients with FEV1 > 65% (256 vs 53 mL year−1, P = 0.001)C2
Tonelli et al. [95] Observational with control group (= 164)

FEV1 decline


FEV1 gain of 11 ± 21 mL year−1 vs loss of 37 ± 12 mL year−1 (P = 0.05)

No differences between groups in mortality

Randomized studies
Dirksen et al. [96]250 mg kg−1 per 28 daysDouble-blind, randomized, placebo controlled in COPD patients with FEV1 30–80% (= 56)FEV1 and DLCO decline. Loss of pulmonary density measured by computed tomography (TC)

Loss of pulmonary tissue of 2.6 g L−1 years−1 with placebo and 1.5 g L−1 year−1 with therapy (P = 0.07)

No differences in FEV1 and DLCO decline

Dirksen et al. [97]60 mg kg−1 per 7 daysDouble-blind, randomized, placebo controlled (n = 77) (COPD with FEV1 = 25−80%)Lung function, quality of life, exacerbations and loss of lung density by TC

Reduction in lung density in treated patients (= 0.049)

No differences in FEV1 and DLCO decline, neither in number of exacerbations, but they were milder amongst treated.

Chapman et al. [98] Patients in augmentation therapy at the Canadian Registry versus controls (= 1509)FEV1 declineSignificant smaller decline of FEV1 in treated subjectsB1
Gotzsche & Johansen [99]60 mg kg−1per 7 díasAnalysis of 2 randomized, placebo-controlled studies (= 140)FEV1 and DLCO decline, loss of pulmonary density measured by TC, ExacerbationsLoss of pulmonary density smaller in treated patients (= 0.03). No difference in pulmonary function tests and exacerbationsB2
Stockley et al. [100]60 mg kg−1 per 7 daysIntegrated analysis of previous studies about pulmonary density TC-measured (= 119)Loss of pulmonary density and FEV1 decline

Significantly smaller loss of pulmonary density amongst treated patients (1.7 vs. 2.7 g L−1, = 0.006)

No difference in FEV1 declination

Marciniuk, et al. [78] Analysis of all available studies about pulmonary density TC-measuredFEV1 and DLCO decline, pulmonary density measured by TC, quality of life, exacerbations, mortalitySignificantly smaller loss of pulmonary density and mortality amongst treated patientsB1
Lieberman [101]Weekly infusions 55% subjects, biweekly in 37%, and monthly in 8%Observational (internet survey) (= 89)Exacerbations frequencyDecrease of exacerbation number and severity from 3–5 to 0–1 per year with augmentation therapyC2
Barros-Tizón et al. [102]180 mg Kg−1 per 21 daysRetrospective pre-post augmentation therapy)Frequency and severity of exacerbations and hospitalisation costsSignificant decrease of number and severity of exacerbations, as well as of hospitalization costsC1

Two clinical trials compared treatment with human AAT versus placebo in a randomized double-blind design. The first trial [100] included 58 patients treated for 3 years and showed that patients who received AAT had a lower annual loss of lung density measured by chest CT. This loss was 1.50 g L−1 as compared with 2.57 g L−1 in those control patients who received a placebo. The second study [101] included 77 patients and showed a trend towards slowing of the loss of lung density with active treatment, measured by CT. The similarities between the two studies permitted a joint analysis of the data to gain statistical power [103]. This analysis included 119 patients and showed significantly less loss of lung tissue in patients treated with human AAT. The remaining data derived from observational comparative studies show a significant decrease in FEV1 in patients with FEV1 between 30% and 60%. In addition, the National Heart, Lung and Blood Institute Registry in the US reported a significant reduction of 36% in mortality in patients receiving replacement therapy [96].

The effect of replacement therapy in critically ill patients (FEV1 <30%) is difficult to assess, because these individuals die or undergo a lung transplant before one can track long enough. In the case of mildly ill patients (FEV1 >60%), it is not easy to evaluate the effect of treatment because there is a bias for indication. The small numbers of patients receiving treatment at this early stage are index cases with severe symptoms or accelerated loss of lung function. By contrast, some index cases are untreated because they have no symptoms or maintain stable lung function.

A recent evaluation by The Canadian Thoracic Society, basing their recommendations on the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system [142, 143], concluded that placebo-controlled clinical trials demonstrate the effectiveness of augmentation therapy in slowing the progressive loss of lung density that characterizes emphysema. This positive effect of augmentation therapy supports any recommendation for treatment [81]. Therefore, AAT treatment should be considered in nonsmokers or ex-smokers with COPD and emphysema with documented AAT deficiency concentrations <50 mg dL−1. In these cases, this treatment recommendation is based on the demonstrated benefits in reduced lung density loss (grade of recommendation: 2B) and improved survival (grade of recommendation: 2C).

Augmentation treatment is not indicated for heterozygotes with phenotype Pi*MZ because of the lack of studies on the subject. However, in Pi*SZ heterozygotes, augmentation treatment can be considered in those rare cases of COPD with low serum concentrations of AAT (<50 mg dL−1).

In summary, intravenous AAT augmentation for chronic treatment of emphysema due to AAT deficiency has been proven to be safe. No cases have been found of transmission of hepatitis A, B, C or D viruses, or HIV, or prion disease [106].

Gene therapy

Gene therapy is still in the initial phases of development in humans. Four phase 1 and phase 2 studies, performed in five, 12, nine and nine patients, respectively [65, 134-136], have not given totally satisfactory results so far due to the low serum levels of AAT achieved (3–5% of the normal serum concentration) and the frequent development of antibodies against vector viruses (Table 5).

Table 5. Current status of four selected studies on AAT gene therapy
PhaseVectorAdm. routePatients numberMain resultsYear Author [ref]
  1. rAAV2, recombinant adeno-associated virus serotype 2 capsid; IM, intramuscular.

IPlasmid-Cationic Liposome complexIntranasal5AAT concentration increased in nasal lavage fluid at ⅓ normal, returning to baseline by day 14Brigham [134]
IrAAV2-AATIM12Only one subject had low-level expression of AAT in serum 30 days after an injection of 2.1 × 1013 vector genomes. Anti-AAV2 capsid antibodies rose after vector injectionBrantly [135]
IrAAV2-AATIM9AAT seen in 2 subjects at subtherapeutic levels (0.1% of normal) at 1 year. Anti-AAV capsid antibodies developed all patients by day 14Brantly [136]
IIrAAV1-AATIM9Dose-dependent increase in AAT levels, but still subtherapeutic. Transient creatinine kinase risesMuller and Flotte [63]

Gene correction of AAT deficiency in induced pluripotent stem cells (iPSCs)

A final promising therapy involves the gene correction of AAT deficiency in iPSCs by means of genetic engineering. As shown schematically in Fig. 8 (modified from [137]), iPSCs are formed from dermal fibroblasts from an AAT-deficient person. In these iPSCs, the Pi*Z mutant base is removed and replaced by a normal Pi*M sequence (by transposon excision and cut and paste procedures) to generate a normal Pi*M gene. These corrected iPSCs are then cloned and differentiated into hepatocytes in vitro and finally transplanted into the liver of the original person, where these corrected iPSCs-derived hepatocytes might replace abnormal hepatocytes. This approach has been demonstrated to work properly in vitro and in vivo in mice, as well as in human cells from individuals with hereditary AAT deficiency, providing sustained serum AAT expression. However, the risk of developing other mutations during iPSC culture cannot be ruled out, which has precluded its clinical use to date [64, 137].

Figure 8.

Person with alpha-1 antitrypsin deficiency treated with autologous corrected iPSCs. The most advanced and promising therapy consists of correction of the mutated gene Z by genetic engineering, combining zinc finger nuclease and transposons ‘piggy back’ techniques. Schematically, the process is as follows: (i) skin fibroblasts of the person with AATD are taken. (ii) These fibroblasts are induced to become iPSCs. (iii) In these iPSCs cells, the mutated nucleotide DNA base Z gene is deleted, and the correct nucleotide M is introduced to generate a normal gene M. (iv) The corrected iPSCs are cloned and differentiated into hepatocytes. (v) These iPSC-derived hepatocytes are transplanted into the liver of the same person. (vi) The hepatocytes derived from the corrected iPSCs replace abnormal hepatocytes, virtually eliminating the chance of rejection.

Potential new therapeutic uses of AAT

In the past decade, an impressive variety of in vitro and in vivo experimental studies have found that AAT is a multifunctional pan-antiprotease, anti-inflammatory, immunomodulatory, anti-infective and tissue-repair molecule [57, 59, 107]. All these activities have resulted in AAT being considered as a promising biological therapeutic agent for treating many diseases of an inflammatory, autoimmune or infectious nature, both inside and outside the field of AAT deficiency (Table 5) [60-62, 83, 132]. However, its real usefulness in clinical practice has yet to be demonstrated with well-designed, randomized, double-blind, placebo-controlled trials, with sufficient sample size to reach the highest level of evidence before being applied to patients (Table 6).

Table 6. Potential use of AAT therapy in treatment of human diseases
DiseaseTheoretical therapeutic targets and potential therapeutic effectsReferenceEvidence level
  1. LTB4, leukotriene B4; TNF-alpha, tumour necrosis factor alpha; VEGF, vascular endothelial growth factor; NGF, nerve growth factor; SP, substance P; NO, nitric oxide; CRF, corticotropin releasing factor; HIV, human immunodeficiency virus.

Aerosol AAT for alpha and non-alpha1 lungs
Non-alpha-related COPD (AAT administered by inhalation)Neutralization of serine proteinases (i.e. elastase, proteinase-3, cathepsin G and alpha-defensins) from neutrophils. Reduction in LTB4 release from macrophages. Reduction in neutrophil chemotaxis

Stockley [108]

Hubbard [109]

Bucurenci [110]

Stockley [69]

Bergin [59]

Cystic fibrosis (AAT administered by inhalation)Neutralization of proteinases from neutrophils and reduction in LTB4. Reduction in IL-8, IL-1-beta and TNF-alpha. AAT inhaled decreased sputum neutrophil counts, elastase activity and IL-8 concentration

Hansen [111]

Martin [112]

Griese [113]

Kerem [114]

Bronchiectasis (AAT administered by inhalation)Same theoretical reasoning above citedWanner [60]C2
Bronchial asthmaDecreased eosinophil count. Reduction in IgE-induced histamine release from mast cells. Reduction of number and severity of exacerbations.

He [115, 116]

Blanco [117]

Autoimmune diseases
Type-1 diabetesProtection of beta cells against inflammatory cytokines (IL-1 blockade) and apoptosis (caspase-3 inhibition).

Lewis [70]

Kalis [118]

Graft-versus-host diseaseInhibition of IL-1 production and activity. Inhibition of proteinase-3-related IL-32 activation. Increased release of IL-1 receptor antagonist and IL-10.

Marcondes [119]

Dinarello [120]

Tawara [121]

Transplant rejectionInhibition of IL-32 activation. Inhibition of IL-1 production and activity. Increased release of IL-1 receptor antagonist and IL-10.Shahaf [122]C2
Multiple sclerosis  C2
Connective tissue and rheumatic diseases
Systemic vasculitisInhibition of proteinase-3 from neutrophils

Baslund [123]

Hernández [81]

Fibromyalgia/Chronic fatigue syndromeMast cell stabilization and neutralization of its mediators (chymase, tryptase, histamine, TNF-alpha, NGF, SP, NO, CRF, etc.)

Blanco [124]

Blanco [58]

Schmechel [125]

Rheumatoid arthritisInhibition of IL-6, IL-1-beta and TNF-alpha. Neutralization of serine proteinases from neutrophils, and Aggrecanase-1 (ADAMTS-4)

Grimstein [126]

Yoshida [127]

Grimstein [128]

Infectious diseases
HIV/AIDSBinding to the gp41 fusion peptide of virus makes AAT the major plasma-derived inhibitor of HIV-1,

Shapiro [129]

Congote [71]

Münch [130]

Forssmann [131]

Other diseases
Relapsing panniculitisNeutralization of elastase and other neutrophil proteases in subcutaneous tissue

Blanco [58]

Al-Niaimi [132]

Acute myocardial infarctionReduction in caspase-1 activityToldo [133]C2

Several studies with aerosolized human AAT have demonstrated its ability to inhibit neutrophil elastase and decrease IL-8 concentration in the alveolar fluid, as well as to reduce neutrophil counts in sputum, with high safety, efficacy and tolerance [111-114]. Also, a high-purity liquid formulation of AAT has been developed for aerosol treatment, to respond to the interest of patients in the development of commercially available inhaled AAT replacement products [69]. It is expected that with this approach, local lung concentrations of AAT can be achieved at low doses, sparing both systemic delivery of AAT and cost.

There are currently three ongoing trials with weekly intravenous infusions of AAT in young patients with type 1 diabetes (NIH clinical trial registry codes NCT01304537, NCT01319331 and NCT01183468), with promising initial results, but long-term results are still pending.

At present, only human plasma-derived AAT is available for clinical use. Given the limited supply and the potential for extended use of this product, there will be a need for new formulations of AAT in the future. Using transgenic sheep that carry the genetic sequence for human AAT under the control of mammary gland promoters, biologically active human AAT has been generated in milk and purified and introduced to humans by intravenous infusion [138]. Using this method, it is estimated that a population of 4500 sheep would be able to provide 5000 kg of human AAT annually [61]. Unfortunately, a clinical trial in humans with AAT from the milk of transgenic sheep produced intolerable allergic reactions that forced their discontinuation.

Recombinant AAT has been derived from plants, yeast, fungi, animals, insect cells, bacteria and mammalian cells and modified and conjugated with polyethylene glycol to prolong its half-life in blood [139, 140], but none of these changes has managed to match the traditional plasma-derived affinity-purified form of human AAT. Recently, a recombinant AAT formulation has been developed that has achieved near-identical glycosylation when compared to plasma-derived AAT [63]. However, further studies are warranted to demonstrate the clinical safety and efficacy of this and similar products.


Since the first description of AAT deficiency 50 years ago, many researchers and health professionals have significantly contributed to a better understanding of the most important clinical and molecular aspects of this under-recognized genetic condition. At the same time, some pharmaceutical companies have made progress in the development of plasma-derived and recombinant AAT to treat this orphaned condition. In turn, many patients with AAT deficiency and their families have taken steps to enhance education, research, detection and treatment of their illness, to improve not only their own lives but also the lives of generations to come, always in the hope that they can benefit from the next research advances [141].

Conflict of interest

The authors of this manuscript have no conflict of interest to declare.