|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Diagnostics and Genetics |
1-Antitrypsin Deficiency Identifies 2 Previously Unidentified Null AllelesDepartment of Clinical Chemistry, Meander Medical Center, Amersfoort, The Netherlands.
aAddress correspondence to this author at: Department of Clinical Chemistry, Meander Medical Center, Utrechtseweg 160, P.O. Box 1502, 3800 BM Amersfoort, The Netherlands. Fax 00-31-78-6523156; e-mail janke.prins{at}hetnet.nl.
| Abstract |
|---|
|
|
|---|
1-Antitrypsin (
1AT) deficiency predisposes individuals to chronic obstructive pulmonary disease (COPD) and/or liver disease. Phenotyping of the protein by isoelectric focusing is often used to characterize
1AT deficiency, but this method may lead to misdiagnosis (e.g., by missing null alleles). We evaluated a workup that included direct sequencing of the relevant parts of the gene encoding
1AT, SERPINA1 [serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1], for patients with
1AT concentrations
1.0 g/L.
Methods: During a 5-year period, we identified 66 patients with
1AT concentrations
1.0 g/L and amplified and sequenced exons 2, 3, and 5 of the
1AT gene in these patients. To ensure that no relevant genotypes were missed, we sequenced the same exons in 48 individuals with
1AT concentrations between 1.0 and 1.5 g/L.
Results: Sequence analysis revealed 18 patients with combinations of disease-associated
1AT alleles: 8 homozygous for the deficient Z allele and 10 compound heterozygotes for various deficient or null alleles. We identified and named 2 new null alleles, Q0soest (Thr102
delA, which produces a TGA stop signal at codon 112) and Q0amersfoort (Tyr160
stop). No relevant disease-associated allele combinations were missed at a 1.0-g/L threshold.
Conclusions: Up to 22% of the alleles in disease-associated
1AT allele combinations may be missed by conventional methods. Genotyping by direct sequencing of samples from patients with
1AT concentrations
1.0 g/L detected these alleles and identified 2 new null alleles.
| Introduction |
|---|
|
|
|---|
1-Antitrypsin (
1AT)1
deficiency is a clinically underrecognized genetic disorder that confers a predisposition for chronic obstructive pulmonary disease (COPD) and/or liver disease (1)(2)(3)(4). In general, serum reference values for
1AT concentration range from 1.0 g/L to 3.5 g/L, and a clinical threshold of 0.8–1.0 g/L is often used for diagnostic purposes (4)(5)(6). Genetic epidemiologic surveys suggest that
1AT deficiency may affect 1 in about 1 500 individuals in Europe (7). Worldwide,
1AT deficiency has been estimated to affect 3.4 million individuals, in all racial subgroups (7).
1AT, an acute-phase protein of 394 amino acid residues (52 kDa), is produced in the liver and reaches the lungs by diffusion from the circulation;
1AT is also produced locally in macrophages and bronchial epithelial cells (8). The SERPINA12
[serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1] gene is located on the long arm of chromosome 14 (14q31–32.3). It spans 12.2 kb and is organized into 4 coding exons (2, 3, 4, and 5) and 3 noncoding exons (1a, 1b, and 1c). At least 100 different
1AT alleles have been identified (5). The alleles associated with disease can be classified according to their pathogenic mechanism. The deficient alleles lead to intracellular
1AT accumulation (e.g., the Z allele, Glu342
Lys) or degradation, such that less (e.g., the S allele, Glu264
Val) or none (e.g., the Mheerlen allele, Pro369
Leu) of the protein is released into the circulation (4)(5)(6)(7)(8)(9). The null alleles are characterized by an absence of circulating
1AT protein because of transcriptional or translational errors that interrupt synthesis. The dysfunctional alleles have an abnormal
1AT function, characterized by a decreased or absent elastase-inhibitory activity. Some deficient
1AT variants, such as the Z and Mmalton alleles, not only produce a predisposition to lung emphysema but also cause the development of liver cirrhosis, owing to pathologic polymerization of these specific
1AT variants within the endoplasmic reticulum of hepatocytes (3)(8)(9)(10).
Although
1AT deficiency is one of the most prevalent and potentially lethal hereditary disorders, it is underrecognized by clinicians and is seldom diagnosed before clinical symptoms, such as COPD, have developed.
1AT concentrations in individuals with the 2 most frequently occurring deficiency genotypes (ZZ homozygotes and SZ compound heterozygotes) are thought to be insufficient to ensure lifetime protection of the lungs from proteolytic damage by elastase, especially in smokers (11). Besides the SZ and ZZ allele combinations, a combination of other deficient, dysfunctional, or null alleles at the
1AT locus may lead to a deficient
1AT concentration (4)(5)(6). The evidence suggests that only 0.41% and 0.35% of ZZ homozygotes and SZ compound heterozygotes, respectively have been recognized (12). In addition, surveys have revealed that the mean (SD) intervals between the first symptoms and the initial diagnosis of
1AT deficiency range from 5.6 (8.5) years to 8.3 (6.9) years (13)(14). Because a delay in the diagnosis of
1AT deficiency also delays opportunities for specific counseling and therapy, efforts to enhance clinicians diagnostic recognition of the disorder are expanding worldwide. The guidelines of the American Thoracic Society and the European Respiratory Society therefore strongly recommend testing for
1AT deficiency in all individuals with COPD, asthma, or emphysema and advise testing for family members of
1AT-deficient patients (15).
In many laboratories, a diagnostic workup currently implies quantification of
1AT antigen and subsequent phenotyping of the
1AT protein by isoelectric focusing (IEF) when concentrations are low or when a pedigree analysis is needed to clarify familial patterns. This approach inevitably leads to misdiagnoses; for instance, clinically relevant null alleles will be missed with IEF (16)(17). Therefore, we replaced IEF with direct sequencing of the relevant parts of the
1AT gene, enabling an efficient and reliable approach to a risk inventory of the index patient and their family.
| Patients and Methods |
|---|
|
|
|---|
1AT antigen concentration in serum are made mainly by pulmonologists and pediatricians, and occasionally by family physicians. In addition, some
1AT-deficient patients are identified by protein electrophoresis. We measured the concentration of
1AT antigen by immunonephelometry with a BN ProSpec Nephelometer calibrated with Certified Reference Material 470 (Dade Behring; CV, 1.6% at 0.9 g/L). Whenever we measured an
1AT concentration below the lower limit of our reference interval (1.0–3.5 g/L) during the last 5 years, we consulted the requesting physician to determine whether further characterization of the finding by genotyping was indicated. This workup was approved by the hospital ethics committee. In addition, we genotyped 48 consecutive patients with
1AT concentrations between 1.0 g/L and 1.5 g/L to confirm that we had missed no relevant disease-associated combinations of
1AT alleles at the 1.0-g/L threshold.
1AT genotyping: dna isolation and sequence analysis
After obtaining informed consent, we isolated the patients DNA and amplified and sequenced relevant parts of the
1AT gene (exons 2, 3, and 5) (see Table 1
for primer sequences). When requested, we also sequenced the
1AT gene in family members of index patients who had been demonstrated to be homozygotes or compound heterozygotes for disease-associated alleles. We used the Gentra Puregene Blood reagent set (Qiagen) according to the manufacturers instructions to prepare genomic DNA from EDTA-anticoagulated whole blood. A detailed description of the applied procedure can be obtained at: www1.qiagen.com/HB/GentraPuregene.
|
All amplification and sequencing primers were obtained from Applied Biosystems. PCR reaction mixtures contained 10 mmol/L Tris-HCl (pH 8.3), 50 mmol/L KCl, 1.5 mmol/L MgCl2, 0.2 mmol/L of each deoxynucleoside triphosphate, 0.3 mol/L of each primer, and 2.5 U of AmpliTaq Gold DNA Polymerase (Applied Biosystems). The thermocycling program included a 10 min incubation at 95 °C, followed by 35 cycles of 94 °C for 30 s, 56 °C for 30 s, and 72 °C for 30 s, finished by 72 °C for 10 min. After PCR, samples were purified using the QlAquick columns (Qiagen) according to the PCR purification protocol provided. Purified PCR products were sequenced with an ABI Prism 310 Genetic Analyzer (Applied Biosystems) using the Bigdye Terminator Cycle Sequencing v1.1 reagent set (Applied Biosystems) according to the manufacturers instructions.
| Results |
|---|
|
|
|---|
1AT quantifications) who had an
1AT concentration
1.0 g/L. Because the
1AT gene is organized into 4 coding exons and 3 noncoding exons, as described above, and because the polymorphisms that affect
1AT concentration and/or function are mainly located in coding exons 2, 3, and 5, we decided to sequence these exons first. We planned to sequence the rest of the gene in cases of discrepancies between the observed concentration and the genotype, but we observed no such discrepancies in our study population.
Genotyping by direct sequencing revealed not only the frequently occurring S and Z alleles but also the Mprocida, Mpalermo, M6passau, Mwurzburg, and Mheerlen alleles. We also discovered 2 previously undescribed null alleles (Tables 2
and 3
), which we named Q0soest and Q0amersfoort, after the residences of the respective index patients. To confirm that we had missed no clinically relevant allele combinations by applying the 1.0-g/L threshold, we selected 48 individuals with
1AT concentrations between 1.0 g/L and 1.5 g/L and sequenced
1AT gene exons 2, 3, and 5. The results confirmed that we had missed no disease-associated allele combinations by placing the threshold at 1.0 g/L (Table 2
). The various alleles observed in this study and their respective effects on the concentration and/or function of the
1AT protein concentration are listed in Table 3
(4)(5)(6)(11)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28).
|
|
Table 2
demonstrates that the wild-type MM genotype and the MS genotype (i.e., heterozygous for the S allele, which produces a minor
1AT deficiency) rarely exhibit
1AT concentrations
1.0 g/L (the lowest observed concentrations were 0.92 g/L and 0.86 g/L, respectively). This result is consistent with the fact that the MS genotype has not been associated with an increased risk for lung disease. The combination of the M allele with any deficient or null allele (Z, Mprocida, Mpalermo, Mheerlen, and the newly identified Q0soest and Q0amersfoort alleles) produces mild to intermediate
1AT deficiencies, with concentrations of 0.64–1.3 g/L. Such allele combinations may have importance, however, because Dahl et al. (29) has demonstrated that the presence of an intermediate
1AT deficiency will not affect lung function in the average individual but may produce marked aggravation of airway obstruction in individuals prone to develop COPD. The combination of the S allele with a deficient allele (Z, Mheerlen, or Mwurzburg) produces intermediate
1AT deficiencies with concentrations of 0.52–0.85 g/L.
1AT concentrations in SZ compound heterozygotes are known to be insufficient to ensure lifetime protection of the lungs from proteolytic damage by elastase, especially in smokers (11). Deficient
1AT concentrations (<0.05 g/L–0.78 g/L) are observed in both ZZ homozygotes and compound heterozygotes for other deficient or null alleles (ZMheerlen, ZQ0soest, and MheerlenQ0amersfoort).
The index patient for the Q0soest allele was a 46-year-old man (JK) who had severe COPD and an
1AT concentration of 0.22 g/L (Fig. 1A
). Sequencing revealed compound heterozygosity for the Z allele and this new null allele, Q0soest. In this allele, the deletion of the A nucleotide in codon 102 (ACC
-CC) causes a 5' frameshift to produce a stop at codon 112 (Thr102
stop112). We suggest that premature termination in exon 2 produces an allele with no detectable mRNA production (21)(22)(23). The Q0soest allele is based on the M1 (Ala213) founder allele. A risk inventory of the family of the index patient revealed that the partner of the index patient had the MM genotype and an
1AT concentration of 1.4 g/L and that their 3 children had the MQ0soest genotype with intermediately deficient
1AT concentrations of 0.58–0.76 g/L, as expected for carriers of a null allele. The mother of the index patient had the MZ genotype and an
1AT concentration of 1.2 g/L, whereas the only sister of the index patient had the MM genotype and an
1AT concentration of 1.5 g/L. These results suggest that the deceased father carried the MQ0soest genotype. Coincidentally, a few months later we encountered a patient who had an
1AT concentration of 0.73 g/L and also appeared to be heterozygous for the Q0soest allele (MQ0soest genotype). As far as we know, the 2 index patients are not related.
|
The index patient for the Q0amersfoort allele was a 47-year-old female patient (GH-B) who had an
1AT concentration of 0.24 g/L and chronic pulmonary disease (Fig. 1B
). Sequencing revealed compound heterozygosity for the deficient Mheerlen allele and the other new null allele, Q0amersfoort. In this allele, a nonsense mutation creates a stop at codon 160, in the same codon as a mutation previously described for the Q0granite falls allele (21) (Table 3
). As with the Q0granite falls allele, the Q0amersfoort mutation produces a premature termination in exon 2, leading to an allele with no detectable mRNA production (21)(22)(23). A risk inventory of the family of the index patient revealed that the partner of the index patient had the MM genotype (
1AT concentration, 1.4 g/L). Their 3 children carried the MMheerlen genotype, with
1AT concentrations of 0.61–0.69 g/L. The mother of the index patient had the MQ0amersfoort genotype and an
1AT concentration of 0.65 g/L. Of the 2 sisters and 3 brothers of the index patient, 1 sibling had the MQ0amersfoort genotype and an
1AT concentration of 1.0 g/L. Coincidentally, a few months later we encountered an index patient who had an
1AT concentration of <0.05 g/L and also appeared to be a compound heterozygote for the MheerlenQ0amersfoort genotype. Again, the 2 index patients do not appear to be related.
| Discussion |
|---|
|
|
|---|
1AT deficiency is well known to reduce the protection of lung tissue from the activity of neutrophilic elastase, thereby leading to progressive destruction of lung tissue and finally to overt COPD (3)(5)(31)(32). Besides smoking, the environmental risk factors for individuals with
1AT deficiency include passive exposure to tobacco smoke, especially as a child, and exposure to mineral dust (33)(34).
Clinically relevant
1AT deficiency is often caused by homozygous inheritance of the
1AT Z allele, but
1AT deficiency can also be due to a combination of other deficient, dysfunctional, or null alleles at the
1AT locus. For instance,
1AT concentrations in SZ compound heterozygotes are thought to be insufficient to ensure lifetime protection of the lungs from proteolytic damage, especially in smokers (1)(4). In addition, Dahl et al. (29) demonstrated that in the larger population even the MZ genotype (associated with intermediately deficient
1AT concentrations) is associated with reduced pulmonary functions in individuals with clinically established COPD. Risk ratios for COPD range from 1.5- to 12-fold, depending on whether the deficient allele is present in heterozygous or homozygous allele combinations (29)(32).
The
1AT protein is a positive acute-phase responder, and we recommend that clinicians keep this fact in mind when setting a threshold for additional analyses to identify patients at risk. In our study, the highest concentration we observed in patients with disease-associated allele combinations was 0.78 g/L (observed in both a ZZ homozygote and an SMheerlen compound heterozygote). On the basis of the data in Table 2
, we suggest a threshold of 0.8 g/L (instead of the 1.0-g/L threshold in this study) for detecting all patients at risk, although additional analyses (either phenotyping or genotyping) are generally carried out whenever the
1AT protein concentration is below the lower limit of the reference interval (4)(15)(16).
In the present study, our sequence analysis of relevant parts of the
1AT gene in patients with
1AT concentrations below the lower limit of the reference interval (
1.0 g/L) revealed various deficient and null alleles at the
1AT locus besides the frequently observed S and Z alleles. We identified 18 patients who were homozygotes or compound heterozygotes for alleles that produce a deficient
1AT concentration. The frequencies of these deficient alleles among the 36 alleles of these 18 patients were as follows: Z, 61%; S, 17%; Mheerlen, 11%; Q0amersfoort, 6%; Mwurzburg, 3%; and Q0soest, 3%. The gold standard for the identification of
1AT variants, according to the American Thoracic Society and the European Respiratory Society (15), is the phenotyping of serum samples by IEF on thin-layer gels in a pH gradient (pH 4–5). Phenotyping is technically challenging because of the complex microheterogeneity of the
1AT protein, and severely deficient alleles and null alleles cannot be identified by this approach (16)(17). In our study, up to 22% of the alleles that we found in disease-associated allele combinations (namely, Mheerlen, Q0amersfoort, Mwurzburg, and Q0soest alleles) would have been missed with IEF-based methods (16)(17). In addition, many commercially available or laboratory-specific methods for
1AT genotyping focus solely on the more prevalent S and Z alleles (16)(35). These S/Z-genotyping assays require additional analysis (phenotyping or sequencing) when the patient with the observed genotype does not exhibit the expected serum
1AT concentration. Such results prompt the question of what threshold to apply in such cases. Snyder et al. (16) suggest expected concentrations of >1.0 g/L for non-S/non-Z genotypes, >0.7 g/L for Z/non-S and S/non-Z genotypes, <1.0 g/L for the SS genotype, and <0.7 g/L for the ZZ genotype. The application of these thresholds to our group of 66 index patients would have required additional analysis for 20 (30%) of these patients. In addition, 2 of the patients (with the SMheerlen and SMwurzburg genotypes) would have been incorrectly typed as having the MS genotype. Of the 20 samples requiring additional analysis, 11 samples would have yielded interpretation problems with phenotyping because of the presence of severely deficient or null alleles, such as Mheerlen, Mwurzburg, Q0soest, and/or Q0amersfoort (17)(27)(28). Given the availability of DNA-sequencing equipment in many hospitals, we therefore advocate direct sequencing of the relevant parts of the
1AT gene for patients with suspected
1AT deficiency. Direct DNA sequencing offers the advantage of clear-cut and efficient detection of any alteration in the wild-type sequence, including null alleles and the identification of previously unknown alleles. This approach not only leads to the unambiguous identification of combinations of alleles responsible for
1AT deficiency in index patients but also offers the possibility of a risk inventory for their families. This capability is beneficial because recent studies have demonstrated that significant positive effects are associated with the early identification of
1AT-deficient individuals (36). Importantly, such effects would include an increased willingness to reduce exposure to environmental risk factors, such as cigarette smoking.
In conclusion,
1AT deficiency is a common genetic disorder that predisposes an affected individual to COPD and/or liver disease. In many laboratories, a characterization of
1AT deficiency implies the phenotyping of the protein by IEF and/or S/Z genotyping. Such an approach may lead to misdiagnosis because some alleles (e.g., null alleles) will be missed. We advocate a workup that includes genotyping by direct sequencing of the coding regions of the
1AT gene for patients who have
1AT concentrations
1.0 g/L. Application of this approach means that no relevant disease-associated allele combinations will be missed. Finally, with this approach we have identified 2 previously unidentified null alleles, Q0soest and Q0amersfoort, that produce
1AT deficiency.
| Acknowledgments |
|---|
Financial Disclosures: None declared.
| Footnotes |
|---|
1AT,
1-antitrypsin; COPD, chronic obstructive pulmonary disease; and IEF, isoelectric focusing.
2 Human genes: SERPINA1, serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1. ![]()
| References |
|---|
|
|
|---|
1-antitrypsin deficiency. Acta Med Scand 1978;204:345-351.[Web of Science][Medline]
[Order article via Infotrieve]
1-antitrypsin accumulation in the liver. Nature 1992;357:605-607.[CrossRef][Medline]
[Order article via Infotrieve]
1-Antitrypsin deficiency. Lancet 2005;365:2225-2236.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
1-Antitrypsin deficiency. 2: genetic aspects of
1-antitrypsin deficiency: phenotypes and genetic modifiers of emphysema risk. Thorax 2004;59:259-264.
1-antitrypsin deficiency. Chest 2002;122:1818-1829.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
1-Antitrypsin polymerization and the serpinopathies: pathobiology and prospects for therapy. J Clin Invest 2002;110:1585-1590.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
1-Antitrypsin deficiency. 4: molecular pathophysiology. Thorax 2004;59:529-535.
1-antitrypsin deficiency: an aggregated protein induces mitochondrial injury. J Clin Invest 2002;110:1579-1583.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
1-antitrypsin deficiency.
1-Antitrypsin Deficiency Registry Study Group. Am J Respir Crit Care Med 1996;154:1718-1725.[Abstract]
1-Antitrypsin deficiency. 1: epidemiology of
1-antitrypsin deficiency. Thorax 2004;59:164-169.
1-antitrypsin deficiency and augmentation therapy [abstract]. Am J Respir Crit Care Med 2004;169:A517.
1-antitrypsin deficiency: a continuing problem. Chest 2005;128:1989-1994.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
1-antitrypsin deficiency [in German]. Pneumologie 2005;59:36-68.[CrossRef][Medline]
[Order article via Infotrieve]
1-antitrypsin deficiency: an algorithm of quantification, genotyping and phenotyping. Clin Chem 2006;52:2236-2242.
1-Antitrypsin deficiency as a result of compound heterozygosity for the Z and MHeerlen alleles. Clin Chem 2001;47:978-979.
1-antitrypsin deficiency allele Mprocida. J Biol Chem 1988;263:15528-15534.
1-antitrypsin variants, including two PI*Q0 alleles and one deficient PI*M allele. Am J Hum Genet 1994;55:1113-1121.[Web of Science][Medline]
[Order article via Infotrieve]
1-antitrypsin. Am J Hum Genet 1989;44:894-902.[Web of Science][Medline]
[Order article via Infotrieve]
1-Antitrypsin NullGranite Falls, a nonexpressing
1-antitrypsin gene associated with a frameshift to stop mutation in a coding exon. J Biol Chem 1987;262:11999-12004.
1-antitrypsin null alleles. Resp Med 2000;94(Suppl C):S7-S11.
1-antitrypsin and the gene for the S variant. Biochemistry 1984;23:4828-4837.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
1-antitrypsin deficiency associated with the common S-type (Glu264Val) mutation results from intracellular degradation of
1-antitrypsin prior to secretion. J Biol Chem 1989;264:10477-10486.
1-antitrypsin deficiency variant PI MHeerlen. Hum Genet 1989;81:264-268.[Web of Science][Medline]
[Order article via Infotrieve]
1-antitrypsin alleles PI Mwürzburg (Pro369Ser), Mheerlen (Pro369Leu), and Q0lisbon (Thr68Ile). Eur J Hum Genet 1999;7:321-331.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
1-antitrypsin deficiency: MZ individuals with chronic obstructive pulmonary disease may have lower lung function than MM individuals. Clin Chem 2001;47:56-62.
1-antitrypsin deficiency: the promise of early diagnosis and intervention. Clin Chem 2006;52:2180-2181.
1-Antitrypsin deficiency. Thorax 2004;59:92-93.
1-antitrypsin heterozygote schoolchildren. Thorax 2003;58:237-241.
1-antitrypsin deficiency. Am J Respir Crit Care Med 2000;162:553-558.
1-antitrypsin genotyping by multiplex real-time fluorescence PCR with the LightCycler. Clin Chem 2000;46:156-161.
1-antitrypsin deficiency. Genet Med 2004;6:204-210.[Web of Science][Medline]
[Order article via Infotrieve]The following articles in journals at HighWire Press have cited this article:
![]() |
M. L. Snyder and C. R. Fantz Cirrhosis Originally Diagnosed as Nonalcoholic Steatohepatitis Clin. Chem., August 1, 2008; 54(8): 1395 - 1398. [Full Text] [PDF] |
||||
![]() |
M. Dahl Commentary Clin. Chem., August 1, 2008; 54(8): 1398 - 1399. [Full Text] [PDF] |
||||
![]() |
M. R. Snyder and W. E. Highsmith Jr Commentary Clin. Chem., August 1, 2008; 54(8): 1399 - 1399. [Full Text] [PDF] |
||||
![]() |
M. Zorzetto, E. Russi, O. Senn, M. Imboden, I. Ferrarotti, C. Tinelli, I. Campo, S. Ottaviani, R. Scabini, A. von Eckardstein, et al. SERPINA1 Gene Variants in Individuals from the General Population with Reduced {alpha}1-Antitrypsin Concentrations Clin. Chem., August 1, 2008; 54(8): 1331 - 1338. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |