Background:Systemic lupus erythematosus (SLE) is caused by a combination of environmental and genetic factors; studying the association between regulatory genes and this disease may determine the genetic causes of interfering with SLE. In different populations, studies have shown that the tumor necrosis factor α (TNF-α) gene (as a candidate gene) can contribute to the formation and progression of lupus disease.
Objectives:This study aimed to indicate the possible association between the increased rate of SLE hazard and 2 single-nucleotide polymorphisms (SNPs) of rs1800629 and rs1800630 genetic polymorphisms in the TNF-α promoter gene in the Lor population.
Methods:According to the American College of Rheumatology (ACR) criteria, 120 unrelated SLE patients and 120 healthy controls with no family or personal history of autoimmune diseases were selected. DNA was genotyped for the TNF-α promoter (-308 G/A and -863 C/A) by the tetra-primer amplification-refractory mutation system (tetra-primer ARMS)–polymerase chain reaction (PCR) method.
Results:The frequency difference between allele A (mutant allele) and allele C (normal allele) at position -863 of the TNF-α promoter gene (odds ratio [OR] = 3.426; 95% CI, 1.985 - 5.914) was notably higher in SLE patients than in control subjects. Also, a significant relation was obtained among the rs1800830 AA genotype and increased risk of SLE (OR = 4.489; 95% CI, 2.464 - 8.177; P < 0.0001). Our results for rs1800629 at position -308 were not remarkably different.
Conclusions:We found a significant correlation between allelic and genotype frequencies between rs1800830 (-863 C/A) TNF-α SNP and SLE in our study. However, no significant correlation was observed between the rs1800629 (-308 G/A) TNF-α promoter and the increase of SLE hazard in the Lor population. No remarkable association was obtained between TNF-α gene rs1800629 (-308 G/A) and rs1800630 (-863 C/A) SNPs and anti-double-stranded DNA (anti-dsDNA) or antinuclear antibody (ANA), which are some of the symptoms of SLE.
Systemic lupus erythematosus (SLE) is one of the complex diseases with a wide range of clinical criteria, including antinuclear antibody (ANA) and anti-double-stranded DNA (anti-dsDNA) antibodies (1-3). The incidence of SLE disease varies between different populations. The cause of SLE disease is unknown. Environmental and genetic factors are among the risk factors for SLE. The prevalence of SLE is high in Asians and notably in Iranian populations (4, 5). Recently, many efforts have been made to find a link between SLE susceptibility and genetic variants (6-9). The tumor necrosis factor α (TNF-α) gene produces an inducible pro-inflammatory cytokine (10, 11), which is fixed in human chromosome 6 within the major histocompatibility complex (MHC) class III region (12). It seems that the TNF-α gene is connected to the pathogenesis of inflammatory disorders (13, 14). Several studies on different diseases have shown the effect of single-nucleotide polymorphisms (SNPs) on the TNF-α promoter region (15, 16). However, studies have not definitely determined the association between TNF-α promoter gene SNPs and SLE (17-20). Rs18008629 at position -308 G/A polymorphism is related to increased potential and intensity in a variety of autoimmune diseases (19, 21-23). The second polymorphism, a common functional polymorphism is located at position -863 C/A (24). Several studies have analyzed the association between rs1800630 at the position -863 C/A TNF-α promoter gene polymorphism and inflammatory disorders such as SLE (18, 25-28). However, different populations have shown different results. No studies have been performed to study the association with rs18008629 at position -308 G/A and also rs1800630 at position -863 C/A TNF-α gene polymorphisms in the case of SLEin the Iranian Lor Population.
We studied the association between rs1800629 TNF-α at position -308 G/A and rs1800630 TNF-α at position -863 C/A promoter polymorphisms and the susceptibility of SLE hazard.
According to the American College of Rheumatology (ACR) criteria, 120 unrelated SLE patients were selected. Also, 120 healthy individuals with no personal and family history of autoimmune disease were selected as controls. Both serological factors (ANA autoantibodies and anti-dsDNA) associated with the disease were also determined by diagnostic tests in patients, such as ANA autoantibodies (n = 114) and anti-dsDNA (n = 116). For this study, we selected all cases and controls from the Iranian Lor population.
Genomic DNA was isolated from peripheral blood leukocytes using the salting-out method (29). Briefly, 500 µL of blood was transferred to 1.5-µL microfuge tubes, and 1-mL cold water was added. The solutions were gently mixed and centrifuged at 13 000g for 1 minute at room temperature. Then, the supernatant was discarded. The procedure was repeated twice. Next, 300 µL TES buffer (pH = 7.5; NaCl [150mM], Tris-base [10mM], EDTA [10mM]), 320 µL SDS 10%, and 25 µL proteinase K (CinnaGen, Iran) were added, and the mixture was incubated at 37°C for 2 hours. Then, 220 µL of saturated NaCl was added with gentle mixing, and the mixture was centrifuged at 13 000g for 15 minutes. The supernatant was transferred to a new microfuge tube, where 550 µL of cold isopropanol was added and centrifuged at 13000 g for 2 minutes. The supernatant was discarded, and 1 mL of cold ethanol 70% was added. The suspension was gently mixed and centrifuged at 13000 g for 1 minute. Finally, pellets were dried before dissolving in 50 µL of TE buffer (Tris base [10mM], EDTA [1mM]) and preserved at -20°C. Recognition of TNF-α -308 and -863 polymorphisms was performed by the tetra-primer amplification-refractory mutation system (tetra-primer ARMS)–polymerase chain reaction (PCR) method. Four primers were tested, of which 2 inner primers (ie, inner forward and reverse) and 2 outer primers (ie, the outer forward and reverse) were the same primers. All 4 primers were designed using the NCBI bioinformatics database and then blasted (Table 1).
|Name||Primer Sequence||Tm||PCR Product|
|Outer forward||5′- GGACCCAAACACAGGCCTCAG -3′||60.2||323 bp|
|Outer reverse||5′- TCCTCCCTGCTCCGATTCC-3′||61.2|
|Inner forward||5′- GGCAATAGGTTTTGAGGGCGAGGG-3′||62.2||217 bp|
|Inner reverse||5′- GGAGGCTGAACCCCGTACT-3′||62.1||106 bp|
|Outer forward||5′-GGCTCTGAGGAATGGGTTAC-3′||57.67||200 bp|
|Inner forward||5′-TCGAGTATGGGGACCCCCA-3′||60.46||121 bp|
|Inner reverse||5′- ATGGCCCTGTCTTCGTTAAGG-3′||62.5||158 bp|
As a positive control, 323-base pairs (bp) and 200-bp constant DNA fragments (for TNF-α -308 and -863 polymorphisms, respectively) were amplified with outer primers. For the TNF-α -308 polymorphism, the thermal cycle of the test was as follows 94°C(4 minuts) (primary denaturation), annealing (for 30 cycles): 94°C (denaturation), 56°C(annealing), and 72°C (30 s each) (extension) and final elongation 72°C (7 minuts). For the TNF-α -863 polymorphism, the thermal cycling condition was followed by primary denaturation at 94°C (4 minutes), annealing (for 30 cycles): 95°C (denaturation), 60°C (annealing), and 72°C (30 s each) (extension) and final elongation 72°C (7minuts). Then, the products were visualized using electrophoresis in 2% and 2.5% agarose gels (TNF-α -308 and -863 polymorphisms, respectively) and stained with DNA safe stain (CinnaGen, Iran; Figure 1).
3.3. Statistical Analysis
The distribution of rs1800629 and rs1800630 genotypes was checked to analyze the deviation from Hardy-Weinberg equilibrium in SLE cases and controls using the chi-square test. Using chi-square and logistic regression tests, allelic and genotypic dispensation between the patients and healthy controls was analyzed. P-values less than 0.05 were considered statistically significant. The odds ratio (OR) and 95% CIs were also evaluated. Besides these 2 polymorphisms, the eventual correlation with 2 clinical manifestations was examined by the chi-square test. Statistical analysis of the data was performed using SPSS version 24 (SPSS Inc, Chicago, Ill, USA).
In this study, TNF-α -308 G/A and -863 C/A SNPs were analyzed in 120 healthy controls and 120 cases with SLE. The -308 G/A polymorphism showed a significant departure from Hardy-Weinberg equilibrium among controls and patients in genotype distribution (P < 0.0001). Genotype distribution analysis for -863 C/A polymorphism Hardy-Weinberg equilibrium showed a significant deviation among patients but not controls (P < 0.0001 and P > 0.05, respectively).
4.1. Characteristics of Cases and Controls
The chi-square test (χ2 test) showed no significant association between gender and disease incidence (P = 0.244). The t-test for both independent samples indicated that age dissimilarity observed between the patient and control groups was statistically significant (P = 0.01). It means that the average age was higher in the control group than in the patient group. However, since this disease often occurs at an early age, it is concluded that our control subjects are perfectly matched for differentiation with our patients.
4.2. Allele and Genotype Frequencies of TNF-α Genetic Polymorphisms
For both TNF-α genetic polymorphisms, allele and genotype frequencies were calculated (Table 2). The -308 G/A SNP (rs1800629) allele frequency was not remarkably different. The frequency analysis of the A allele at position -863 of the TNF-α gene was remarkably higher in SLE cases than in healthy controls (OR = 3.426; 95% CI, 1.985 - 5.914). Accordingly, it is concluded that the AA genotype is correlated with an increased hazard of SLE disease (OR = 4.489; 95% CI, 2.464 - 8.177; P < 0.0001).
|Gene Name SNP Database ID (Cucleotide Change)||SLE||Controls||P-Value||Odds Ratio (95% CI)|
|rs1800629 (-308G>A)||0.341||1.322 (0.744 - 2.347)|
|A/A||30 (0.25)||35 (31.7)|
|G/A||90 (0.75)||85 (68.3)|
|Alleles||1.178 (0.810 - 1.712)|
|A||150 (62.5)||158 (65.83)||0.391|
|G||90 (37.5)||82 (34.167)|
|rs1800630 (-863C>A)||< 0.0001||4.489 (2.464 - 8.177)|
|C/A||57 (47.5)||19 (15.84)|
|Alleles||3.426 (1.985 - 5.914)|
|A||183 (91.67)||220 (76.25)||< 0.0001|
|C||57 (7.916)||19 (23.75)|
4.3. TNF-α Genetic Polymorphisms and Clinical Features of SLE
The possible association between rs1800629 and rs1800630 TNF-α SNPs and both clinical features of SLE patients (ie, ANA [n = 114] and anti-dsDNA antibodies [n = 116]) were analyzed (Table 3). No statistically significant correlation was observed between these 2 polymorphisms and the clinical features of SLE patients.
|Serological Features||Rs1800629||Serological Features||Rs1800630|
|No. (%)||P-Value||No. (%)||P-Value|
|AA (%)||100 (43.5)||A (%)||87 (38.16)|
|G (%)||14 (6.141)||C (%)||27 (11.48)|
|AA (%)||104 (45.61)||AA (%)||30 (26.32)|
|GA (%)||14 (24.57)||CA (%)||27 (23.68)|
|AA (%)||47 (41.22)||AA (%)||31 (27.19)|
|GA (%)||10 (8.78)||CA (%)||26 (22.8)|
|A (%)||99 (42.67)||A (%)||85 (36.64)|
|G (%)||13 (5.61)||C (%)||27 (11.64)|
|A (%)||109 (46.98)||A (%)||94 (40.52)|
|G (%)||11 (4.75)||C (%)||26 (11.20)|
|AA (%)||43 (37.06)||AA (%)||29 (25)|
|GA (%)||13 (11.20)||CA (%)||27 (23.28)|
|AA (%)||49 (42.25)||AA (%)|
|GA (%)||11 (9.48)||CA (%)|
In different ethnic groups, studies have suggested a correlation between TNF-α polymorphisms and SLE risk (30, 31). However, the effect of the TNF-α polymorphism on the ability of SLE disease is yet unclear. Some studies have shown that the TNF-α -308 A allele has a significant transcriptional effect, but others have claimed that this polymorphism has no effects on TNF-α function (31-35). The results of this study confirmed the association between the TNF-α -308 G/A allele and SLE, as it has been found in most populations, including Taiwanese patients (25). The current study indicated that none of the genotype and allele frequencies of the re1800629 polymorphism at position -308 G/A were remarkably associated with Lor SLE patients compared to controls. Generally, the results of association studies on different populations between re1800629 and rs1800630 and susceptibility of SLE risk are different. In Caucasian SLE patients, Tsuchiya et al showed that -863A, -308G haplotypes were associated with disease intensity (36), whereas McHugh et al showed no significant association between -863A, -308G and disease risk (37). Our results suggested that allele and genotype frequencies of rs1800630 were significantly associated with SLE in the Iranian Lor population. Functional analysis of the rs1800630 polymorphism in the promoter domain of TNF-α at position -863 showed contradictory results. Although the association between TNF-α gene polymorphisms and SLE is ambiguous in different ethnic histories, the -863 C allele may play a role in susceptibility to SLE in the Lor population, partially through their higher promoter occupation of TNF-α production.
Our statistical analysis showed no significant relationship between anti-dsDNA and ANA with the type of the genotype. Consequently, further studies are required in this respect between various Iranian populations and the increased potential of SLE hazard.
This study showed that TNF-α -863 SNP was associated with SLE in the examined patients. No significant association was observed between clinical features of SLE patients and these TNF-α promoter gene polymorphisms. This indicates that further studies with larger sample sizes on different populations are needed to find the exact role of this gene.
Petri M, Orbai AM, Alarcon GS, Gordon C, Merrill JT, Fortin PR, et al. Derivation and validation of the Systemic Lupus International Collaborating Clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum. 2012;64(8):2677-86. doi: 10.1002/art.34473. [PubMed: 22553077]. [PubMed Central: PMC3409311].
Ejtehadi A, Roghanian R, Bonakdar SZ. Association of Helicobacter pylori Infection with Systemic Lupus Erythematosus. J Clin Cell Immunol. 2017;8(6):2. doi: 10.4172/2155-9899.1000528.
Shahrokhi SZ, Kazemi Nezhad SR, Baharvand Ahmadi S, Akhoond MR. Association Study of the PTPN22 Gene Polymorphisms with Systemic Lupus Erythematosus in Lorestan Province of Iran. Gene Cell Tissue. 2017;In Press(In Press). doi: 10.5812/gct.12023.
Fourati H. Genetic Factors Contributing to Systemic Lupus Erythematosus in Tunisian Patients. J Clin Cell Immunol. 2012;3(4). doi: 10.4172/2155-9899.1000129.
Lin YJ, Wan L, Sheu JJ, Huang CM, Lin CW, Lan YC, et al. G/T polymorphism in the interleukin-2 exon 1 region among Han Chinese systemic lupus erythematosus patients in Taiwan. Clin Immunol. 2008;129(1):36-9. doi: 10.1016/j.clim.2008.05.011. [PubMed: 18650128].
Lin YJ, Wan L, Lee CC, Huang CM, Tsai Y, Tsai CH, et al. Disease association of the interleukin-18 promoter polymorphisms in Taiwan Chinese systemic lupus erythematosus patients. Genes Immun. 2007;8(4):302-7. doi: 10.1038/sj.gene.6364387. [PubMed: 17361200].
Lin YJ, Wan L, Sheu JJ, Huang CM, Lin CW, Lan YC, et al. A/C polymorphism in the interleukin-18 coding region among Taiwanese systemic lupus erythematosus patients. Lupus. 2008;17(2):124-7. doi: 10.1177/0961203307086031. [PubMed: 18250135].
Cunninghame Graham DS, Graham RR, Manku H, Wong AK, Whittaker JC, Gaffney PM, et al. Polymorphism at the TNF superfamily gene TNFSF4 confers susceptibility to systemic lupus erythematosus. Nat Genet. 2008;40(1):83-9. doi: 10.1038/ng.2007.47. [PubMed: 18059267]. [PubMed Central: PMC3705866].
Ruddle N. Tumor necrosis factor (TNF-α) and lymphotoxin (TNF-β). Curr Opin Immunol. 1992;4(3):327-32. doi: 10.1016/0952-7915(92)90084-r.
Illei GG, Tackey E, Lapteva L, Lipsky PE. Biomarkers in systemic lupus erythematosus. I. General overview of biomarkers and their applicability. Arthritis Rheum. 2004;50(6):1709-20. doi: 10.1002/art.20344. [PubMed: 15188346].
Spies T, Morton CC, Nedospasov SA, Fiers W, Pious D, Strominger JL. Genes for the tumor necrosis factors alpha and beta are linked to the human major histocompatibility complex. Proc Natl Acad Sci USA. 1986;83(22):8699-702. doi: 10.1073/pnas.83.22.8699. [PubMed: 3464978]. [PubMed Central: PMC386998].
Ramos PS, Shaftman SR, Ward RC, Langefeld CD. Genes associated with SLE are targets of recent positive selection. Autoimmune Dis. 2014;2014:203435. doi: 10.1155/2014/203435. [PubMed: 24587899]. [PubMed Central: PMC3920976].
Hohaus S, Giachelia M, Di Febo A, Martini M, Massini G, Vannata B, et al. Polymorphism in cytokine genes as prognostic markers in Hodgkin's lymphoma. Ann Oncol. 2007;18(8):1376-81. doi: 10.1093/annonc/mdm132. [PubMed: 17496310].
Flori L, Delahaye NF, Iraqi FA, Hernandez-Valladares M, Fumoux F, Rihet P. TNF as a malaria candidate gene: polymorphism-screening and family-based association analysis of mild malaria attack and parasitemia in Burkina Faso. Genes Immun. 2005;6(6):472-80. doi: 10.1038/sj.gene.6364231. [PubMed: 15931230].
Rood MJ, Van Krugten MV, Zanelli E, Van Der Linden MW, Keijsers V, Schreuder GMT, et al. TNF-308A and HLA-DR3 alleles contribute independently to susceptibility to systemic lupus erythematosus. Arthritis Rheum. 2000;43(1):129-34. doi: 10.1002/1529-0131(200001)43:1<129::Aid-anr16>3.0.Co;2-s.
Hirankarn N, Avihingsanon Y, Wongpiyabovorn J. Genetic susceptibility to SLE is associated with TNF-alpha gene polymorphism -863, but not -308 and -238, in Thai population. Int J Immunogenet. 2007;34(6):425-30. doi: 10.1111/j.1744-313X.2007.00715.x. [PubMed: 18001298].
Zuniga J, Vargas-Alarcon G, Hernandez-Pacheco G, Portal-Celhay C, Yamamoto-Furusho JK, Granados J. Tumor necrosis factor-alpha promoter polymorphisms in Mexican patients with systemic lupus erythematosus (SLE). Genes Immun. 2001;2(7):363-6. doi: 10.1038/sj.gene.6363793. [PubMed: 11704801].
Cuenca J, Cuchacovich M, Perez C, Ferreira L, Aguirre A, Schiattino I, et al. The -308 polymorphism in the tumour necrosis factor (TNF) gene promoter region and ex vivo lipopolysaccharide-induced TNF expression and cytotoxic activity in Chilean patients with rheumatoid arthritis. Rheumatology (Oxford). 2003;42(2):308-13. doi: 10.1093/rheumatology/keg092. [PubMed: 12595628].
Yang ZC, Xu F, Tang M, Xiong X. Association Between TNF-alpha Promoter -308 A/G Polymorphism and Systemic Lupus Erythematosus Susceptibility: A Case-Control Study and Meta-Analysis. Scand J Immunol. 2017;85(3):197-210. doi: 10.1111/sji.12516. [PubMed: 27943420].
Sadaf T, John P, Bhatti A, Malik JM. Lack of association of -863C/A (rs1800630) polymorphism of tumor necrosis factor-a gene with rheumatoid arthritis. Arch Med Sci. 2019;15(2):531-6. doi: 10.5114/aoms.2018.76946. [PubMed: 30899307]. [PubMed Central: PMC6425217].
Skoog T, van't Hooft FM, Kallin B, Jovinge S, Boquist S, Nilsson J, et al. A common functional polymorphism (C-->A substitution at position -863) in the promoter region of the tumour necrosis factor-alpha (TNF-alpha) gene associated with reduced circulating levels of TNF-alpha. Hum Mol Genet. 1999;8(8):1443-9. doi: 10.1093/hmg/8.8.1443. [PubMed: 10400991].
Lin YJ, Chen RH, Wan L, Sheu JC, Huang CM, Lin CW, et al. Association of TNF-alpha gene polymorphisms with systemic lupus erythematosus in Taiwanese patients. Lupus. 2009;18(11):974-9. doi: 10.1177/0961203309105361. [PubMed: 19762398].
Katkam SK, Rajasekhar L, Tasneem FSD, Kutala VK. Synergetic Interaction of HLA-DRB1*07 Allele and TNF-Alpha - 863 C/A Single Nucleotide Polymorphism in the Susceptibility to Systemic Lupus Erythematosus. Indian J Clin Biochem. 2021;36(1):59-66. doi: 10.1007/s12291-019-00854-9. [PubMed: 33505128]. [PubMed Central: PMC7817716].
Ramirez-Bello J, Cadena-Sandoval D, Mendoza-Rincon JF, Barbosa-Cobos RE, Sanchez-Munoz F, Amezcua-Guerra LM, et al. Tumor necrosis factor gene polymorphisms are associated with systemic lupus erythematosus susceptibility or lupus nephritis in Mexican patients. Immunol Res. 2018;66(3):348-54. doi: 10.1007/s12026-018-8993-8. [PubMed: 29611038].
El-Dakrony AM, El-Gazzar I, Sayed S, Sadek W, El-Fishawy H, Shaker O. Clinical significance of tumor necrosis factor-α-863 C/A and -1031 T/C promoter polymorphisms in systemic lupus erythematosus patients: Relation to disease activity and damage. Egypt Rheumatol. 2017;39(4):223-6. doi: 10.1016/j.ejr.2017.04.008.
Shahvali Koohshori M, Kazemi Nezhad SR, Rajaei E, Akhoond MR. Association of the MTHFR C677T Polymorphism (rs1801133) With Risk of Rheumatoid Arthritis in the Khuzestan Province of Iran. Gene Cell Tissue. 2015;2(4). doi: 10.17795/gct-28421.
Kummee P, Tangkijvanich P, Poovorawan Y, Hirankarn N. Association of HLA-DRB1*13 and TNF-alpha gene polymorphisms with clearance of chronic hepatitis B infection and risk of hepatocellular carcinoma in Thai population. J Viral Hepat. 2007;14(12):841-8. doi: 10.1111/j.1365-2893.2007.00880.x. [PubMed: 18070287].
Jimenez-Morales S, Velazquez-Cruz R, Ramirez-Bello J, Bonilla-Gonzalez E, Romero-Hidalgo S, Escamilla-Guerrero G, et al. Tumor necrosis factor-alpha is a common genetic risk factor for asthma, juvenile rheumatoid arthritis, and systemic lupus erythematosus in a Mexican pediatric population. Hum Immunol. 2009;70(4):251-6. doi: 10.1016/j.humimm.2009.01.027. [PubMed: 19480843].
Wilson AG, Symons JA, McDowell TL, McDevitt HO, Duff GW. Effects of a polymorphism in the human tumor necrosis factor alpha promoter on transcriptional activation. Proc Natl Acad Sci USA. 1997;94(7):3195-9. doi: 10.1073/pnas.94.7.3195. [PubMed: 9096369]. [PubMed Central: PMC20345].
Kroeger KM, Carville KS, Abraham LJ. The −308 tumor necrosis factor-α promoter polymorphism effects transcription. Mol Immunol. 1997;34(5):391-9. doi: 10.1016/s0161-5890(97)00052-7.
He B, Navikas V, Lundahl J, Söderström M, Hillert J. Tumor necrosis factor α-308 alleles in multiple sclerosis and optic neuritis. J Neuroimmunol. 1995;63(2):143-7. doi: 10.1016/0165-5728(95)00138-7.
Bayley JP, de Rooij H, van den Elsen PJ, Huizinga TW, Verweij CL. Functional analysis of linker-scan mutants spanning the -376, -308, -244, and -238 polymorphic sites of the TNF-alpha promoter. Cytokine. 2001;14(6):316-23. doi: 10.1006/cyto.2001.0902. [PubMed: 11497492].
Tsuchiya N, Kawasaki A, Tsao BP, Komata T, Grossman JM, Tokunaga K. Analysis of the association of HLA-DRB1, TNFalpha promoter and TNFR2 (TNFRSF1B) polymorphisms with SLE using transmission disequilibrium test. Genes Immun. 2001;2(6):317-22. doi: 10.1038/sj.gene.6363783. [PubMed: 11607787].
McHugh NJ, Owen P, Cox B, Dunphy J, Welsh K. MHC class II, tumour necrosis factor alpha, and lymphotoxin alpha gene haplotype associations with serological subsets of systemic lupus erythematosus. Ann Rheum Dis. 2006;65(4):488-94. doi: 10.1136/ard.2005.039842. [PubMed: 16107511]. [PubMed Central: PMC1798099].