Conceived and designed the experiments: MCT BT GF. Performed the experiments: MCT BT VN SC AM. Analyzed the data: MCT BT VN SC AM GF. Contributed reagents/materials/analysis tools: VN AM GF. Wrote the paper: MCT BT VN AM GF. Clinical data collection: FF EG SLB SA.
The authors have declared that no competing interests exist.
The
A total of 396 RA cases and 477 controls, all of Italic ancestry, were genotyped for
The
The
Genetic factors are thought to be responsible for up to 50–60% of the rheumatoid arthritis (RA) liability
Although the role of
To date, no data are available regarding the possible association between
The ethical approval for the study was obtained from the Catholic University of the Sacred Heart Ethical Committee. All subjects gave their written informed consent on the analysis of the
Cases were recruited from the Division of Rheumatology of the Catholic University of the Sacred Heart of Rome. Patients fulfilled at least four of the American College of Rheumatology criteria for RA
In order to calculate the sample size, the following parameters were used: power 80%, level of confidence 95%, estimated frequency of
All patients' sera were tested for the presence of anti-CCP, IgM RF (rheumatoid factor) and IgA RF autoantibodies (ELISA method, Axis-Shield Diagnostics, Dundee, UK for anti-CCP and Orgentec diagnostika, Mainz, Germany for IgM and IgA RFs).
Genomic DNA was isolated from whole blood through FlexiGene DNA kit (Qiagen, Valencia, CA) according to the manufacturer instructions.
The
The electronic medical databases used for the search were Pubmed, Embase and the Cochrane Library. In the research, we used the keywords: “arthritis”, “rheumatoid”, “
Data regarding the
Three different meta-analyses were carried out using StatDirect statistical software Version2.7.8. The first one evaluated the association of
Forest-plots graphs were produced in order to estimate the pooled association between the
The studied sample was composed of 396 RA patients and 477 controls (the power of the study was 75% for the cases). The genotype distribution of the
In our center, the analysis of RA patients and controls, all of Italic ancestry, showed a
Genotype | ||||||||
Group | C/C | C/T | T/T | T Allele | HWE ( |
OR (95%CIs) |
Allelic |
|
|
RA patients (n = 313) (mean age: 55.7±13.7) | 276 (91.6%) | 36 (11.5%) | 1 (0.3%) | 38 (6.1%) |
|
|
|
Healthy controls (n = 377) (mean age: 55.5±14.1) | 348 (92.3%) | 28 (7.4%) | 1 (0.3%) | 30 (4.0%) |
|
|||
|
RA patients (n = 83) (mean age: 59.7±12.2) | 76 (91.6%) | 7 (8.4%) | 0 (0.0%) | 7 (4.2%) |
|
|
|
Healthy controls (n = 100) (mean age: 59.4±12.0) | 95 (95.0%) | 5 (5.0%) | 0 (0.0%) | 5 (2.5%) |
|
|||
|
RA patients (n = 396) (mean age: 56.6±13.4) | 352 (88.9%) | 43 (10.9%) | 1 (0.2%) | 45 (5.7%) |
|
|
|
Healthy controls (n = 477) (mean age: 56.3±14.0) | 443 (92.9%) | 33 (6.9%) | 1 (0.2%) | 35 (3.7%) |
|
HWE: Hardy-Weinberg equilibrium.
Odds ratios expressed as carriers of T1858 allele
Fisher's exact test of odds ratios.
The association of the
Study | Ref. | Year | Country | Quality | RA C/T-T/T | Controls C/T-T/T | RA total | Controls total |
Seldin |
16 | 2005 | Finland | 10 | 372 | 406 | 1030 | 1400 |
Plenge |
17 | 2005 | Sweden | 11.5 | 432 | 203 | 1513 | 874 |
Hinks |
18 | 2005 | UK | 11 | 289 | 114 | 886 | 595 |
Steer |
19 | 2005 | UK | 10 | 84 | 62 | 302 | 374 |
Wesoly |
20 | 2005 | Netherlands | 10 | 93 | 155 | 416 | 891 |
Zhernakova |
21 | 2005 | Netherlands | 10 | 42 | 88 | 151 | 528 |
Orozco |
22 | 2005 | Spain | 11.5 | 163 | 146 | 826 | 1036 |
Johansson |
5 | 2006 | Sweden | 12 | 35 | 71 | 89 | 360 |
Harrison |
23 | 2006 | UK | 10 | 179 | 109 | 686 | 566 |
Pierer |
24 | 2006 | Germany | 11.5 | 148 | 67 | 390 | 349 |
Viken |
25 | 2007 | Norway | 10.5 | 264 | 119 | 861 | 557 |
Lie |
26 | 2007 | Norway | 12.5 | 75 | 119 | 221 | 555 |
Kokkonen |
27 | 2007 | Sweden | 12 | 166 | 209 | 504 | 970 |
Majorczyk |
28 | 2007 | Poland | 11 | 61 | 118 | 173 | 543 |
Wesoly |
29 | 2007 | Netherlands | 10.5 | 183 | 55 | 661 | 284 |
Eike |
30 | 2008 | Norway | 11 | 213 | 191 | 686 | 952 |
Farago |
31 | 2008 | Hungary | 11 | 158 | 24 | 399 | 107 |
Starck |
32 | 2009 | Slovakia | 12.5 | 158 | 63 | 514 | 302 |
Sahin |
7 | 2009 | Turkey | 10.5 | 11 | 9 | 167 | 177 |
Chabchoub |
8 | 2009 | Tunisia | 11 | 7 | 12 | 150 | 236 |
Sfar |
33 | 2009 | Tunisia | 11 | 33 | 2 | 133 | 100 |
Morgan |
34 | 2009 | UK | 10 | 1324 | 705 | 4789 | 3630 |
Majorczyk |
35 | 2010 | Poland | 11 | 259 | 118 | 371 | 543 |
Present study | - | 2011 | Italy | - | 44 | 34 | 396 | 477 |
In the first meta-analysis a significant and positive association between the
Forest plot of published studies in relation to the first meta-analysis (24 studies). The association of the
The funnel plot (
Funnel plot of published studies in relation to the first meta-analysis (24 studies).
There was not a significant difference between the first meta-analysis and the second one: combined OR = 1.80 with 95%CI = (1.61–2.02) using random effect estimate (Cochran's Q test: χ2 = 79.42, df = 22,
The studies with a quality score ≥11 were 14. The relationship between
When looking at the T allele frequency in RA patients and controls, we noticed a North-South gradient with higher values in Finland, Germany, Hungary, and lower values in Spain, Italy, Tunisia, Greece and Turkey (
Geographical distribution of the T allele frequency at
The association between the T1858 allele at
Furthermore, it is also noteworthy that while in Germany, the frequency of the T1858 allele was significantly higher in RA patients (21.3%) compared to controls (10.0%; with an OR of 2.43) and the association was present irrespective of the presence or absence of anti-CCP and RF
Genetic differences within European populations have been once more underlined by a recent work of Rodríguez-Rodríguez
Our data revealed a higher frequency of the T1858 allele in RA Italian patients compared to the controls cohort. On the other hand, the frequency in controls was lower than that observed in France or in Germany and similar to Turkish, Greek and Tunisian populations.
Interestingly, Mediterranean populations are genetically linked by a common history of migrations, like the abiding one of Saracens and Moors. In fact a recent work, estimating the medieval North African contribution over Mediterranean countries through the analysis of the Y chromosome short tandem repeats, suggested a general correlation between historical and genetic data of Iberia, Sicily, Turkey and North Africa
No relationship arose between the C/T-T/T genotypes presence and auto-antibodies positivity. The demonstration of a gain-of-function conferred by the T1858 allele in suppressing TCR (T cell receptor) function in T cells and BCR (B cell receptor) function in B cells raises new hypotheses on the role of tyrosine phosphatases. The T1858 allele might increase the threshold for a persistent activation of both autoreactive T and B cells thus leading to a more defined autoimmune subset of RA
In conclusion, the geographical distribution of SNPs in the world, linked to different population origins, should be taken into account in studies regarding genetic associations. Given that specific therapies directed toward Lyp will be available in the near future for various autoimmune diseases
(TIF)
(DOC)
(DOC)