Abstract:Results A total of 2158 cases and 2088 controls in 12 case-control studies were included. The results indicated that the A allele carriers (AA + AG) had a 27% increased risk of AD, when compared with the homozygote GG (OR = 1.27, 95%CI = 1.02–1.58 for AA + AG vs. GG). In the subgroup analysis by ethnicity, significant elevated risks were associated with A allele carriers in Europeans (OR = 1.27 and 95%CI = 1.01–1.59 for AA + AG vs. GG), but not in Asians (OR = 1.37 and 95%CI = 0.32–5.88 for AA + AG vs. GG). Conclusions This meta-analysis suggested that the −1082G/A polymorphism of IL-10 gene would be a risk factor for AD. To further evaluate gene-to-gene and gene-to-environmental interactions between polymorphisms of IL-10 gene and AD risk, more studies with large groups of patients are required. Keywords Alzheimer's disease IL-10 Meta-analysis Polymorphism Risk 1 Introduction Alzheimer's disease is the major cause of cognitive deterioration in the elderly and presents a substantial public health problem [1–3] . Its major characteristics include progressive synapse and neuronal loss, formation of neuritic plaques and neurofibrillary tangles [1,4] . There are growing evidences to suggest that inflammatory process play a crucial role in the pathogenesis of AD [5] . Many inflammatory factors have been detected by pathological studies [6] . Multiple cytokines such as IL-6 , TNF-α , IL-1β are overexpressed in activated microglia surrounding plaques in AD [7] . In addition, evidence suggests that the risk of AD is associated with genetic variants in inflammatory mediators, such as IL-1β [8] , IL-6 [9] , TNF-α [10] and IL-10 [11] . Genetic variants in these genes may affect biological activity by transcription, translation and secretion. IL-10 , a main anti-inflammatory cytokine, has been suggested to play an important role in neuronal homeostasis and cell survival [12] . It promotes neuronal and glial survival and downregulates the expression of pro-inflammatory cytokine genes through activation and/or induction of transcription factors [13] . IL-10 -positive cells around beta plaques are increased significantly with age in AD animal models [14] . The human IL-10 gene is located on chromosome 1q31–32 and consists of five exons [1] . The IL-10 promoter region is highly polymorphic and many polymorphisms have been identified, such as −1082G/A(rs1800896), −829C/T(rs1800871) and −592C/A (rs1800872) [1] . A number of studies have reported the associations between −1082G/A polymorphism in IL-10 gene and AD risk [1,2,4,11,13,15–21] , but the results were inconclusive. This polymorphism was suggested as a risk factor for AD in Italian population [21] and Chinese population [19] , but not in German population [13] . This is partly due to the different genetic effects in different ethnic populations [22] . In addition, a small sample size may result in weak statistical power [22] . In order to derive a more precise conclusion, we performed a meta-analysis to pool all available results of the association between IL-10 −1082G/A polymorphism and AD risk. This is, to our knowledge, the first genetic meta-analysis conducted with respect to the association between the IL-10 polymorphism and AD risk. 2 Methods and materials 2.1 Publication search We carried out a search in Pubmed and Embase, covering all the papers published before 2010, with a combination of the following words: polymorphism or mutation or variant , and interleukin-10 or IL-10 , and Alzheimer's disease (last search was updated on September 17th, 2010). We evaluated potentially relevant publications by examining their titles and abstracts and all studies matching the eligible criteria were retrieved. All languages were included. Studies included in the current meta-analysis had to meet all of the following inclusion criteria: (1) evaluation of the −1082G/A polymorphism in IL-10 gene and AD risk, (2) using of a case-control design, and (3) genotype distributions in both cases and controls were available for estimating an odds ratio (OR) with 95% confidence interval (CI). Accordingly, the following exclusion criteria were also used: (1) abstracts and reviews, (2) genotype frequency not reported, and (3) repeat or overlapping publications. 2.2 Data extraction Data were independently abstracted by two reviewers (JZ and YGZ) using a standard protocol and data-collection according to the inclusion criteria. The following data were abstracted from each study: the name of first author, year of publication, country origin, ethnicity, definition of cases, source of control, genotyping methods, total number of cases and controls, genotype distributions in cases and controls. Different ethnicity descents were categorized as European and Asian. 2.3 Statistical analysis ORs with 95%CI were used to assess the strength of the association between the IL-10 −1082G/A polymorphism and AD risk. The pooled ORs were calculated for dominant model (AA + AG vs. GG), recessive model (AA vs. AG + GG), heterozygote comparison (AG vs. GG), homozygote comparison (AA vs. GG) and allele comparison (A vs. G). The pooled ORs were calculated by a fixed-effects model (the Mantel–Haenszel method) or a random-effects model (the Dersimonian and Laird method) according to the heterogeneity. Heterogeneity was evaluated by a X 2 based Q statistic and was considered statistical significant at P value less than 0.10. When the P value is greater than 0.10, the pooled OR estimate of each study was calculated by the fixed-effects model, otherwise, a random-effect model was used. To evaluate the ethnicity-specific effects, subgroup analyses were performed by ethnic group. Publication bias was analyzed by several methods. Visual inspection of asymmetry in funnel plots was carried out. The Begg's test and Egger's test were also used to statistically assess publication bias [23] . Sensitivity analysis was also performed by sequence excluding individual study to check the stability of the result. All statistical tests were performed by using the Revman 4.2 software and STATA 10.0. 3 Results 3.1 Study characteristics Pubmed and Embase databases were comprehensively searched. There were 158 records relevant to our searching strategy. After screening the titles and abstracts, 18 eligible articles were identified. During the extraction of data, 5 articles were excluded, because they did not provide the usable data of −1082G/A polymorphism for OR calculation. Additionally, one case-control was excluded for data repeated with another study. Thus, a total of 12 case-control studies were identified which met our inclusion criteria [1,2,4,11,13,15–21] . The characteristics of selected studies are summarized in Table 1 . Genotype distributions for each case-control are also shown in Table 1 . There were 11 case-control studies of Europeans, one of Asians. 3.2 Meta-analysis results The main results of this meta-analysis are shown in Fig. 1 (a, AA + AG vs. GG; b, AA vs. AG + GG; c, AA vs. GG; d, AG vs. GG; e, A vs. G) summarized in Table 2 . Significant increased AD risk was observed in dominant model (OR = 1.27, 95%CI = 1.02–1.58, P = 0.03). However, no significant association between this polymorphism and AD risk was observed in other comparison models. In the subgroup analysis by ethnicity, significant increased AD risks were found among Europeans for dominant model (OR = 1.27, 95%CI = 1.01–1.59, P = 0.04), but not for other genetic models. For Asians, no significant association between this polymorphism and AD risk was observed in all comparison models ( Table 2 ). 3.3 Publication bias Begg's funnel plot and Egger's test were performed to assess the publication bias of the literature. The shapes of the funnel plots seemed approximately symmetrical (AA + AG vs. GG) and the Egger's test did not show any evidence of publication bias (t = 0.04 and P = 0.966 for AA + AG vs. GG) ( Fig. 2 ). 3.4 Sensitivity analysis No significant associations were obtained after sequentially excluding five case-control studies [2,13,15,16,21] , suggesting that these studies were responsible for the significant of association (data not shown). 4 Discussion AD is the leading cause of cognitive deterioration in the elderly, and affects approximately 10% after age of 65 [6] . Accumulated evidences indicate that neruoinflammation and cytokines appear to play a critical role in the development of AD. IL-10 , an important cytokine, has an anti-inflammatory role in the brain and may be involved in the pathogenesis of AD. Genetic studies showed that the −1082G/A polymorphism in the promoter region was associated with AD risk, and a number of studies have been performed to investigate that association. However, inconclusive results were obtained. In order to confirm the association between IL-10 −1082G/A polymorphism and AD, we performed a meta-analysis. In the current meta-analysis, we found a significant relationship between IL-10 −1082G/A polymorphism and AD risk. Our results indicated that individuals who carry variant A allele (AA + AG) had a nearly 27% increased risk of AD. In the subgroup analysis by ethnicity, results indicated that there was significant association between IL-10 −1082G/A polymorphism and AD risk in Europeans, but not in Asians, suggesting genetic diversity among ethnicities. However, because there was only one study performed in Asians, these results may not be valid for Asians. It is widely accepted that AD is a complex disease and both genetic factors and environmental factors contribute to its inception and evolution [1–4,11,13,15,17–21,24] . Environmental factors include ethnicity, gender, age, diet, family history of disease and so on. It is suggested that there might be interaction between the environmental factors and IL-10 gene. Further analyses should be conducted to validate the environmental modification of IL-10 −1082G/A on AD risk. Some limitations of this meta-analysis should be considered when explaining our results. First, because a lack of original information for each included individuals, data were not stratified by other variables, such as, gender, mean age of AD patients at onset, MMSE, criteria used to determine AD patients. Second, only one included study was performed in Asians, so the association needs to be validated in other ethnicities too, such as Asians and Africans. Third, because some relevant published and unpublished studies which were not indexed by the selected databases were not included, a possible bias could not be ruled out, although our results of publication bias showed no significance. In conclusion, this meta-analysis provided evidence that −1082G/A polymorphism in IL-10 gene is associated with an increased risk of AD, especially in Europeans. However, larger well-designed studies are of great value to confirm these findings, especially in Asians and Africans. Moreover, gene–gene and gene–environmental interactions should also be considered. Conflict of Interest None declared. Acknowledgements This study was supported by grants # 30470761 , 30871117 from National Natural Science Foundation of China . References [1] D. Culpan J.A. Prince S. Matthews L. Palmer A. Hughes S. Love Neither sequence variation in the IL-10 gene promoter nor presence of IL-10 protein in the cerebral cortex is associated with Alzheimer's disease Neurosci Lett 408 2006 141 145 [2] E.M. Ramos M.T. Lin E.B. Larson I. Maezawa L.H. Tseng K.L. Edwards Tumor necrosis factor alpha and interleukin-10 promoter region polymorphisms and risk of late-onset Alzheimer disease Arch Neurol 63 2006 1165 1169 [3] Y. Qian W. Chen J. Wu T. Tao L. Bi W. Xu Association of polymorphism of DNA repair gene XRCC1 with sporadic late-onset Alzheimer's disease and age of onset in elderly Han Chinese J Neurol Sci 295 2010 62 65 [4] G. Ribizzi S. Fiordoro S. Barocci E. Ferrari M. Megna Cytokine polymorphisms and Alzheimer disease: possible associations Neurol Sci 31 2010 321 325 [5] H. Akiyama S. Barger S. Barnum B. Bradt J. Bauer G.M. Cole Inflammation and Alzheimer's disease Neurobiol Aging 21 2000 383 421 [6] W. Swardfager K. Lanctot L. Rothenburg A. Wong J. Cappell N. Herrmann A meta-analysis of cytokines in Alzheimer's disease Biol Psychiatry 2010 10.1016/j.biopsych.2010.06.012 [7] E.E. Tuppo H.R. Arias The role of inflammation in Alzheimer's disease Int J Biochem Cell Biol 37 2005 289 305 [8] D. Di Bona A. Plaia S. Vasto L. Cavallone F. Lescai C. Franceschi Association between the interleukin-1 beta polymorphisms and Alzheimer's disease: a systematic review and meta-analysis Brain Res Rev 59 2008 155 163 [9] M. Wang J. Jia The interleukin-6 gene -572C/G promoter polymorphism modifies Alzheimer's risk in APOE epsilon 4 carriers Neurosci Lett 482 2010 260 263 [10] D. Di Bona G. Candore C. Franceschi F. Licastro G. Colonna-Romano C. Camma Systematic review by meta-analyses on the possible role of TNF-alpha polymorphisms in association with Alzheimer's disease Brain Res Rev 61 2009 60 68 [11] E. Cousin S. Macé C. Rocher C. Dib G. Muzard D. Hannequin No replication of genetic association between candidate polymorphisms and Alzheimer's disease Neurobiol Aging 2009 10.1016/j.neurobiolaging.2009.09.004 [12] K. Strle J.H. Zhou W.H. Shen S.R. Broussard R.W. Johnson G.G. Freund Interleukin-10 in the brain Crit Rev Immunol 21 2001 427 449 [13] C. Scassellati R. Zanardini R. Squitti L. Bocchio-Chiavetto C. Bonvicini G. Binetti Promoter haplotypes of interleukin-10 gene and sporadic Alzheimer's disease Neurosci Lett 356 2004 119 122 [14] J. Apelt R. Schliebs Beta-amyloid-induced glial expression of both pro- and anti-inflammatory cytokines in cerebral cortex of aged transgenic Tg2576 mice with Alzheimer plaque pathology Brain Res 894 2001 21 30 [15] B. Arosio L. Mastronardi C. Vergani G. Annoni Intereleukin-10 promoter polymorphism in mild cognitive impairment and in its clinical evolution Int J Alzheimers Dis 2010 854527 2010 [16] P. Vural S. Degirmencioglu H. Parildar-Karpuzoglu S. Dogru-Abbasoglu H.A. Hanagasi B. Karadag The combinations of TNFalpha-308 and IL-6 −174 or IL-10 −1082 genes polymorphisms suggest an association with susceptibility to sporadic late-onset Alzheimer's disease Acta Neurol Scand 120 2009 396 401 [17] O. Combarros P. Sanchez-Juan J.A. Riancho I. Mateo E. Rodriguez-Rodriguez J. Infante Aromatase and interleukin-10 genetic variants interactively modulate Alzheimer's disease risk J Neural Transm 115 2008 863 867 [18] S. Bagnoli E. Cellini A. Tedde B. Nacmias S. Piacentini V. Bessi Association of IL10 promoter polymorphism in Italian Alzheimer's disease Neurosci Lett 418 2007 262 265 [19] S.L. Ma N.L. Tang L.C. Lam H.F. Chiu The association between promoter polymorphism of the interleukin-10 gene and Alzheimer's disease Neurobiol Aging 26 2005 1005 1010 [20] C. Depboylu Y. Du U. Muller A. Kurz R. Zimmer M. Riemenschneider Lack of association of interleukin-10 promoter region polymorphisms with Alzheimer's disease Neurosci Lett 342 2003 132 134 [21] D. Lio F. Licastro L. Scola M. Chiappelli L.M. Grimaldi A. Crivello Interleukin-10 promoter polymorphism in sporadic Alzheimer's disease Genes Immun 4 2003 234 238 [22] H. Liu M. Yang G.M. Li Y. Qiu J. Zheng X. Du The MTHFR C677T polymorphism contributes to an increased risk for vascular dementia: a meta-analysis J Neurol Sci 294 2010 74 80 [23] Y. Zhang J. Zhang J. Huang X. Li C. He C. Tian Polymorphisms in the transforming growth factor-beta1 gene and the risk of asthma: a meta-analysis Respirology 15 2010 643 650 [24] B.R. Del S. Ghezzi E. Scarpini N. Bresolin G.P. Comi VEGF genetic variability is associated with increased risk of developing Alzheimer's disease J Neurol Sci 283 2009 66 68

The −1082G/A polymorphism in IL-10 gene is associated with risk of Alzheimer's disease: A meta-analysis