Clusterin, TNF- α , and IL-6 polymorphism and implications on Alzheimer’s disease risk determination in Saudi population

23 Background: In the wake of the warning by WHO that the prevalence of dementia may rise 24 by 125% in Middle East by 2050, identification of genetic risk factors in Arab population is 25 urgent. 26 Methods: To genotype the Single Nucleotide Polymorphisms (SNPs) in clusterin (CLU) 27 rs11136000, rs1532278; tumor necrotic factor (TNF- α ) -308 rs1800629 A/G and -857 28 rs1799724 T/C; interleukin-6 (IL-6) rs1800796 G/C (-572 G/C) and rs1800795 G/C (-174 29 G/C), and to determine their association with Alzeimer’s Disease (AD), DNA was isolated 30 from the blood of 42 elderly Saudi AD patients (19 male, 23 female) and 23 healthy controls 31 (11 male, 12 female), recruited for this study. Total serum cholesterol, LDL-C, HDL-C, and 32 triglyceride levels were measured using an autoanalyzer. Serum concentrations of beta- 33 amyloid 1–40 (A β 1-40), beta-amyloid 1–42 (A β 1-42), CLU, and inflammatory biomarkers (IL- 34 6, TNF- α , and the C-reactive protein) were assessed by ELISA. The gene polymorphisms were 35 analyzed by RT-PCR using the TaqMan assay. 36 Results : The results show that in the rs1532278 SNP of CLU gene, GA heterozygous allele 37 was significantly higher in AD patients (57.1%) than in the control subjects (26.1%; OR = 3.67, 38 CI = 1.10-12.32; p=0.036), thus it may be a risk factor for AD. On the other hand, the AD 39 patients who carried genotype GG for TNF- α SNP rs1800629 showed significantly higher 40 levels of serum IL-6 ( p = 0.04), and hence may increase susceptibility to AD in Saudi 41 population.


Statistical analysis 145
All statistical analyses were performed using IBM SPSS (version 21) statistical package (IMB,  146 Armonk, NY, USA). Biochemical parameter data were presented as the mean ± standard 147 deviation (SD) for normal variables, whereas the median (1 st Quartile to 3 rd Quartile) was given 148 for non-normal variables. A p-value ≤ 0.05 was considered to be significant. P-values were 149 obtained using the Mann-Whitney U test when two groups were compared, and the Kruskal-150 Wallis test when three groups were compared. Data were presented using Graph Pad Prism 8. 151 For genotyping, data were presented as N (%), odds ratio (OR), and 95% confidence intervals 152 (CIs), with a p-value ≤0.05 considered to be significant. P-values were obtained from logistic 153 regression analyses. The Hardy-Weinberg equilibrium was tested using the Chi-square test. 154 155 rs1800629 Chr6:31575254 A G Promoter

Participant characteristics 158
A significantly (p = 0.004) enhanced expression of CLU was detected in the AD patients (340.4 159 µg/ml ± 74.6) compared with that of the control group (265.0 µg/ml ± 80.9). The serum levels 160 of other biochemical markers of Alzheimer's disease, such as Aβ1-42, Aβ1-40, CRP, and 161 cytokines TNF-α and IL-6 were comparable between the two tested groups. No significant 162 differences could be observed except in case of CLU (Table 2A). 163

175
Furthermore, the clinical data suggests no significant difference between the Alzheimer's

Biochemical parameters in relation to genotypic characteristics of participants 184
No significant differences were observed for any measured biochemical parameters in relation 185 to the CLU gene SNP rs1532278 or the CLU gene SNP rs11136000 genotypes in the two tested 186 groups of participants (Table 3; Table 4). 187 For the control group, the carriers of GG genotype for IL-6 gene SNP_rs1800796 had 191 significantly lower systolic blood pressure (p = 0.009) and triglycerides (p = 0.04) compared 192 to carriers of CC genotype for this SNP (Table 5). 193  238 239

Genotyping analysis 240
The genotype analysis for the CLU gene SNP rs11136000 revealed no significant differences 241 in distribution between the two groups (

258
The present study aimed to evaluate the associations of selected polymorphisms in the CLU, 259 IL-6, and TNF-α genes with the risk of AD in the Saudi population. The CLU gene is one of 260 the ten non-APOE genome-wide significant risk loci that carries SNPs with functional 261 evidence. Serum CLU levels are implicated in the onset, prevalence, progression, and severity 262 of AD, as well as brain atrophy, through several pathways such as lipid metabolism, beta-263 amyloid aggregation and clearance, apoptosis, neuroinflammation, and neuronal cell cycle 264 control [20]. A significant association between the CLU gene SNP rs11136000 and a reduced 265 risk of LOAD has been reported in the European population [21][22]. The present study found 266 that the heterozygote GA for CLU rs1532278 was significantly higher in patients compared to 267 that of the controls (p = 0.036). Our result finds support in a study that reported a strong and 268 significant association between rs1532278 and an increased risk of AD through its involvement promoter showed no significant difference between AD patients and the healthy control group in the present study. However, it was interesting to note that in our study, the levels of TNF-α 308 in AD patients with the CC genotype was higher than in AD patients with GG genotypes for 309 SNPs rs1800796 of the IL-6 gene. This finding showed a trend towards significance (p=0.06; 310 Table 5). These findings also support the hypothesis, which suggests that interactions between 311 cytokines can affect cytokine production through T-cell activation, which in turn produces 312 more mediators. Several studies in the past have attempted to find an association between the 313 IL-6 rs1800795 polymorphism with AD, but the results have been contradictory. Studies among the subjects of the present study. We found that TNF-α rs1800629 polymorphism lacks 330 an association with susceptibility for increased or decreased risk of AD in Saudi subjects.
progression [38]. The level of IL-6 was found to be significantly low among AD subjects 333 carrying AG genotypes compared to those carrying GG genotypes for SNP_rs1800629 of TNF-334 α gene (p=0.040). This finding suggests that TNF-α -308 A/G polymorphism 335 (SNP_rs1800629) may affect the level of IL-6 in AD patients with specific genotypes. This 336 result can be explained by the fact that TNF-α is known to not only strongly induce the secretion 337 of IL-6 but it may also modify IL-6 signaling and vice versa via a crosslink [39]. 338 Conclusion: Striking finding of this study is the observation that genetic variants in rs1532278 339 SNP of CLU gene, GA heterozygous allele variant, and the serum levels of the CLU protein 340 may be associated with an increased risk of Alzheimer's disease among Saudi Arabian subjects. 341 Secondly, AD patients who were carriers of GG genotype for TNF-α SNP rs1800629 had 342 significantly higher serum levels of IL-6, highlighting the possible association of this genotype 343 with AD. Further studies using a larger sample size are needed to confirm the present finding. 344 Ethics approval and consent to participate: 345 The study was approved by the Ethical Committee, at the College of Science and the internal 346 review board at the college of medicine, King Saud University (# KSU-SE-18-24) and all 347 Figure 1 Genotype analysis of CLU gene SNP rs 11136000 and rs15322278 shows no signi cant difference in patient (A) and control groups (B) with reference to rs11136000 while GA was signi cantly higher in patients (D) than in control (C) in case of rs15322278.