Individuals with AD (n=40) were recruited from PSG Institute of Medical Sciences and Research (PSG IMS&R), Coimbatore and Alzheimer’s and Related Disorders Society of India (ARDSI), Thrissur, India. The present study was approved by the Institutional Ethics Committee of PSG IMS&R and all procedures involving human subjects were in accordance with the Declaration of Helsinki. Informed consents were obtained from control subjects and from the relatives/legal guardians of the AD patients for their inclusion in the study. The cognitive status of each subject was assessed using Mini-Mental State Examination (MMSE) (16) and the patients were categorized according to the criteria of NINCDS-ADRDA and DSM-IV for dementia (17). AD patients showing co-morbidity with cardiovascular disease and diabetes were excluded from the study. The control subjects (n=40) included in the study were in the age group ranging between 65-85 years with a negative history of cardiovascular disorders, diabetes and neurological disorders. Subjects with habits of smoking or alcohol consumption or having certain advanced malignancies such as AIDS and other major illnesses were excluded from the study. The blood samples (5 ml) from AD subjects and age matched healthy controls were collected individually in EDTA tubes and 4 ml was utilized for separation of plasma, PBMC and RNA. The remaining 1 ml was utilized for DNA isolation.

Peripheral blood mononuclear cells (PBMC) were isolated by HiSep LSM density gradient separation (Himedia, Mumbai, India). Total RNA from PBMC was isolated using RNA-XPress™ reagent (Himedia, Mumbai, India). Quality of RNA was analyzed using a nanospectrometer, (Imple, USA). 250 ng of total RNA was then reverse-transcribed using RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific, USA) and stored at -20º C until further use.

Expression analysis was performed for total RAGE (tRAGE), sRAGE and TNF-α using Quantitative real-time PCR (CFX96 Touch™ Real-Time PCR - Bio Rad) with 2X SYBR Green qPCR master mix (Applied Biosystems, USA). Information regarding the primer sequences used for real-time PCR analysis can be found in Table 1. The primers for tRAGE were designed with reference to the fRAGE gene (GenBank ID – NM001136) using Primer-Blast tool ensuring that it was comprised the exon-exon junction to avoid amplification of contaminating genomic DNA. On the other hand, the primers for soluble RAGE (containing intron 9) were designed to have an exon-intron junction and an exon-exon junction, ensuring it amplified all the prominent soluble isoforms (Figure 1). Though primers used for tRAGE can amplify other variants excluding sRAGE, the canonical variant fRAGE has been reported to be the major variant (70% to 80%) expressed in human tissues (18).

Figure 1

Schematic representation of qPCR primer designed to amplify tRAGE and sRAGE variants. RAGE gene exons, intron; Forward primer: FP, Reverse primer: RP.

The PCR conditions were used as follows: 50ºC for 2 min, 95ºC for 10 min, followed by 40 cycles of 95ºC for 10s, and 60ºC for 1 min. Relative quantification of gene transcripts was calculated using 2^-ΔΔCt method and normalized to Glyceraldehyde-3-phosphate dehydrogenase (GADPH) transcripts.

Samples of blood collected in EDTA tubes were centrifuged at 400 g for 15 minutes to separate plasma, which was stored at -20 °C in aliquots till use. Aβ42 and sRAGE were quantified in plasma using a commercially available ELISA kit from CUSABIO (CSB-E10684h) and Quantikine R&D systems (DRG00, USA), respectively, according to the manufacturer’s instructions.

The Polymerase chain reaction – restriction fragment length polymorphism (PCR- RFLP) was designed to identify G82S RAGE polymorphism. The genomic DNA was isolated using Xpress DNA kit (Magenome, India), according to the manufacturer’s instruction. The PCR was carried out in EmeraldAmp® PCR Master Mix (Takara, Japan) with genomic DNA (200 ng) and 10 µM of forward (GTCCCTCACCCACCCTGTC) and reverse primers (GCCACACACCCACACCCAC) for amplification of exon 3 region of RAGE gene containing G82S polymorphism. The amplified product (817 bp) was subjected to restriction digestion by Alu I (5 units) for 2h at 37 ºC followed by electrophoresis on 2% agarose gel.

Statistical analysis was performed using the statistical package SPSS Version.16, Chicago, USA. An independent sample t-test was used to compare cases between AD and control. Receiver operating characteristic (ROC) curve analysis was performed to study the reliability of sRAGE in plasma as marker for AD diagnosis. Pearson’s correlation was done wherever applicable. For genotype and allelic frequency, the level of significance was analyzed using chi-squared test. The disease relationship with genotype and allelic frequencies is presented as an odds ratio with 95% confidence interval.

Expression of tRAGE, sRAGE and TNF-α were quantified using qPCR with gene specific primers. To analyze mRNA expression of RAGE variants, the comparative threshold cycle (Ct) method was applied with GAPDH as the endogenous reference gene. The first, synthesized strand cDNA was verified by the amplification of GAPDH. Melt curve analysis of sRAGE, tRAGE and TNF-α revealed a single peak which represents a single amplicon and specific amplification (Figure 2A). The results revealed that while the expression sRAGE was down regulated by 3.4-fold, tRAGE expression was upregulated by 1.9-fold in AD samples compared to controls. There was also an upregulation in the expression of TNF-α (2.2 fold) in the AD group associated with lower expression of sRAGE in AD (Figure 2B, Table 3).

Figure 2

Expression profile of RAGE variants and TNF-α. (A) Melt curves from qPCR for sRAGE, tRAGE and TNF-α. (B) Analysis of RAGE variants and TNF-α mRNA expression in PBMC. Bar graph depicts mean gene expression levels and error bars represents the standard error mean obtained from the average of 40 nos. Expression values are obtained by 2^-ΔΔCT.

The levels of sRAGE in plasma was significantly lower in patients AD (151.91±15.9 pg/mL, p<.001) as compared to the levels found in cognitively healthy age matched control (753.79±58.12 pg/mL) (Figure 3A). The plasma Aβ42 level was found to be higher in AD (191.57 ± 9.5 pg/ml, p<.0001) compared to controls (88.48 ± 6.7 pg/ml) Table 4. The correlation analysis between plasma sRAGE and Aβ levels has shown no significant relation (AD R = 0.122; age matched healthy control subjects R = -0.080) (Figure 3B). ROC curve analysis was conducted to assess the cutoff points of sRAGE to differentiate the AD group from the controls. The area under the curve was found to be 0.981, indicating good diagnostic accuracy for sRAGE with a cutoff value of ≤ 278.09 pg/ml (Sensitivity 100 %, Specificity 85%) suggesting that plasma sRAGE could be used to distinguish AD from controls (p = 0.001; 95% CI, 0.960 – 1.00, Figure 3C).

Figure 3

Plasma sRAGE levels in AD. (A) Plasma sRAGE in AD and control. The values are shown as mean ± SEM and were found to be statistically significant compared to control, (p<0.001). (B) The correlation between concentration of sRAGE and Aβ42 (i) AD (R=0.122) ii) controls (R= -0.080). (C) Receiver operating characteristic curve analysis of plasma sRAGE: AUC: 0.981 (p < 0.001), cutoff value of ≤ 278.09 (Sensitivity 100 %, Specificity 85%). The quality of diagnosis test is determined by AUC: 0.9-1 excellent, 0.8-0.9 good, 0.7-0.8 fair, 0.6-0.7 poor, 0.5-0.6 fail.

In the present study PCR-RFLP was adopted to analyze the G82S polymorphism. Information on the primers used, nucleotide sequence of PCR products and AluI restriction site in the PCR product are all represented in Figure 4A. The representative gel image of six samples after digestion of the PCR-amplified products (Figure 4B) and schematic pattern of RFLP after digestion with AluI enzyme can be found in Figure 4C. The distribution of genotype was in accordance with the Hardy–Weinberg equilibrium. The genotypic and allelic frequency of G82S RAGE polymorphism in AD and control group along with the resulting odds ratio and significance levels is represented in Table 5.

Figure 4

G82S RAGE Polymorphism. (A) Nucleotide sequence of the PCR-amplified RAGE gene showing the restriction site of AluI that are highlighted in grey and yellow color and indicated by arrow. The yellow color represents the restriction site that occurred due to single nucleotide change from G to A that shifted Gly 82 to Ser. Primer sequences are highlighted in gray. (B) PCR RFLP results of G82S RAGE gene polymorphism. PCR product: Lane - 2,4,6,8,10,12; Restricted samples (AluI digest): Lane 3,5,7,9,13- Genotype GG, Lane 11 - Genotype GS, Lane 1- 100 bp Ladder. (C) Restriction pattern of RAGE for homozygous GG and SS and heterozygous GS. After AluI digestion, the G82G genotype of RAGE gives two fragments and S82S genotype gives three fragments. Due to single nucleotide polymorphism, the heterozygous GS allele gives four fragments.

The S82S genotype was not observed in both groups. Allelic frequency of S allele in the AD and control group was 0.225 and 0.037 respectively, whereas for G allele, it was 0.775 and 0.963 respectively. The observed difference in genotype distribution and allelic frequency between controls and the AD group was statistically significant (χ² test, p <0.05) (Table 5). Reduced sRAGE in subjects with GS genotype was also observed compared to GG genotype and the results were statistically significant (Figure 5).

Figure 5

Association of G82S RAGE genotype on sRAGE and Aβ42 expression in AD. (A) sRAGE in different RAGE genotypes. It is found that the expression of sRAGE was reduced in AD, which is further reduced in G82S RAGE genotype compared to G82G RAGE in AD and control. (B) Aβ42 levels in different RAGE genotypes. The expression of Aβ42 was observed to be high in AD in both genotypes compared to control. The values are shown as mean ± SEM. The effects of G82S polymorphism on the levels of sRAGE and Aβ42 expression in AD were found to be statistically significant with the p value, p<0.0001