Glycation end-products specific auto-antibodies in Systemic Lupus Erythematosus

Systemic lupus erythematosus (SLE) is an autoimmune disease, which is highly inflammatory. Compared to a healthy control group, SLE patients exhibit a higher concentration of advanced glycation end products (AGEs) and a lower concentration of receptors for AGEs (RAGE) in serum, however, the exact aetiology is still unclear. In the present study, non-enzymatic glycation induced modification of human serum albumin (HSA) has been studied by biophysical techniques. Glycated HSA (G-HSA) was used as an antigen, and serum autoantibody levels were estimated in SLE and normal humans (NH) against it, using direct binding ELISA and competitive inhibition ELISA. Compared to N-HSA, remarkable structural modifications were observed in G-HSA. Modified HSA also showed increased pentosidine fluorescence (213.7 ± 13.4 AU). Glycation of HSA induced a conversion of α-helix and random coil to β-sheet and β-turns. Serum immuno assays results exhibited significantly (p < 0.001) higher binding of G-HSA with serum autoantibodies from SLE patients when compared with native HSA (N-HSA). Furthermore, competitive ELISA results showed significantly (p < 0.001) high percent inhibition of serum IgG from SLE patients with modified antigen. Chronic inflammation with excessive oxidative stress in SLE patients could be a possible reason for structural alterations in blood proteins, generating highly immunogenic unique new-epitopes. These in turn induce the generation of specific autoantibodies against G-HSA that may serve as a potential biomarker for SLE pathogenesis.

could be a possible reason for structural alterations in blood proteins, generating highly immunogenic unique new-epitopes. These in turn induce the generation of specific autoantibodies against G-HSA that may serve as a potential biomarker for SLE pathogenesis.

Background:
Systemic lupus erythematosus (SLE) is a chronic inflammatory disease, which is also autoimmune in nature. The exact aetiology of the disease is still unknown thus, the available medicines and approved therapies are not sufficient to properly manage the disease. B cells are involved in the production of autoantibodies; hence, their depletion or inactivation is also a potential SLE treatment option [1]. A range of different factors, such as immune dysfunction of cells such as B cells, dendritic cells and neutrophils as well as genetic and environmental effects can be implicated in the pathogenesis of SLE is attributed to dysfunction of different immune cells such as B cells, dendritic cells [2]. The formation pathogenic immune complexes mediate the disease by molecular mimicry or endogenous antigen modification followed by autoantibody expression, that in turn cause tissue damage [3,4]. Non-enzymatic glycation is a condensation reaction of monosaccharides and reactive amino acid groups located on intracellular and extracellular proteins. The resultant Schiff base intermediates undergo slow Amadori rearrangement to yield stable glycated protein adducts [5]. Further irreversible chemical changes to these protein-glucose adducts may lead to the formation of advanced glycation end products (AGEs).Advanced glycation of proteins causes gluco-oxidation and renders them immunogenic [6]. Evidence that AGEs are antigenic has led scientists to hypothesize that in vivo AGE structure may wield a larger autoimmune response [7]. Recently autoantibodies against glycated proteins such as immunoglobulins have been detected in rheumatoid arthritis (RA), diabetes, as well as in the elderly healthy individuals [6, [8][9][10][11]. Gluco-oxidative damage to proteins is increasingly being implicated in diabetes mellitus, RA, artherosclerosis, Alzheimers, amyloidosis and aging [8,9,[12][13][14]. Albumin is the most commonly found protein in human blood and hence most susceptible to non-enzymatic glycation [15]. Thus, it is important to validate the presence of autoantibodies against G-HSA in SLE patient sera to infer the role of gluco-oxidative HSA in inducing autoimmunity in SLE.

Collection of blood samples:
Out of 25  Biophysical characterization of G-HAS: Tryptophan fluorescence: An excitation wavelength of 285 nm was used to determine fluorescence intensity for tryptophan residue in the N-HSA and G-HSA. Emission was recorded between 290-440 nm [17]. Slit widths used were 10 nm. Protein solutions used were of identical concentration (100 μM). Hitachi model F2700 spectrofluorometer (Japan) was used to analyse the samples.

AGE-pentosidine fluorescence:
Fluorescence specific for pentosidine was applied to detect the pentosidine residue in both glycated and non-glycated samples. Excitation wavelengths of 375 nm and peaks were observed between 330-420 nm [18]. Samples used in this assay were at the concentration of 60 μM Circular dichroism (CD): CD of both (N-HSA and G-HSA) samples (2.5μM) were recorded as published previously [15,19]. N-HSA and G-HSA results for CD analysis were expressed in milli degrees. Each sample was recoded three times and average ± SD values were calculated and given. A wavelength of between 200 -230 CD was used to record the samples with 5 mm/milli degree sensitivity. Preparation of sample solution was in 20 mM sodium phosphate buffer, pH 7.4. Secondary structural elements were quantified based on Chen and Yang equation [19].
Biochemical analysis of serum samples: Serum carbonyl content: Carbonyl content bound to protein in SLE (n= 25) and NH (n=25) serum samples was quantified [20]. Results for all the samples were calculated in number of nanomoles of carbonyl/mg of protein using ε379 = 22,000 M -1 .cm -1 .

Serum pentosidine detection by ELISA:
Competition ELISA kit was used to detect pentosidine in serum samples of individuals of all groups (FSK pentosidine ELISA kit; Fushimi Pharmaceutical, Kagawa, Japan) as described previously [11]. Briefly, pronase was added to sera samples and incubated at 55⁰C for 90 min. To facilitate enzyme inactivation, the mixture was heated in a water bath for 15 min. PBS containing 0.5 ml/l Tween 20 buffer was used to was samples. The pretreated serum sample and antibodies specific for pentosidine were mixed and incubated at 37⁰C for 60 min. Polyclonal rabbit anti-human IgG peroxidase conjugated antibodies were then added and incubated at room temperature for 60 min. After color development was arrested as per kit instructions, absorbance was measured at 450 and 630 nm, respectively. Standard curve was prepared by measuring the same standard solution of pentosidine used to quantify pentosidine in sera samples [11].

Competitive ELISA:
A previously published competitive-binding assay was used to determine antigen-antibody interaction specificity [6, 10 & 11]. Hundred microliters of antigen (N-HSA or G-HSA) at a concentration of 5μg/ml were coated onto microplate's wells for 2-4 hours at room temperature. TBS-T was used to wash plates (3-5 x) after incubation. Skimmed milk (2 %) was used for blocking followed by incubation for 4-6 hours at room temperature. Micro titre plates were washed (3 x) with TBS-T. In serum test tube varying concentrations (0-20 μg/ml) of inhibitors (N-HSA or G-HSA) were mixed with identical concentrations of serum IgG (10 μg/ml) for 2-4 hours at room temperature. Serum IgG without antigen served as control. This mixture was added to antigencoated microplates. Residual antibody levels were detected at 410 nm using Accuris USA Absorbance Microplate Reader (MR9600-E).
Percent inhibition = 1 -(A inhibited / A uninhibited) X 100 Where, A inhibited is the absorbance at 20 µg/ml of inhibitor concentration and A uninhibited the absorbance without inhibitor.

Statistical evaluation:
Values are represented as arithmetic mean ± SD. Multiple comparisons were made by student t test using SPSS16 software program. Values of p< 0.05 were considered to be statistically significant.

Results:
Biophysical characterization of glycated HAS: Tryptophan fluorescence: Tryptophan fluorescence was quantified in both native and modified samples of HSA. The fluorescence intensity of a sole tryptophan residue located in HSA molecule was chosen to probe subtle structural and conformational changes. Analyses were conducted by using an excitation wavelength of 285 nm on samples (HSA or G-HSA). The emission maxima of native HSA (25.1 ± 3.1 AU) and G-HSA (60.9 ± 5.3 AU) samples were 330 and 320 nm, respectively. A blue shift of 10 nm for G-HSA (Table 1) was observed. The fluorescence intensity of G-HSA significantly (p < 0.001) increased. 11.5 ± 0.58 11.1 ± 0.47 (-3.5) β-turns 18.9 ± 0.76 19.3 ± 0.87 (+2.1) All the calculations were done using N-HSA as standard. Each sample was run three times and values represent arithmetic mean ± SD. Values in parenthesis represents percent change; + and -signs represent increase and decrease of values, respectively. *P< 0.05, **P < 0.01, ***P < 0.001.

Pentosidine specific fluorescence:
Pentosidine is a fluorescent AGE compound generated by the sequential glycation of proteins. Thus, detection of pentosidine using specific fluorescence is proof of in vitro glycation of HSA in our reaction samples. The optimum excitation wavelength for pentosidine (275 nm), was used on samples. A significantly high (p < 0.001) pentosidine-specific fluorescence (213.7 ± 13.4 AU) was observed for G-HSA. However, the fluorescence for native HSA (6.7 ± 0.7 AU) ( Table 1) was insignificant.

Circular Dichroism:
Glycation induced secondary structural changes in HSA (Figure 1) can be uncovered by CD. The CD signal of proteins in the spectral region of 200-250 nm, can be attributed to secondary structure. Native HSA and modified HSA exhibited markedly different CD spectra. This is indicative of substantial secondary structural modifications in the modified HSA sample. A dip at 212.5 nm observed in the spectrum for G-HSA signifies a stable β-sheet structure. Software based on Yang equation was used to calculate secondary structural differences between native and modified HSA samples [15]. Appreciable changes were observed in HSA with respect to α-helix, β-sheet, β-turns and random coil. A decrease in α-helix and random coil structures (5.6% and 3.5% respectively) was detected in modified HSA, compared to the native form. Where as β-sheet and β-turns both showed increases of 8.0% and 2.1% respectively, after glycation. These findings suggest that glycation induces structural changes, accompanied by a partial destruction of the secondary structure of HSA.

Biochemical analysis of serum samples: Carbonyl contents:
Carbonyl contents present in vivo is a biomarker for the high levels of free radicals i.e., oxidative stress. Estimation of oxidative stress levels in SLE and NH subject samples is given in Table 2. A significant increase in protein bound carbonyl contents (p < 0.001) in SLE subject sera as compared to NH subject sera was found. SLE patients exhibited higher quantities of (2.6 ± 0.6nmoles/mg protein) of carbonyl compound sin comparison to NH (1.4 ± 0.3 nmoles/mg protein) subjects ( Table 2).

Pentosidine levels in serum samples:
Pentosidine levels were detected in each serum sample obtained from SLE and NH subjects. Increased levels of pentosidine were found in subjects with SLE (0.0267 ± 0.0021μg/ml) compared to the pentosidine levels in NH subjects (0.0248 ± 0.0025μg/ml) ( Table 2).

FBG and HbA1C levels in serum samples:
For each sample, FBG was estimated by glucose oxidase method, and HbA1Cbycapillary electrophoresis method. Levels of FBG and HbA1Cwere slightly increased in SLE samples, compared to the samples from NH subjects ( Table 2). This increase may be attributed to inflammatory conditions in SLE patients.   (Figure 2).

Competition ELISA:
Specificity of serum autoantibodies was further investigated by competition ELISA. Native and modified HSA were used as competitors in the assay. Competition ELISA results showed high specificity between serum IgG from SLE and G-HSA, summarized in Table 3. The mean maximum percent inhibition (MMPI) was found to be 54.9± 2.9 and 11.9 ±2.1 for G-HSA and N-HSA, respectively in SLE subjects. However, IgG from control subject's sera showed very little reactivity with either of the antigens (10.7 ± 2.1 and 9.6 ±2.0 for G-HSA and N-HSA, respectively). Antigens (5µg/ml) were used to coat each microplate. Competitive ELISA of each serum IgG was done thrice, and the values are means ± SD Statistically significant percent competition with G-HSA (*** p<0.001) than N-HSA for SLE subjects.

Discussion:
Presence of autoantibodies, such as anti-nuclear antibodies (anti-dsDNA), is well-known markers for SLE [20]. This proves the abnormal functioning of the immune system. This immune imbalance leads to chronic inflammation and hence causes other secondary complications (nephropathy, atherosclerosis, and osteoarthritis). The search for a better prognostic potential marker for SLE diagnosis that could be used in early diagnosis and disease management is ongoing. Oxidative stress is involved in the immunopathogenesis of SLE [21]. A recent study showed excessive production of ROS, along with many other factors may induce SLE [22]. Increase in cell death and delay in clearance of these dead cells, as well as apoptotic cells are commonly associated with SLE induced excessive production of ROS . Thus, G-HSA was chosen as an antigen and serum autoantibodies were screened in SLE sera samples. Healthy individuals served as controls and were also screened for autoantibodies. Direct binding ELISA results from this study exhibited significantly increased levels of serum autoantibodies in SLE patients against G-HSA, whereas healthy individuals showed negligible binding. To further confirm that the autoantibodies in the SLE sera are specific to modified antigen (G-HSA), competitive ELISA assays were run for all SLE samples as well as in NH serum samples. Results showed that IgG from SLE patients exhibited strong recognition of G-HSA as compared to N-HSA. Negligible binding patterns were observed for NH serum IgG with both N-HSA and G-HSA. According to these immune assay findings and other clinical data, (carbonyl content, pentosidine, FBG and HbA1C) from SLE patients, is has been suggested that generation of higher amount of serum autoantibodies against G-HSA can be correlated with oxidative stress and chronic inflammation in SLE patients.

Conclusion:
Role of G-HSA in SLE immunopathogenesis was established in this study. G-HSA has been found to be immunogenic and produce increased amounts of autoantibodies in SLE patients. It can be suggested that glycation of blood proteins might cause imbalance in humoral immunity and may further lead to the generation of antigen-antibody immune complexes, in turn contributing to the pathogenesis of SLE disease.

Conflicts of Interest:
The authors have no conflicts of interest to declare.