Abstract
Background:
Frontotemporal dementia and parkinsonism-linked to chromosome-17 are a group of diseases with tau mutations leading to primary tauopathies which include progressive supranuclear palsy, corticobasal syndrome, and frontotemporal lobar degeneration. Alzheimer’s disease is a non-primary tauopathy, which displays tau neuropathology of excess tangle formation and accumulation. FTDP-17 mutations are responsible for early onset of AD, which can be attributed to compromised physiological functions due to the mutations. Tau is a microtubule-binding protein that secures the integrity of polymerized microtubules in neuronal cells. It malfunctions owing to various insults and stress conditions-like mutations and post-translational modifications.
Objective:
In this study, we modified the wild type and tau mutants by methyl glyoxal and thus studied whether glycation can enhance the aggregation of predisposed mutant tau.
Methods:
Tau glycation was studied by fluorescence assays, SDS-PAGE analysis, conformational evaluation, and transmission electron microscopy.
Results:
Our study suggests that FTDP-17 mutant P301 L leads to enhanced glycation-induced aggregation as well as advanced glycation end products formation. Glycation forms amorphous aggregates of tau and its mutants without altering its native conformation.
Conclusion:
The metabolic anomalies and genetic predisposition have found to accelerate tau-mediated neurodegeneration and prove detrimental for the early-onset of Alzheimer’s disease.
INTRODUCTION
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder affecting the cognitive functions. Protein misfolding and aggregation in the brain is the major cause for neuronal dysfunction in AD. Amyloid-β (Aβ) and tau proteins aggregate and accumulate causing neuronal death and brain lesions. Aβ pro-aggregant peptide Aβ1-42 forms senile plaques extracellularly. Recently, these senile plaques have been implicated in enhancing seeded tau aggregation in mouse brain. Tau aggregation and accumulation hampers neuronal functioning and microtubule stability. Tau is a natively unfolded protein and is involved in microtubule dynamics by association with the labile domain. Tau is a microtubule-associated protein playing a pivotal role in stabilizing the axonal microtubules [1, 2]. Tau protein is broadly divided into two major domains. The N-terminal projection domain consisting of two inserts and a polyproline rich region. The projection domain maintains the flexibility and the hydrophilicity of the protein. The C-terminal region consists of 4 imperfect repeats which function in binding to the microtubules and stabilizing them [3, 4]. Tau has 6 isoforms based on the alternative splicing of exon 2 and 10 [5]. Tau pathology in AD is sporadic as well as genetic. The genetic mutations leading to tau pathology belong to frontotemporal dementia group of mutations. Frontotemporal dementias are the group of disorders characterized by degeneration of cerebral cortex and sub-cortical regions of substantia nigra [6]. The initial reports of familial linkage of disease mutations to the chromosome locus 17q21-22 were published in 1994 for disinhibition-dementia-parkinsonism-amyotrophy complex (DDPAC) [7], which was subsequently grouped under frontotemporal dementia and parkinsonism-linked to chromosome-17 (FTDP-17) in 1996 [8]. FTDP-17 is an autosomal dominant disorder phenotypically characterized by behavioral, motor, and memory impairments [9–11]. Tau mutations are missense, deletion, or silent which either affects the protein coding or the splicing events depending on the locus [12, 13]. The mutations that affect the splicing sites alters the 3 repeat to 4 repeat tau ratios [14]. Moreover, the combination of tau isoforms and the FTDP-17 mutations affect the parameters like aggregation kinetics and aggregate length, thus, having important implication in the disease pathogenesis [15]. Tau missense mutations affect its functions in two ways. Some of these mutations reduce the affinity of tau to microtubules [16] and others render tau to increased aggregation forming neurofibrillary tangles in the brain [17]. The missense FTDP-17 mutations include P301 L, G272 V, V337M, K369I, and R406 W. The mutations G272 V, P301 L, G303 V, S320F, and S352 L, among others, demonstrate increased aggregation propensity compared to the wild type tau [18, 19]. Moreover, the ability to polymerize tau varies with the mutation. The mutations P301 L, G303 V, and S320F induce tubulin polymerization at a slower pace compared to wild type, whereas V337M and E342 V enhanced the rate of tubulin assembly [18]. The slight decrease in the affinity of tau to microtubules can completely hamper the stability of microtubules. Moreover, decreased affinity can render the increased cytoplasmic concentration of tau accelerating its self-assembly and aggregation [20]. Familial AD accounts for 2-5% of the total AD cases and is associated with the early onset of the disease, before 60–65 years of age [21, 22] which includes genetic mutations. In addition, one of the most important aspects of AD is the abnormal post-translational modifications (PTMs) of tau, which hampers its functions and increases the propensity of self-aggregation [2, 23]. Abnormal phosphorylation of tau in AD is studied in its complete depth and is one of the key player of tau pathology in AD [24, 25]. Another important tau PTM is glycation, which reduces tau’s affinity for microtubules [26–28]. Glycation affects the proteins by forming cross-links and rendering it protease resistant [29, 30]. It also leads to protein aggregation and accumulation [31, 32]. Few reports suggest that tau glycation might not trigger its fibrillization but may enhance the already initiated fibrillization process [33]. The effect of PTMs on the dementia mutants is still unexplored, whether it can enhance or modulate the tau pathology. Here, we report the effect of glycation on selected tau dementia mutants G272 V, P301 L, and R406 W (Fig. 1A). Glycation enhanced the aggregation of P301 L tau with subsequent increase in advanced glycation end products (AGEs). However, glycation did not alter the global conformation tau.

MG-induced aggregation of tau and FTDP-17 mutants. A) Tau protein showing distinct functional domains with two inserts (blue blocks) at the N-terminal. Polyproline rich region denoted by green rings followed by 4 imperfect repeats (red arrows) involved in microtubule-binding and also involved in tau aggregation. The tau missense mutations are denoted in red, G272 V, P301 L, and R406 W, which are involved in tau dysfunction. B) The polyol pathway showing the formation of reactive metabolites like glyoxal and methyl glyoxal (MG) that are precursors of advanced glycation end products. C) Structure of MG, a reactive aldehyde accumulating in aging brain. D) ThS fluorescence analysis for MG-induced aggregation showing enhanced aggregation of mutants G272 V and P301 L as compared to control. E) ThT fluorescence assay shows similar observation as ThS analysis wherein the mutants show enhanced aggregation propensity as compared to control. F) The hydrophobicity transitions probed by ANS fluorescence reveal no distinct change in wild-type tau versus mutants. G) The comparative analysis of all fluorophores show that R406 W shows suppressed aggregation in presence of MG as compared to wild-type whereas G272 V and P301 L follow the wild-type tau with increased aggregation by MG (p < 0.01).
MATERIALS AND METHODS
Chemicals and reagents
Methyl glyoxal, ThS, ThT, ANS, Glycine, MES, BES, and SDS were obtained from Sigma; Luria-Bertani broth (Himedia); Ampicillin, Heparin, NaCl, Phenylmethylsulfonylfluoride (PMSF), MgCl2, Sodium azide, APS, Ethanol (Mol Bio grade), were purchased from MP biomedicals; IPTG and Dithiothreitol (DTT) were from Calbiochem; EGTA, Tris base, Acrylamide, and TEMED were obtained from Invitrogen. Protease inhibitor cocktail was purchased from Roche. Methyl glyoxal should be handled with care. Any contact with skin and eyes should be avoided. Inhalation of vapors must be avoided. It is advised to wear gloves and safety goggles while handling the chemical.
Tau purification
Protein purification for full-length tau and the FTDP-17 mutants was carried out as previously described. Full-length recombinant tau and mutants were expressed in E. coli BL21* strain. The cell pellets were homogenized under high pressure (15,000 psi) in a microfluidics device (Constant Cell Disruptor). The obtained lysate was heated at 90°C for 15 min after addition of 0.5 M NaCl and 5 mM DTT. The heated lysate was then cooled and centrifuged at 40,000 rpm for 50 min in Optima XPN-100 ultracentrifuge (Beckman coulter). The supernatant was collected and dialyzed in Sepharose A buffer overnight. The obtained dialyzed sample was subjected to a second round of ultracentrifugation and the supernatant was loaded onto the cation exchange column (Sepharose fast flow GE healthcare) for further purification. The bound protein was eluted using an ionic gradient. The eluted fractions were pooled and concentrated for size exclusion chromatography (16/600 Superdex 75pg GE healthcare). The concentration of respective proteins was measured using BCA method.
MG-induced tau aggregation assay
20μM full-length tau and its FTDP-17 mutants were incubated in BES buffer pH 7.4 with 2.5 mM methyl glyoxal as a glycating agent. The extrinsic fluorophores were added to the reaction mixtures to a concentration of 2.5μM for thioflavin S and thioflavin T and 400μM for ANS respectively. The reaction mixture was supplemented with 25 mM NaCl, 0.01% sodium azide, protease inhibitor cocktail, and 1 mM DTT. The mixtures were made in opaque tubes to avoid fluorescence bleaching at 37°C. For measurement of extrinsic fluorophores (ThS, ThT, and ANS), samples were diluted to 5μM of tau in ammonium acetate (pH 7.0). The excitation/emission for the fluorophores was as follows: ThS 441/521, ThT 435/485, and ANS 375/490. The measurements were acquired in triplicates every 24 h. Background fluorescence was subtracted from the obtained readings. The fluorescence measurements were carried in TECAN Infinite series Pro plate reader.
Measurement of AGEs-specific fluorescence
The reaction mixture was set as mentioned above except the addition of the extrinsic fluorophores. For measurement of AGEs-specific fluorescence, the 50μL of 20μM reaction mixture was aliquoted in 384 black-well plate and reading was acquired at excitation/emission 370 nm/430 nm in TECAN Infinite series Pro plate reader. Buffer background was subtracted from the obtained readings.
SDS-PAGE
The glycated tau and mutants were analyzed on 10% SDS-PAGE at the interval of 24 h. 10μL of the reaction mixtures were analyzed at each time point. Gels were stained with 0.1% Coomassie brilliant blue.
Transmission electron microscopy (TEM)
The glycated proteins were visualized by TEM. 2μM of glycated proteins were applied to 400 mesh carbon coated copper grids for 1 min. The grids were rinsed with ultra-pure water and stained with 2% uranyl acetate for 2 min. Further the grids were dried and scanned in Tecnai G2 20 S-Twin transmission electron microscope.
CD spectroscopy
Glycation-induced conformational changes were mapped by CD spectroscopy. The modified and unmodified tau proteins were diluted to 3μM in 50 mM phosphate buffer pH 6.8 in a cuvette with a path length of 1 mm. The scanning was done using Jasco J-815 CD spectrometer under nitrogen atmosphere. The scan was carried out with following parameters; bandwidth 1 nm, scan speed 100 nm/min; scan range 190 to 250 nm, and an average of 5 acquisitions. All the scans were done at 25°C. The buffer baseline was set with phosphate buffer, pH 6.8.
Statistical analysis
The statistical analyses were carried out using unpaired T-test by SigmaPlot 10.2. The error bars represent mean±SD values. 95% confidence intervals were maintained for the analyses.
RESULTS AND DISCUSSION
Tau mutant P301 L shows enhanced MG-induced aggregation
Tau FTDP-17 mutations decrease the microtubule-binding affinity of tau [34] and also enhance the heparin-induced assembly of tau in vitro [35]. In our study, we investigated the glycation-induced aggregation of tau and its mutants. Glycation of proteins by sugars and their reactive intermediates, cross-links and aggregates proteins [36]. Methyl glyoxal (MG) is a highly reactive byproduct of glucose metabolism in glycolysis and polyol pathway (Fig. 1B) and a precursor of AGEs [37] (Fig. 1C). AD brains are reported to show excess accumulation of MG and AGEs [38–40]. In order to study MG-induced protein aggregation, tau and its FTDP-17 mutants G272 V, P301 L, and R406 W were subjected to MG treatment in vitro and aggregation kinetics was studied using three fluorescent probes. These extrinsic fluorophores have discreet binding affinities, which give information about varied folding states during the aggregation [41]. We employed three extrinsic fluorophores, Thioflavin S (ThS), Thioflavin T (ThT), and ANS, to study aggregation of tau and its mutants. ThS and ThT are structurally different. ThS has two quaternary nitrogen atoms. Both these dyes have similar binding sites. They bind to β-sheets formed during protein aggregation and fluorescence according to the extent of aggregation. ThS binds to mature fibrils and not to monomers and oligomers. ThS aggregation studies showed that MG enhanced the aggregation of the dementia mutants G272 V and P301 L as compared to wild-type tau (Fig. 1D). R406 W tau had a very low ThS signal intensity suggesting decreased MG-induced aggregation as compared to wild-type control (Fig. 1D). ThT kinetic studies followed the similar trend with P301 L showing highest aggregation propensity and R406 showing the least (Fig. 1E). ThS and ThT followed the similar pattern of kinetics for glycation-induced aggregation. Further, we checked the effect of MG-induced hydrophobicity changes in tau and its mutants by ANS dye. ANS binds to hydrophobic regions of aggregated proteins, which are exposed during the transition of monomers to polymers [42]. These species are transient and might be short lived and hence not captured by ThS and ThT. ANS assay did not show any difference between the tau wild-type and its mutants suggesting that there is no difference of hydrophobicity of proteins while transitioning from monomers to polymers (Fig. 1F).
One of the recent finding suggests that the FTDP-17 mutants enhance oligomer formation under heparin induction especially the P301 L mutant show small and more granular oligomers as compared to wild-type control [43]. But glycation-induced aggregation did not show enhanced oligomerization by ANS fluorescence. Thus, the comparative analyses of tau and its mutants for the extrinsic fluorophores suggest that P301 L and G272 V show more glycation-induced aggregation as compared to control (Fig. 1G). Glycation enhances aggregation propensity of tau mutants (P301 L and G272 V) as compared to wild-type tau.
P301L tau mutant shows more AGEs formation on MG induction
Glycation of proteins leads to formation of brown and fluorescent protein adducts, which can be quantitated by AGEs-specific fluorescence. AGE-specific fluorescence was initially utilized to study the extent of diabetic complications [44–46]. Further this technique was used as a non-invasive tool to determine the skin fluorescence in diabetic patients [47]. AGE-specific fluorescence is detected at the excitation/emission 300-420/420-600 nm. We hypothesized whether the tau mutations would enhance AGEs formation or suppress it as compared to the wild-type control (Fig. 2A). The glycation induced aggregation studies revealed enhanced aggregation of P301 L mutant over the wild type. Further, the AGEs formation of tau and its mutants was investigated in the presence of MG by the AGEs-specific fluorescence. The tau mutant P301 L showed increased fluorescence as compared to control. G272 V and R406 W mutants showed decreased AGEs-specific fluorescence than wild-type (Fig. 2B). The comparative analysis of AGEs-specific fluorescence at 168 h revealed that P301 L showed enhanced AGEs formation as compared to control (Fig. 2C). Although tau glycation for FTDP-17 mutants is unexplored, MG-induced glycation has been found to accelerate tau aggregation in presence of hyperphosphorylation [48]. Thus, dual modification might have an additional impact on the tau pathology. The mutant R406 W showed least AGEs fluorescence as compared to other mutants. This mutant also exhibited least MG-induced aggregation, which might be attributed to the local conformational changes caused by this mutation. Similar reports have been recently published wherein the tau R406 W mutant resists phosphorylation by various kinase [49]. A transgenic experimental model of diabetes mellitus expressing P301 L tau showed increased tau hyperphosphorylation and tangle deposition [50]. Our in vitro glycation data suggests P301 L mutant showed enhanced AGEs formation, which supports the in vivo findings. Thus, this mutation might play an important role in accelerating brain pathology in early onset of AD.

Advanced glycation end products formation propensity of tau and its mutants. A) Hypothetical model to check the effect of glycation on AGEs formation. The mutants can either demonstrate an increased or decreased propensity for AGEs formation as compared to wild type tau that is quantitated by AGEs-specific fluorescence. B) The AGEs-specific fluorescence reveals increase in intensity with time for tau and its mutants. P301 L mutant shows highest AGEs formation as compared to control. C) The comparison of last time point fluorescence intensity reveals that the mutant P301 L shows more propensity for glycation as compared to wild-type tau (p < 0.05).
Protein glycation forms protein cross-linking and aggregation [51] which accumulates steadily over time due to slow clearance of these proteins as they are resistant to most proteases. Moreover, glycation renders more stability [52] to the proteins leading to bulk deposition over the years [53, 54]. In order to study the nature of these aggregates SDS-PAGE analysis was performed at 24-h time intervals. At 0 time point, we observed a single soluble protein band for tau and its FTDP-17 mutants. As the time progressed, the soluble protein was no longer intact and had formed higher order aggregates (Fig. 3A). The aggregates were SDS-resistant and migrated on the separating gel in a continuous trail of cross-linked proteins. The extent of glycation differed between the mutants and the wild type. Mutant P301 L had enhanced glycated products on the SDS-PAGE as compared to control. G272 V and R406 W mutants showed decreased glycated proteins on SDS-PAGE as compared to control. Thus, the FTDP-17 mutant P301 L showed enhanced glycation as compared to control. The heparin induced aggregation of tau protein leads to formation of the fibrillar aggregates as visualized by TEM [55]. We visualized the glycation-induced tau aggregates by TEM for tau and its mutants. Glycation led to the formation of amorphous aggregates completely discreet from the heparin-induced fibrillar aggregates (Fig. 3B). The morphology did not alter for tau and its mutants. Similar observation has been previously made for a 2 repeat tau peptide which forms dimers on glycation but does form filamentous aggregates [56].

SDS-PAGE and morphological evaluation of glycated tau proteins. A) SDS-PAGE evaluation of glycated tau shows formation of SDS-resistant glycated proteins, which aggregate and can be visualized as a trail on the separating gel at higher molecular weight than the soluble protein. B) Electron micrographs of the glycated tau and its mutants reveal formation of amorphous aggregates.
Glycation does not alter the conformation of tau and its mutants
The modification of proteins in AGEs alters the structure and function of the proteins rendering it prone to aggregation and protease resistant. [31, 57]. Tau is natively random coil lacking a rigid structure. On aggregation, it adopts a partial β-sheet structure [58, 59]. In order to study the effect of glycation on tau and its mutants, we studied the global conformation of glycated and non-glycated soluble tau and its mutants by CD spectroscopy. The conformational analysis of soluble proteins revealed a typical random coil structure (Fig. 4A). The maximum ellipticity was observed at around 198 nm for all the proteins suggesting a complete random coil structure (Fig. 4A zoom). The glycated proteins also showed CD spectra of random coil conformation (Fig. 4B). The maximum ellipticity was unaltered at 198 nm, same as that of soluble protein. Thus, glycation does not alter tau native conformation. This is in conjunction with previous studies wherein glycation can induce the 3D structural changes differently for different proteins [52, 60].

Effect of glycation on global conformation of tau proteins. A) The native conformation of tau and its FTDP-17 mutants show native random coil conformation and maximum ellipticity at 198 nm. B) Glycation does not alter the global conformation of tau and its mutants. Glycated proteins show native conformation with maximum ellipticity at 198 nm.
The connection between AD and glycation have been suggested in the earlier reports [61]. Moreover, tau pathology is enhanced due to insulin-mediated hyperphosphorylation suggesting a link between diabetes and tauopathy [62–64]. Hyperphosphorylated tau is known to undergo enhanced glycation [48]. Thus, metabolic anomalies and genetic predisposition have found to accelerate tau-mediated neurodegeneration and prove detrimental for the early-onset of AD (Fig. 5).

Enhanced MG-induced glycation of tau P301 L. The glycating agent MG show enhanced glycation for the tau dementia mutant P301 L with large aggregate formation as compared to wild-type. This might be due to differentially exposed amino acids in WT versus the P301 L mutant that leads to enhanced modification by MG. The other mutant G272 V showed basal level of glycation as control whereas R406 W mutant showed less glycation as compared to control. This model shows additional pathological feature of a predisposed tau dementia mutant, which might enhance tau dysfunction in a combinatorial manner.
ACKNOWLEDGMENTS
This project is supported in part by grants from the Department of Biotechnology from Neuroscience Task Force (Medical Biotechnology-Human Development & Disease Biology (DBT-HDDB))-BT/PR/19562/MED/122/13/2016 and in-house CSIR-National Chemical Laboratory grant MLP029526. SS acknowledges DBT for the fellowship.
Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/19-1348r2).
