Identification of the Al-binding proteins that account for aluminum neurotoxicity and transport in vivo

Abstract

Studies have shown that aluminum (Al) is the most abundant neurotoxic element on Earth, and is implicated in the pathogenesis of Alzheimer's disease (AD). However, the mechanisms underlying Al-induced neurotoxicity are still largely elusive. Based on affinity analyses with Al and LC-LTQ-MS, we have found that serum albumin, brain CK-B and 14-3-3ζ protein have a high affinity for Al³⁺, and albumin has a much stronger affinity for Al than transferrin. The normal activity of CK-B, and physiological combination of 14-3-3ζ with tau can be severely perturbed by Al. We anticipate that our assay will provide a new focus concerning the mechanism underlying Al-induced neurotoxicity, and aid the design of strategies to prevent AD and other human diseases related to Al overload.

---

1. Introduction

The roles of aluminum (Al) in neurodegenerative diseases, such as Alzheimer's disease, have been controversial for decades. Primary cultures of human brain endothelial cells have been found to have an extremely high affinity for Al when compared to other types of brain cells.¹ Studies on elderly subjects, investigating the association between exposure to Al in drinking water and Alzheimer's disease, showed that cognitive decline was greater in subjects with a higher daily Al intake, confirming that high consumption of Al may be a risk factor for Alzheimer's disease.^2,3 Nevertheless, some critical details in the mechanisms behind Al neurotoxic effects are still unclear. Although Al accumulation in the brains of AD patients has been observed,⁴ its transport mechanism still remains controversial. Experimental reports have suggested that at equilibrium, 80–90% of total Al in the plasma is carried by transferrin;⁵ however, Al does not replace iron ions under physiological conditions, since the binding affinity of iron to transferrin is about 100 times greater than that of Al,⁶ and other proteins that can perform Al transport in serum may be ignored.

Several hypotheses regarding the role of Al in the development of AD have been proposed, including Al facilitating iron-mediated oxidative injury in neurons by co-deposition of abeta peptides with iron and Al,⁷ causing the conformational change of abeta peptides into a beta-sheet structure, and/or promoting abeta peptide aggregation and accelerating the formation of amyloid by interacting with their acidic groups.^8,9 All of these and the other hypotheses can only partly elucidate the mechanisms behind Al neurotoxicity.^10–12 Indeed, little information has been revealed about whether Al is involved in abnormal phosphorylation of tau, which is now recognized as one of the key features of neurofibrillary tangles (NFTs) in AD.^13,14

Understanding the biomolecule interactions of metals with proteins has been a very active research area during the last two decades and continues to be so.^15,16 Considering that numerous toxic effects have been attributed to Al³⁺ and that the chemistry of A1³⁺ shows a strong preference for binding to a wide variety of proteins, much interest has been stimulated by the potential perturbation of protein function by A1³⁺.¹⁷ Al has been shown to enter and permanently occupy binding sites, which in healthy systems are served by other metal cations with specific binding and charge properties.

As a classic separation method, the native polyacrylamide gel electrophoresis (native-PAGE) method is a powerful technique for isolation, separation, and detection of protein complexes in the brain.¹⁸ Under non-reducing conditions, protein complexes are maintained in their native state, which is the key feature of native-PAGE, and proteins are resolved in the gel according to their native molecular masses, net surface charges, and molecular shapes.¹⁹ Our previous studies have shown that the native-PAGE procedure is suitable for the routine separation and quantification of four main whey proteins in different heat-treated bovine milks.²⁰ On the other hand, the chelating agent 8-hydroxyquinoline (8-HQ) is one of the most sensitive organic ligands used for the determination of A1³⁺ by fluorimetric detection.²¹ Researchers have utilized the fluorescent complex formed between Al and 8-HQ for the analysis of Al in seawater, foodstuffs,²² and gastric mucosa.²³

In the present study, we have detected Al-complexed proteins in rat brains and serum using a novel native-PAGE staining procedure, which relies on the reaction of Al with 8-HQ. Identification of proteins with high affinity for Al should help to comprehensively characterize the mechanisms underlying Al neurotoxicity, and aid the design of strategies to prevent AD.

2. Materials and methods

Research was conducted under an Institutional Animal Care and Use Committee approved protocol in compliance with the Animal Management Rules of the Ministry of Health of the People's Republic of China relating to animals and experiments involving animals. Human serum from healthy volunteers (6 males and 6 females) was collected to identify Al-binding protein in human serum.

2.1. Chemicals

Aluminum chloride, Tris, acrylamide, Triton X100, PMSF, Coomassie Brilliant Blue R-250 and 8-hydroxyquinoline were bought from Sigma-Aldrich Chemical Co. (United States). All other chemicals of the highest grades were obtained commercially.

2.2. Animals and treatment schedule

Wistar rats (seven-week-old males) were purchased from Animal Department of Academy of Military Medical Sciences (China). All the rats were housed in a temperature-controlled room (25 ± 2 °C) at a relative humidity of 60 ± 5% with a 12 h dark/light cycle were and allowed free access to food and water for 7 days before the experiments. They were randomly divided into two groups (six rats each group), as follows.

Group 1 served as an untreated control; each of the rats received deionized water and a normal chow diet. Group 2 experienced subchronic exposure to Al; each of the rats received deionized water and 171.8 mg Al per kg per day (1/5 LD₅₀) in a formulated diet.²⁴ The study was carried out for a period of 10 weeks.

2.3. Protein extraction

Human serum, rat serum or brain tissue was homogenized in a buffer (1 mM PMSF, 0.04 M Tris-HCl, pH 7.8) (1 : 8), centrifuged (10 000g at 4 °C for 15 min) and the supernatant was used as a protein sample for polyacrylamide gel electrophoresis (PAGE) and subsequent analysis.

2.4. Native-PAGE

Native-PAGE was performed as previously reported by our laboratory.²⁰ The serum and brain regions of the control group rats were homogenized in extracting buffer (1 mM PMSF, 0.04 M Tris-HCl, pH 7.8) (1 : 8), the homogenate was centrifuged (10 000g at 4 °C for 15 min) and the supernatant was used for the native-PAGE electrophoresis. Protein samples were mixed with the non-reducing loading buffer (0.1 M Tris-HCl, pH 6.8; 20% (v/v) glycerol, 0.01% bromophenol blue), and 10 μL of mixed sample was loaded in each lane of 9% native-PAGE (9% separating gel and 4% stacking gel).

2.5. 8-Hydroxyquinoline staining procedure and imaging

After the native-PAGE procedure, an aluminum chloride solution (30 mL, 4 mM AlCl₃, 60 mM Tris-HCl, pH 7.5) was prepared, added to the gel and allowed to shake at 40 °C for 10 min (formation of protein-Al complex). After the gel was washed with purified water three times for 15 min, 25 ml of 8-hydroxyquinoline (0.4 g L⁻¹ 8-HQ, 0.25 M Tris-HAc, pH 7.2) was added to the gel and the mixture was shaken at 40 °C for 10 min. Then the gel was washed three times with purified water for 10 min. Proteins were visualized by placing the gel on a UV Analyser (UV-IU, BinDa, Beijing), irradiating the gel for 10 seconds, and then the protein bands became visible as green bands against a black background of the gel matrix. A camera from Fuji-shi (Fine Pix J25) was used to take photographs of the gel. To optimize the staining effects, a series of optimization studies was carried out.

2.6. SDS-PAGE

Individual fluorescent protein bands of native-PAGE were confirmed by SDS-PAGE under the same conditions. After 8-hydroxyquinoline staining and imaging, native-PAGE protein bands were stained in 0.1% (w/v) R250 Coomassie Brilliant Blue and destained in a solution of 7.5% (v/v) ethanol and 7.5% (v/v) acetic acid. Fluorescent protein bands were excised horizontally, washed with deionized water three times for 10 min to remove ethanol and acetic acid, and then vibrated with 1% SDS for 30 min before SDS-PAGE. Protein samples with the sample buffer (0.1 M Tris-HCl, pH 6.8; 20% (v/v) glycerol, 2% (v/v) SDS, 5% (v/v) β-mercaptoethanol and 0.01% bromophenol blue) in a 1 : 1 ratio (v/v) were heated in boiling water for 5 min, and then centrifuged at 10 000g for 5 min. The band and supernatant were run on SDS-PAGE. After R250 Coomassie Brilliant Blue staining and destaining, photographs were taken of the gel.

2.7. Gel digestion, liquid chromatography and mass spectrometry

Gel trypsin digestion was carried out using a standard protocol according to Sang et al.²⁵ Gel bands were excised into approximately 1 × 1 mm cubes, washed with 40% acetonitrile/50 mM NH₄HCO₃, and dried under vacuum. 7 μL trypsin (15 ng μL⁻¹) was applied to each gel piece to improve cleavage efficiency, and digestion was carried out for 18 h at 37 °C. The digests were then extracted with 70/25/5% acetonitrile/H₂O/formic acid, speed vacuumed to dryness, and stored at −80 °C for subsequent analysis.

Nano-LC-ESI-MS/MS experiments were performed on LTQ-FT mass spectrometer (Thermo) equipped with a nanospray source, and an Eksigent NanoLC-2D HPLC system. Prior to mass spectrometry, tryptic digests were separated on-line by reversed-phase nano-liquid chromatography (20 μL sample) with a Dionex C18 column (75 μm I.D. × 150 mm, 5 μm, 300 Å, Thermo Fisher). Solvent A was 2% acetonitrile and 0.1% FA in water, and solvent B was 80% acetonitrile and 0.1% FA in water. There was a gradient from 0% B to 40% B at 80 min, to 100% B at 95 min, to 100% B at 115 min, to 100% A at 120 min, and to 100% A at 130 min.

When performing mass spectrometry, the temperature of the ion transfer tube was set at 200 °C, and the spray voltage was set at 1.8 V. One FT full MS scan (mass range, m/z 400–2000) was followed by three data-dependent LTQ MS/MS scans on the three most intense ions. The normalized collision energy was 35% for the MS/MS scan. The dynamical excluding time was 15 seconds.²⁶

2.8. Data analysis and informatics

For data analysis, all MS/MS spectra were converted into the DTA format from the experiment RAW file using Bioworks (Thermo Fisher Scientific), and were then merged into a single file for Mascot (ver. 2.2, Matrix Science, Boston, MA, USA) MS/MS ion search. The precursor ion error tolerance and fragment ion error tolerance in MS/MS were set at 10 ppm and 0.6 Da, respectively. Then, the data were searched against the rat IPI protein database or SwissProt database.²⁷

2.9. Analyzing effect of iron on affinity of Al with proteins

The gel slice containing the protein sample separated by native-PAGE was incubated with 1 mM AlCl₃ plus FeCl₃ at the levels indicated at 40 °C for 10 min, and was then stained with 8-hydroxyquinoline and visualized under UV as described above.

2.10. Creatine kinase activity assay

Activity analysis of creatine kinase (CK) was performed using a colorimetric method according to Hughes,²⁸ utilizing CK to catalyze the conversion of phosphocreatine and ADP to ATP and creatine.

In this study, Al at final concentrations of 0, 5, 7.5, 15 and 20 mM was introduced directly into the control group brain homogenates (0.25 mL, 300 μg protein) in buffer A (0.04 M Tris-HCl, pH 7.8, 150 mM NaCl, 0.25% sodium deoxycholate, 1 mM PMSF, pepstatin). The mixed homogenates (250 μL) were allowed to shake at 37 °C for 2 h. The supernatant fractions of the homogenates were collected by centrifugation (10 000g, 10 min) for CK activity assay.

For the membrane-bound CK activity assay, the centrifugal precipitate of homogenate from the control group or Al-treated group brain tissue was washed with 1 ml of phosphate buffer (10 mM, pH 6.8 in 10 mM KCl), incubated with 40 mM sodium phosphate, 80 mM sodium phosphate, 0.1% Triton X-100, or 0.3% Triton X-100 at 37 °C for 30 min. After centrifugation (10 000g at 4 °C for 30 min), the supernatant was used for the membrane-bound CK activity assay.

2.11. Assay of Al-effect on affinity of 14-3-3ζ protein with tau

The soluble proteins (300 μg) of the rat brains were incubated with 0, 0.1, 0.5, 1, 2 or 5 mM AlCl₃, in a 250 μL buffer (0.04 M Tris-HCl, pH 7.8, 150 mM NaCl, 0.25% sodium deoxycholate, 1 mM PMSF, 1 μM pepstatin) at 37 °C for 2 h, and then centrifuged (10 000g at 4 °C for 10 min) to remove the precipitate of Al-protein complex. Thereafter, each of the supernatants with 5 μg polyclonal antibody against 14-3-3ζ protein (ab32622, Abcam) was incubated at 4 °C for 8 h under shaking, and then incubated with 30 μL PAAB (protein A-agarose beads from Abmart, 50% slurry, washed extensively) to isolate CT14A (complexes of tau-14-3-3ζ protein conjugated on the antibodies). The PAAB-CT14A, which had been centrifugally precipitated (3000g, 10 min), was collected, separated by SDS-PAGE, transferred onto a nitrocellulose membrane and analyzed by western blotting with antibodies against tau (Ab64193, Abcam).

2.12. Statistcal analysis

The results were expressed as the mean ± standard deviation (SD) of triplicate using SPSS 17.0 for Windows. The statistical significance of the data was determined using one-way analysis of variance (ANOVA) followed by Dunnett's contrast. A P value of <0.05 was regarded as significant.

3. Results

3.1. Feasibility analysis and optimization of the 8-hydroxyquinoline (8-HQ) staining conditions

According to the proposed framework, a comparison test was performed to check whether the 8-hydroxyquinoline (8-HQ) staining method is technically feasible. At first, no fluorescent protein bands were found in the general staining procedure without the stage of adding AlCl₃ to the gel (Fig. 1). This demonstrated that the fluorescence of the protein bands was not produced by the protein-HQ complex, but was produced by the protein-Al-HQ complex (Fig. 1). A comparison of the 8-HQ staining method (Fig. 1) with the Coomassie brilliant blue (CBB) method (Fig. 1) shows that 8-HQ staining allowed the proteins to be separated and visualized clearly, and highlighted specific interactions between A1³⁺ and the cerebral proteins.

Fig. 1 Feasibility analysis of the 8-hydroxyquinoline staining procedure. Brain proteins from normal rats were loaded in lanes 1, 2 and 3, separated by 9% native-PAGE. Lane 1: Coomassie brilliant blue method; lane 2: 8-HQ staining procedure; lane 3: the general 8-HQ staining procedure without the Al staining stage.

The influences of different post-electrophoresis staining conditions (temperature of the reaction, concentration of 8-HQ and pH of the 8-HQ staining procedure) were studied according to the general procedure given above, to obtain a maximum reaction yield of the protein-Al-HQ complex and to achieve effective separation and visualization of the Al binding proteins. The effect of temperature (range: 4–50 °C) on the protein-Al complex formation was investigated. As can be seen, the optimum temperature of the stage of formation of the protein-Al complex was 40 °C. The optimum concentration of 8-HQ was found to be 0.4 g L⁻¹ in Tris-HAc buffer solution. In order to evaluate the influence of pH on the formation of the protein-Al-HQ fluorescence complex, a set of tests was carried out by varying the pH using 0.25 M Tris-HAc buffer solution, covering the pH range between 6.4 and 8.8. Based on the results observed, the optimum pH value of 8-HQ to use was 7.2.

3.2. Identification of proteins with high affinity for aluminum

A fluorescent band (Al-RS) could be clearly seen on the gel containing serum proteins that had been incubated with AlCl₃ (Fig. 2A). A similar fluorescent band (Al-HS) was also observed in the human serum sample tested. The proteins of the fluorescent bands (Al-RS, Al-HS) in native gel were isolated and re-separated by SDS-PAGE to be submitted for LC-MS/MS analysis. The LC-MS/MS data showed that the Al-RS or Al-HS was highly homologous to the rat or human serum albumin protein (Table 1).

Fig. 2 Identification of serum proteins with high affinity for aluminum. (A) Serum proteins (lanes 1 and 2 from normal rat, lane 3 from human) were separated by native-PAGE, stained with Coomassie brilliant blue (lane 1), or incubated with 1 mM AlCl₃ (lanes 2 and 3) and then stained with 8-HQ (0.4 g L⁻¹, pH 7.2), Tf: transferrin. (B) Lane 1, normal rat serum protein (15 μg) separated by SDS-PAGE, Albu: albumin; lane 2, protein in the fluorescent band of rat serum in the native gel (A) was isolated and re-separated by SDS-PAGE; lane 3, the Tf protein isolated from the native gel (A) and re-separated by SDS-PAGE. (C) The gel slice of native-PAGE containing rat serum proteins (15 μg) was incubated with 0, 0.1, 1, 4 or 8 mM of AlCl₃, stained with 8-HQ (0.4 g L⁻¹, pH 7.2), photographed on a UV analyzer; the level of fluorescent protein was quantified by densitometry. (D) The gel slice of native-PAGE containing rat serum proteins (15 μg) was incubated with 1 mM AlCl₃ plus 0, 0.1, 10, 20 or 40 mM of FeCl₃, then stained with 8-HQ (0.4 g L⁻¹, pH 7.2), photographed on a UV analyzer; the level of fluorescent protein was quantified by densitometry.

Table 1 LC-MS/MS analysis of protein bands

Band	Database	Accession number	Protein descriptions^a	M, Da	Sequence coverage, %
a The tryptic peptides analyzed by LC-MS/MS were matched with the amino acid sequences available on the database.
Al-B1	IPI	IPI00470288	Creatine kinase B-type	42 983	76
Al-B2	SwissProt	1433Z_RAT	14-3-3 protein zeta	27 925	68
Al-RS	SwissProt	ALBU_RAT	Serum albumin	70 682	60
Al-HS	SwissProt	ALBU_HUMAN	Serum albumin	71 317	50
Tf	SwissProt	TRFE_RAT	Serotransferrin	78 512	45

---

We found that serum albumin has a much stronger affinity for A1³⁺ than the other serum proteins, including transferrin (Fig. 2B and Table 1). Therefore, albumin may play a major role in transporting A1³⁺ in the blood.

The fluorescence intensity of the serum proteins separated on gel was positively correlated with the AlCl₃ concentration up to 8 mM (Fig. 2C). The affinity of Al³⁺ for albumin was remarkably reduced by 10 μM or higher levels of Fe³⁺in vitro (Fig. 2D).

There were two fluorescent bands (Al-B1, Al-B2) found in the gel that contained rat brain proteins and had been incubated with AlCl₃ (Fig. 3A). LC-MS/MS analysis showed that the Al-B1 in rat brain had achieved 76% sequence coverage for creatine kinase b-type (CK-B, IPI00470288), with 289 of 381 amino acids sequenced, while the Al-B2 in rat brain had achieved 68% sequence coverage of 14-3-3ζ protein (1433Z_RAT), with 167 of 245 amino acids sequenced (Table 1). The affinity of Al³⁺ for CK-B or 14-3-3ζ, like serum albumin, was remarkably reduced by 0.1 mM or higher levels of Fe³⁺in vitro (Fig. 3B).

Fig. 3 Identification of brain proteins with high affinity for aluminum. (A) Lane 1, normal rat brain proteins were separated by native-PAGE, incubated with 1 mM AlCl₃ and then stained with 8-HQ (0.4 g L⁻¹, pH 7.2); lane 2, rat brain proteins separated by SDS-PAGE; lane 3 and 4, proteins in the fluorescent band of rat brain in the native gel (lane 1) were isolated and re-separated by SDS-PAGE. (B) Normal rat brain protein was separated by native-PAGE, incubated with 1 mM AlCl₃ plus 0, 0.1, 10, 20 or 40 mM of FeCl₃, then stained with 8-HQ (0.4 g L⁻¹, pH 7.2), photographed on a UV analyzer; the level of fluorescent protein was quantified by densitometry.

Furthermore, as shown in Fig. 4A, the activity of CK-B from normal rats could be inhibited by Al³⁺in vitro (Fig. 4A). Meanwhile in Al-treated rats in vivo, CK-B activity recovered from the precipitate of rat brain extraction was significantly higher than that of the control (Fig. 4B).

Fig. 4 Effect of AlCl₃ on the CK activities and hydrophobic character in the rat brain. (A) CK activities in the rat brain protein extract plus 0, 20, 40, 60, 80 or 100 mM of AlCl₃ was measured, respectively. (B) CK activities in the extract with Phosphate Buffer (PBS) or Triton X-100 (TX) from centrifugal precipitate of brain tissue homogenate of rats treated with Al or without Al (control group). Data are expressed as mean ± SD (n = 6).

Controlling microtubule structure and function in a normal brain depends on tau being phosphorylated properly.^1,14,29 It has been considered that 14-3-3ζ possesses chaperone-like activity and can prevent aggregation of some proteins,³⁰ such as tau proteins.¹⁸ To test how Al may perturb the interaction of 14-3-3ζ with tau, 14-3-3ζ-antibody coupled on APAB (protein A-agarose beads) was incubated with protein extract from rat brain in a buffer containing AlCl₃ (Fig. 5). Thereafter, tau levels absorbed on the complex of 14-3-3ζ-antibody-APAB were analyzed by western blotting with antibodies against tau protein. Compared to the control, the tau absorbed on 14-3-3ζ-antibody-APAB was reduced by 30% and 86.5% or more by 0.5 mM and 0.75 mM or more of AlCl₃, respectively. This result suggests that A1³⁺ can strongly reduce the affinity between tau and 14-3-3ζ, and therefore should promote tau aggregation and Alzheimer's disease.

Fig. 5 Effect of AlCl₃ on the affinity of 14-3-3ζ for tau in vitro. The protein extract (300 μg) from normal rat brain was incubated with 0, 0.1, 0.5, 0.75, 2 or 5 mM of AlCl₃ in a 250 μL buffer at 37 °C for 2 h, then centrifuged (10 000g, 10 min); each of the supernatants was incubated with 5 μg polyclonal antibody against 14-3-3ζ protein under shaking at 4 °C for 8 h, then incubated with 30 μL PAAB (protein A-agarose beads, 50% slurry) to isolate the CT14A (complexes of tau-14-3-3ζ protein conjugated on the antibodies). The PAAB-CT14A was centrifugally precipitated (3000g, 10 min), collected and separated by SDS-PAGE, transferred to a nitrocellulose membrane and analyzed by western blotting with antibodies against tau protein.

4. Discussion

Visualizing proteins on polyacrylamide gels, by utilizing the interaction of metal with proteins, is widely used, in processes such as silver staining. Carano et al.³¹ used copper staining combined with scanning electrochemical microscopy to visualize proteins on polymeric membranes. Meanwhile, techniques to visualize proteins fluorescently in polyacrylamide gels have been developed for protein detection, but the purpose of these techniques is to visualize all the proteins on polyacrylamide gels.³²

Luminescent chemical sensors based on increases in fluorescence in the presence of metals are particularly attractive because of their ease of use, and high sensitivity, even at low metal concentrations.³³ As is well known, 8-hydroxyquinoline (8-HQ), as a desirable fluorophore and binding moiety, is an excellent luminescent material that has found much use in the preparation of a variety of chemosensors for the detection of metal cations such as Zn(II),³⁴ Mg(II),³⁵ Cu(II)³⁶ and Hg(II).³⁷ As a matter of fact, a few reports have been published regarding a HQ derivative as a good fluorophore and chromophore for the simultaneous detection of Al³⁺.³⁸ Buratti et al.³⁹ reported that the fluorescence intensity of the complexation of Al(III) with 8-HQ (excitation wavelength, 380 nm; emission wavelength, 504 nm) remains unchanged for over 48 h at room temperature.

Our study aimed to utilize both the interaction of Al with proteins and the reaction of Al with 8-HQ to separate and visualize the specific Al binding proteins in gel. The cerebral proteins were separated by native polyacrylamide gel electrophoresis (native-PAGE), which has been used successfully for the separation and analysis of soluble protein complexes.⁴⁰ After the native-PAGE procedure, the prepared AlCl₃ solution was added to gel and Al³⁺ ions bound to the proteins. Then, a great amount of bound Al³⁺ ions were washed out of the gel with purified water, and only the strongly bound part was reserved and reacted with 8-HQ. Therefore, the protein in the fluorescent band must have had a much greater affinity for binding with A1³⁺. The 8-HQ staining method is technically feasible and fluorescent protein bands of native-PAGE were clearly observed after a series of optimization studies.

At physiological pH, aluminum ions normally undergo hydrolysis and form the insoluble species Al(OH)₃. The presence of coordinating ligands in blood can, however, completely inhibit hydrolysis, so that aluminum in blood is complexed with biological molecules.⁴¹ Aluminum binds to the iron-binding sites on the transferrin molecule, but under physiological conditions, it does not replace iron ions, since the strength of the aluminum bond is much weaker than that of iron. Fatemi et al.⁴² using difference UV spectroscopy to monitor the degree of saturation of the transferrin (Tf), reported direct competition for Al between Tf and albumin. They concluded that albumin was in fact a stronger Al chelator than citrate, and estimated that it should carry about 34% of the Al in serum. The ²⁷Al NMR results indicate that the Al³⁺ is bound to six oxygen ligands in an octahedral arrangement, which has led to the suggestion that Al might be binding at Ca²⁺ binding sites.⁴³ In our study, based on the 8-HQ staining method, the serum albumin has shown a specific binding affinity of A1³⁺ among the serum proteins. This result confirms the earlier conclusion by Fatemi and implies that protein albumin may play an important role in the transport mechanism of Al.

Creatine kinase (CK) is a key enzyme in energy metabolism and CK reaction has a much higher maximal rate of ATP synthesis than oxidative phosphorylation. In the brain, the dimeric cytosolic form of CK is called brain-type CK (CK-BB). Since CK-BB has been shown to play a fundamental role in the cellular energetics of the brain, any disturbance of this enzyme may exasperate the Alzheimer's disease (AD) process.⁴⁴ It has been reported in the literature that CK in the AD brain is modified in such a way that it loses activity.⁴⁵ However, long-term Al intake decreasing CK activity in the rat brain has not been reported. In our study, it is clear that the CK-B protein has a specific binding affinity for A1³⁺, so we further performed in vivo and in vitro studies to gain a better understanding of the effect of Al on CK activity. Decreased CK activity was found in brain homogenates of the Al-treated rats compared to the control. An inactivating effect was also observed in vitro, when the aluminum compounds were added directly to rat brain homogenates. In the case of enzymes (such as ATP⁴⁻-utilizing kinases and hexokinase), Al³⁺ most often binds at an essential metal ion binding site in the protein.¹⁷ This is clearly related to the fact that CK requires Mg²⁺ for catalyzing the reversible exchange of phosphate between phosphocreatine (PCr) and ADP, or creatine and ATP.⁴⁶ A1³⁺ is known to be an effective surrogate for Mg²⁺. Therefore, the inactivating effect of CK, induced by Al treatment in vivo and in vitro, must be linked to differences in the binding affinities of A1³⁺ and Mg²⁺ for the CK-B protein.

David et al.⁴⁷ reported that CK-BB in the AD brain becomes more hydrophobic and membrane binding results due to an unknown modification. Aksenov et al.⁴⁸ observed that CK-BB in the AD brain undergoes an oxidative posttranslational modification. In an effort to check the change in hydrophobic character of CK in Al-treated rat brains, we found that treatment of the Al group with hydrophobic agents released CK into the supernatant, whereas treatment with increased ionic reagents had no effect. Because of the binding affinity of A1³⁺ for the CK-B protein, the ability of Al to induce reactive oxygen species (ROS)^49,50 and the lack of any apparent change in the molecular weight of CK in the study of David et al.,⁴⁷ we conjecture that Al may have modified the enzyme, shifting it into a more hydrophobic conformation and causing the enzyme to partition into the membrane. This loss of CK activity and the change in hydrophobic character could explain why treatment of astrocytic cells with Al leads to a loss of energy synthesis.⁴⁷

The 14-3-3 proteins bind to a wide variety of intracellular proteins and regulate diverse cellular processes, such as intracellular signal transduction, the cell cycle, and cell survival.⁵¹ Tau protein belongs to the family of microtubule-associated proteins and is a major component of the neurofibrillary tangles (NFTs) in the brains of patients with AD. A 14-3-3 isoform, 14-3-3ζ, is abundant in the brain, binds to tau⁵² and is an effector of tau protein phosphorylation.⁵³ Several reports deal with probable involvement of 14-3-3 in tau functioning. Based on the binding affinity of A1³⁺ for the 14-3-3ζ protein observed in our study, we analyzed the effect of A1³⁺ on the affinity of 14-3-3ζ for tau in in vitro studies. Our findings show that A1³⁺ exerts a strong inhibitory effect on the interaction between tau and 14-3-3ζ, and this inhibitory effect is dependent on the concentration of Al (Fig. 5).

The binding of 14-3-3 proteins with most of their partners mainly depends on the phosphorylation of a Ser or Thr residue in the recognition domain.⁵⁴ More detailed investigations revealed that liberated tau undergoes phosphorylation by cAMP-dependent protein kinase or by protein kinase B, forming potential sites for 14-3-3 binding, and this phosphorylation of tau protein enhances the interaction of tau protein with 14-3-3.⁵⁵ Sluchanko reported that by binding to phosphorylated tau, 14-3-3 might inhibit its dephosphorylation due to protein phosphatases, and by this means, indirectly affect the interaction of tau with microtubules and tau aggregation. In addition, it is worthwhile to mention that 14-3-3 possesses chaperone-like activity and can prevent the aggregation of partially unfolded proteins.³⁰ Chaperone-like activity of 14-3-3 can also be important in the regulation of oligomerization and aggregation of tau proteins. In our study, the reduced affinity of 14-3-3ζ for tau induced by Al would reduce the preventive effect of 14-3-3 on tau aggregation. Although the biochemical pathways underlying this process disturbed by Al in vivo are unknown, it is certain that Al interferes with the function of 14-3-3ζ protein and changes the homeostasis of phosphorylated and unphosphorylated tau.

In summary, we attempted a novel protein staining procedure in native-PAGE and demonstrated that protein CK-B and albumin have a high affinity for Al³⁺. These high-affinity interactions perturb the normal function of CK-BB and albumin, and may explain the neurochemical and neuroanatomical alterations in the brain induced by Al. Further study should focus on molecular studies on the two specific Al binding proteins to elucidate the mechanisms underlying Al biotoxicity.

Conflicts of interest

No conflict of interest exists in the submission of this manuscript.

Ethical statement

The study complied with all institutional and national guidelines, as per Management rules of medical science and technology involving human body of the National Health and Family Planning Commission of PR China (2013, National Health and Family Planning Commission of PR China). The protocol was approved by the Ethics Committees for Human Research of China Agricultural University and Tianjin University of Science and Technology, and all participants provided written informed consent.

Acknowledgements

This work was partly supported by National Natural Science Foundation of China, National Basic Research Program of China (973 Program no. 2013CB127106) and China Postdoctoral Science Foundation (2017M621059).

References

1. S. Bhattacharjee, Y. Zhao, J. M. Hill, F. Culicchia, T. P. Kruck, M. E. Percy, A. I. Pogue, J. R. Walton and W. J. Lukiw, Selective accumulation of aluminum in cerebral arteries in Alzheimer's disease AD, J. Inorg. Biochem., 2013, 126, 35–37.
2. C. Exley, Aluminum and Alzheimer's disease, J. Alzheimers Dis., 2001, 3, 551–552.
3. J. R. Walton, Aluminum's Involvement in the Progression of Alzheimer's Disease, J. Alzheimers Dis., 2013, 35, 7–43.
4. A. Mirza, A. King, C. Troakes and C. Exley, Aluminium in brain tissue in familial Alzheimer's disease, J. Trace Elem. Med. Biol., 2017, 40, 30–36.
5. J. A. Varner, K. F. Jensen, W. Horvath and R. L. Isaacson, Chronic administration of aluminum-fluoride or sodium-fluoride to rats in drinking water: alterations in neuronal and cerebrovascular integrity, Brain Res., 1998, 784, 284–298.
6. J. D. Birchall and J. S. Chappell, The chemistry of aluminum and silicon in relation to Alzheimer's disease, Clin. Chem., 1988, 34, 265–267.
7. A. Khan, J. P. Dobson and C. Exley, Redox cycling of iron by Abeta42, Free Radicals Biol. Med., 2006, 40, 557–569.
8. A. Khan, A. E. Ashcroft, V. Higenell, O. V. Korchazhkina and C. Exley, Metals accelerate the formation and direct the structure of amyloid fibrils of NAC, J. Inorg. Biochem., 2005, 99, 1920–1927.
9. D. Ribes, M. Torrente, P. Vicens, M. T. Colomina, M. Gómez and J. L. Domingo, Recognition Memory and beta-amyloid Plaques in Adult Tg2576 Mice are not Modified After Oral Exposure to Aluminum, Alzheimer Dis. Assoc. Disord., 2012, 26, 179–185.
10. S. Sánchez-Iglesias, R. Soto-Otero, J. Iglesias-González, M. C. Barciela-Alonso, P. Bermejo-Barrera and E. Méndez-Alvarez, Analysis of brain regional distribution of aluminium in rats via oral and intraperitoneal administration, J. Trace Elem. Med. Biol., 2007, 21, 31–34.
11. T. Garcia, J. L. Esparza, M. R. Nogués, M. Romeu, J. L. Domingo and M. Gómez, Oxidative stress status and RNA expression in hippocampus of an animal model of Alzheimer's disease after chronic exposure to aluminum, Hippocampus, 2010, 20, 218–225.
12. J. Iglesias-González, S. Sánchez-Iglesias, A. Beiras-Iglesias, E. Méndez-Álvarez and R. Soto-Otero, Effects of Aluminium on Rat Brain Mitochondria Bioenergetics: an In vitro and In vivo Study, Mol. Neurobiol., 2017, 54, 563–570.
13. A. de Calignon, M. Polydoro, M. Suárez-Calvet, C. William, D. H. Adamowicz, K. J. Kopeikina, R. Pitstick, N. Sahara, K. H. Ashe, G. A. Carlson, T. L. Spires-Jones and B. T. Hyman, Prop-agation of tau pathology in a model of early Alzheimer's disease, Neuron, 2012, 73, 685–697.
14. R. Dixit, J. L. Ross, Y. E. Goldman and E. L. Holzbaur, Differential Regulation of Dynein and Kinesin Motor Proteins by Tau, Science, 2008, 319, 1086–1089.
15. J. M. Ugalde, E. Rezabal, J. M. Mercero and X. Lopez, A study of the coordination shell of aluminumIII and magnesiumII in model protein environments: Thermodynamics of the complex formation and metal exchange reactions, J. Inorg. Biochem., 2006, 100, 374–384.
16. E. Almaraz, Q. A. Paula, Q. Liu, J. H. Reibenspies, M. Y. Darensbourg and N. P. Farrell, Thiolate bridging and metal exchange in adducts of a zinc finger model and PtII complexes: biomimetic studies of protein/Pt/DNA interactions, J. Am. Chem. Soc., 2008, 130, 6272–6280.
17. D. J. Nelson, Aluminum complexation with nucleoside di- and triphosphates and implication in nucleoside binding proteins, Coord. Chem. Rev., 1996, 149, 95–111.
18. N. B. Gusev, A. S. Seit-Nebi and N. N. Sluchanko, Resolving mitochondrial protein complexes using nongradient blue native polyacrylamide gel electrophoresis, Biochem. Biophys. Res. Commun., 2009, 379, 990–994.
19. L. J. Yan and M. J. Forster, Resolving mitochondrial protein complexes using nongradient blue native polyacrylamide gel electrophoresis, Anal. Biochem., 2009, 389, 143–149.
20. S. J. Lin, J. Sun, D. D. Cao, J. K. Cao and W. B. Jiang, Distinction of different heat-treated bovine milks by native-PAGE fingerprinting of their whey proteins, Food Chem., 2010, 121, 803–808.
21. A. R. Bowie, P. R. Haddad, C. V. Butler and J. Tria, Determination of aluminium in natural water samples, Anal. Chim. Acta, 2007, 588, 153–165.
22. A. B. Tabrizi, Cloud point extraction and spectrofluorimetric determination of aluminum and zinc in foodstuffs and water samples, Food Chem., 2007, 100, 1698–1703.
23. K. Kashimura, Y. Mizushima, E. Hoshino and S. Matsubara, Kinetic differentiation mode chromatography using 8-quinolinol and fluorimetric detection for sensitive determination of aluminum adhering to the gastric mucosa, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2003, 791, 13–19.
24. S. S. A. El-Rahman, Neuropathology of aluminum toxicity in rats (glutamate and GABA impairment), Pharmacol. Res., 2003, 47, 189–194.
25. Q. X. A. Sang, A. G. Marshall, P. A. Stewart, X. Wang and S. M. Semaan, Differential phosphopeptide expression in a benign breast tissue, and triple-negative primary and metastatic breast cancer tissues from the same African-American woman by LC-LTQ/FT-ICR mass spectrometry, Biochem. Biophys. Res. Commun., 2011, 412, 127–131.
26. X. H. Qian, Y. Cai, J. J. Dong and S. H. Sui, Comparison of Phosphoproteomic Separation Strategies Based on Strong Cation Exchange Chromatography and Isoelectric Focusing Techniques, Chin. J. Anal. Chem., 2012, 40, 177–183.
27. D. Chang, C. F. Cheng, M. Wang, P. L. Chen and L. S. Chen, Global analysis of modifications of the human BK virus structural proteins by LC-MS/MS, Virology, 2010, 402, 164–176.
28. B. P. Hughes, A method for estimation of serum creatine kinase and its use in comparing creatine kinase and aldolase activity in normal and pathologic sera, Clin. Chim. Acta, 1962, 7, 597–604.
29. D. Berg, C. Holzmann and O. Riess, 14-3-3 proteins in the nervous system, Nat. Rev. Neurosci., 2003, 4, 752–762.
30. D. M. Williams, H. Ecroyd, K. L. Goodwin, H. Dai, H. Fu, J. M. Woodcock, L. Zhang and J. A. Carver, NMR spectroscopy of 14-3-3zeta reveals a flexible C-terminal extension: differentiation of the chaperone and phosphoserine binding activities of 14-3-3zeta, Biochem. J., 2011, 437, 493–503.
31. M. Carano, N. Lion and H. H. Girault, Detection of proteins on membranes and in microchannels using copper staining combined with scanning electrochemical microscopy, J. Electroanal. Chem., 2007, 599, 349–355.
32. K. Berggren, E. Chernokalskaya, T. H. Steinberg, C. Kemper, M. F. Lopez, Z. Diwu, R. P. Haugland and W. F. Patton, Background-freehigh sensitivity staining of proteins in one- and two-dimensional sodium dodecyl sulfate-polyacrylamide gels using a luminescent ruthenium complex, Electrophoresis, 2000, 21, 2509–2521.
33. M. L. Ramos, L. L. Justino, A. I. Salvador, A. R. de Sousa, P. E. Abreu, S. M. Fonseca and H. D. Burrows, NMR, DFT and luminescence studies of the complexation of Al(III) with 8-hydroxyquinoline-5-sulfonate, Dalton Trans., 2012, 41, 12478–12489.
34. R. T. Bronson, J. S. Bradshaw, P. B. Savage, S. Fuangswasdi, S. C. Lee, K. E. Krakowiak and R. M. Izatt, Bis-8-hydroxyquinoline-armed diazatrithia-15-crown-5 and diazatrithia-16-crown-5 ligands: Possible fluorophoric metal ion sensors, J. Organomet. Chem., 2001, 66, 4752–4758.
35. G. Farruggia, S. Iotti, L. Prodi, M. Montalti, N. Zaccheroni, P. B. Savage, V. Trapani, P. Sale and F. I. Wolf, 8-Hydroxyquinoline derivatives as fluorescent sensors for magnesium in living cells, J. Am. Chem. Soc., 2006, 128, 344–350.
36. E. J. Song, G. J. Park, J. J. Lee, S. Lee, I. Noh, Y. Kim, S.-J. Kim, C. Kim and R. G. Harrison, A fluorescence sensor for Zn²⁺ that also acts as a visible sensor for Co²⁺ and Cu²⁺, Sens. Actuators, B, 2015, 213, 268–275.
37. K. G. Vaswani and M. D. Keränen, Detection of aqueous mercuric ion with a structurally simple 8-hydroxyquinoline derived on-off fluorosensor, Inorg. Chem., 2009, 48, 5797–5800.
38. N. Lashgari, A. Badiei and G. Mohammadi Ziarani, A Fluorescent Sensor for Al(III) and Colorimetric Sensor for Fe(III) and Fe(II) Based on a Novel 8-Hydroxyquinoline Derivative, J. Fluoresc., 2016, 26, 1885–1894.
39. M. Buratti, C. Valla, O. Pellegrino, F. M. Rubino and A. Colombi, Aluminum determination in biological fluids and dialysis concentrates via chelation with 8-hydroxyquinoline and solvent extraction/fluorimetry, Anal. Biochem., 2006, 353, 63–68.
40. R. Beriault, D. Chenier, R. Singh, J. Middaugh, R. Mailloux and V. Appanna, Detection and purification of glucose 6-phosphate dehydrogenase Malic enzyme, and NADP-dependent isocitrate dehydrogenase by blue native polyacrylamide gel electrophoresis, Electrophoresis, 2005, 26, 2892–2897.
41. N. D. Priest, The biological behaviour and bioavailability of aluminium in man, with special reference to studies employing aluminium-26 as a tracer: review and study update, J. Environ. Monit., 2004, 6, 375–403.
42. S. J. Fatemi, F. H. Kadir and G. R. Moore, Aluminum transport in blood serum Binding of aluminium by human transferrin in the presence of human albumin and citrate, Biochem. J., 1991, 280, 527–532.
43. S. J. A. Fatemi, D. J. Williamson and G. R. Moore, A ²⁷Al NMR investigation of Al3+ binding to small carboxylic acids and the proteins albumin and transferrin, J. Inorg. Biochem., 1992, 46, 35–40.
44. U. Schlattner, R. Homayouni, K. Gough, M. Rak and A. Szeghalmi, The Creatine Kinase/Creatine Connection to Alzheimer's Disease: CK Inactivation, APP-CK Complexes, and Focal Creatine Deposits, J. Biomed. Biotechnol., 2006, 35936, 1–11.
45. J. Lemire, R. Mailloux and V. D. Appanna, The disruption of L-carnitine metabolism by aluminum toxicity and oxidative stress promotes dyslipidemia in human astrocytic and hepatic cells, Toxicol. Lett., 2011, 203, 219–226.
46. D. Dobrota, T. N. Pham, T. Lipta, J. Horecky, Z. Braunova, S. Kasparova and A. Gvozdjakova, A study of creatine kinase reaction in rat brain under chronic pathological conditions—Chronic ischemia and ethanol intoxication, Brain Res. Bull., 2000, 53, 431–435.
47. S. David, M. Shoemaker and B. E. Haley, Abnormal properties of creatine kinase in Alzheimer's disease brain: Correlation of reduced enzyme activity and active site photolabeling with aberrant cytosol-membrane partitioning, Mol. Brain Res., 1998, 54, 276–287.
48. M. Y. Aksenov, D. A. Butterfield and W. R. Markesbery, Oxidative modification of creatine kinase BB in Alzheimer's disease brain, J. Neurochem., 2000, 74, 2520–2527.
49. C. Exley, The pro-oxidant activity of aluminum, Free Radicals Biol. Med., 2004, 36, 380–387.
50. R. Mailloux, J. Lemire and C. Auger, How aluminum, an ROS generator promotes hepatic and neurological diseases: The metabolic tale, Toxicol. Lett., 2011, 105, 1513–1517.
51. H. Fu, R. R. Subramanian and S. C. Masters, 14-3-3 proteins: structure, function, and regulation, Annu. Rev. Pharmacol., 2000, 40, 617–647.
52. H. K. Paudel, A. Agarwal-Mawal, H. Y. Qureshi, P. W. Cafferty, Z. Yuan, D. Han and R. Lin, 14-3-3 connects glycogen synthase kinase-3 beta to tau within a brain microtubule-associated tau phosphorylation complex, J. Biol. Chem., 2003, 278, 12722–12728.
53. M. Hashiguchi, K. Sobue and H. K. Paudel, 14-3-3zeta is an effector of tau protein phosphorylation, J. Biol. Chem., 2000, 275, 25247–25254.
54. M. B. Yaffe, K. Rittinger, S. Volinia, P. R. Caron, A. Aitken, H. Leffers, S. J. Gamblin, S. J. Smerdon and L. C. Cantley, The structural basis for 14-3-3: Phosphopeptide binding specificity, Cell, 1997, 91, 961–971.
55. N. N. Sluchanko, A. S. Seit-Nebi and N. B. Gusev, Phosphory-lation of more than one site is required for tight interaction of human tau protein with 14-3-3zeta, FEBS Lett., 2009, 583, 2739–2742.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7tx00261k

‡ Present address: Qinghua Donglu no. 17 Beijing, 100083, PR China.

---