Pro-inﬂammatory effects of aluminum in human glioblastoma cells

Inﬂammatory events have been associated with senile plaques, one of the pathological hallmarks of Alzheimer’s disease (AD). It is believed that aggregated beta-amyloid ( b A) proteins, which form the core of these plaques, may be responsible for triggering the inﬂammatory reaction. In the present study, the ability of aluminum (Al) to initiate similar inﬂammatory events was investigated in a human glioblastoma cell line. A 6-day exposure to either lipopolysaccharide (LPS) or aluminum sulfate caused a signiﬁcant increase in the rate of proliferation of the glioblastoma cells. Both treatments also caused activation of the immune-responsive transcription factor NF- k B although there were time-related differences. The levels of secreted cytokines, interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF- a ) were both increased by the LPS treatment although exposure to Al decreased the secretion of the former while elevating the levels of the latter. These events may be due to the activation of glial cells and subsequent stress response to either Al complexes or LPS. Although exposure to either stress factor caused a stimulation of inﬂammatory markers, there were time-dependent differences in the response. This may reﬂect the ability of the cells to discern different stress factors and thus orchestrate an innate immune response proﬁle distinct to each immunogen. (cid:211)


Introduction
adults revaccinated for hepatitis B [9] and there is a correlation between focal lung inflammation and total A direct causal role for aluminum has been established deposited Al in occupational workers [30]. In rats, low in dialysis encephalopathy [1], osteomalacia [6,18] and doses of aluminum, present in parenteral nutrition formula, microcytic anemia [34,16]. It has been proposed that the produced marked portal inflammation that was correlated metal may play a role in the etiology of Alzheimer's with the duration of exposure and the amount of Al disease since an increase in Al has been found in the brain accumulated in the liver [12]. A premise of the current of patients with this disorder [10,38]. However, other study is that aluminum, once inside the brain, may elicit a studies have contradicted this finding [24,4] and the issue similar inflammatory response. remains unresolved. However, since the pathogenesis of In cells of glial origin, Al caused an increase in the rate this age-related disorder is multifactorial and the cerebral of reactive oxygen species production, glutathione deplecontent of the metal has been shown to increase with age tion and increased mitochondrial dehydrogenase activity [31,24], Al may promote events connected with the disease while cells of neuronal origin were not responsive to the process.
same treatment [7]. Al is prone to hydrolysis and will form Systemically, Al has been shown to cause inflammation.
insoluble precipitates within the physiological pH range The metal induced inflammatory nodules in Al-sensitized [32]. It is possible that these extracellular particles may induce an innate immune response in cells of glial origin. Human glioblastoma T98G cells were used to study the inflammation. Al sulfate was chosen because of its bio-logical relevance. The salt is used for water purification as were collected for the subsequent assays. At this time, the well as in clarifying fats and oils, antiperspirants, and same treated cells were re-seeded at a ratio of 1:5, exposed agricultural pesticides. This dose was selected based on to the salts prepared previously, and allowed to grow for previous results in our laboratory [7,8]. Lipopolysaccharide the other time-points. (LPS), in a concentration of 1 mg / ml, was used as a positive control. This component of the cell wall of gram-2.2.2. Cell proliferation assay negative bacteria has been used by several groups to Cells were frozen overnight at 270 8C and the number demonstrate increases in proinflammatory parameters was determined using the CyQuant cell proliferation assay [13,26,36]. To determine the toxic effect of exposure and kit (Molecular Probes, OR). Briefly, cells were thawed at cell number, cell viability and proliferation studies were room temperature and 200 ml of the dye, diluted in cell conducted. Since it was postulated that colloidal Al may be lysis buffer was added. The dye rapidly binds to nucleic causing an inflammatory response and aged aluminum salts acids and exhibits strong fluorescence when bound. The are known to aggregate [32], the effect of Al was studied cells were incubated at room temperature for 5 min and the up to 6 days of exposure. Aluminum and LPS treatment fluorescence was measured at excitation 480 nm and both caused an increase in several glial inflammatory emission 520 nm using a FL600 microplate fluorescence parameters although their responses were different. reader. A standard of cells counted with the hemacytometer was used to determine the actual cell number.

Cell viability assay
A viability / cytotoxicity kit (Molecular Probes) was used 2.1. Materials to measure cell survival after treatment. This assay is based on the simultaneous measurement of the fluorescence of Human glioblastoma (T98G) cells were purchased from two dyes. Live cells were determined by the retention of ATCC (Rockville, MD). All tissue culture supplies were the calcein AM dye, which is non-fluorescent and upon obtained from Gibco (Grand Island, NY). Cell viability enzymatic conversion by ubiquitous intracellular esterase and cell proliferation kits were purchased from Molecular activity, becomes intensely fluorescent. The EthD-1 dye Probes. The NF-kB gel shift assay compounds were can only enter cells with damaged membranes and there it purchased from Promega (Madison, WI). The enzyme binds to nucleic acids and produces a strong red fluoresimmunoassay used for detection of the cytokines was from cence. A final concentration of 1 mM calcein AM and 2 Neogen (Lexington, KY). All other chemicals used were mM EthD-1 was used in the assay. Upon addition of the from Sigma (St. Louis, MO).
dye, samples were incubated at room temperature for 45 min. The fluorescence due to calcein was measured at 2.2. Methods excitation 485 nm and emission 530 nm, while that of EthD-1 was measured at excitation 530 nm and emission 2.2.1. Cell preparation 645 nm using a FL600 microplate fluorescence reader. The Human cell lines were grown in minimum essential percentage of live and dead cells was calculated based on medium (MEM) with a modification consisting of nones-the intensity of fluorescence using a standard of live and sential amino acids, sodium pyruvate, lipoic acid, vitamin dead cells. The dead cells were prepared by treatment with B , biotin and ascorbic acid. The media also contained 0.25% digitonin for 10 min. 12 Earle's salts and L-glutamine. It was supplemented with 10% fetal bovine serum (heat-inactivated). Cells (0.5 2.2.4. Electrophoretic mobility shift assay 2 million) were seeded in 25-cm flasks and incubated at The gel shift assay was utilized to determine the extent 37 8C in a humidified atmosphere of 5% CO . Once the of NF-kB activation in the human glioblastoma (T98G) 2 cells were confluent, they were treated with the salts. The cells using a protocol developed by Promega (Madison, aluminum and LPS solutions were prepared in the cell WI). After treatment, the media was removed and the cells media and immediately sterile-filtered prior to the first were washed with Tris-HCl / 0.7% NaCl (pH 7.4) buffer. dosing. Although some aggregated aluminum complexes, Cells were lysed by several freeze-thaw cycles and the formed when Al sulfate was brought to neutral pH, was resulting cell extract was stored at 270 8C. The amount of removed, we have previously demonstrated that the filtered protein in 1 ml of the cell extract was determined by the solution still contains Al species capable of selectively BioRad protein assay [5] and 20 mg of each sample, 32 binding to the Al-specific Morin dye [8]. The solutions, incubated with P-labeled oligonucleotides containing the along with the media, were then incubated with the treated NF-kB consensus sequence, was loaded onto the gel. A cells to insure parallel conditions for the aging of the Al negative control containing no cell extract, as well as salt for up to 6 days. These were then utilized for further competitor reactions were run simultaneously with the exposure after the cells were harvested for each time point.
samples. The specific competitor contained unlabelled NF-After the initial 7-h dosing, supernatant and 4:5 of the cells kB consensus nucleotide while the nonspecific competitor contained unlabelled SP-1 consensus oligonucleotide. The competitor reactions also contained 20 mg of the 6-day LPS-treated sample. X-ray films were manually developed and the intensity of each band was measured and quantitated using the image analyzer, Eagle Eye, from Strategene (San Diego, CA).

Enzyme immunoassay
Levels of TNF-a and IL-6 were determined using sandwich enzyme immunoassay kits from Neogen (Lexington, KY) for the detection of free forms of the proteins and incubated for 45 min with goat anti-rabbit-conjugated alkaline phosphatase. Then, 200 ml of the color reagent solution was added to the plate and the color generated was determined with a spectrophotometric plate reader set at 490 nm. For the IL-6 assay, the samples were diluted effect either parameter at the 48-h time-point. After a 403 because the level of this cytokine in the cell culture 6-day exposure, both Al and LPS caused an increase in the supernatant was too concentrated for the detection range of rate of proliferation of the glioblastoma cells ( Fig. 1). At the assay.
this later time-point, there was no significant change in cell viability (Fig. 2).

Statistical analysis
3.2. NF-kB activation The difference among groups was assessed using oneway analysis of variance followed by the Student's t-test.
Exposure of the glioblastoma cells for 7 h to LPS caused significant activation of the transcription factor NF-kB. Al exposure also caused a slight increase in the intensity of 3. Results the shifted band at this time point. However, after a 6-day exposure, Al treatment resulted in a significant activation 3.1. Cell viability and proliferation of the transcription factor while the LPS effect was reduced compared to the control (Fig. 3). After a 48-h exposure to LPS (1 mg / ml), cell proliferation was increased although there was a significant 3.3. Cytokine secretion decline in the viability of the human glioblastoma T98G cells ( Figs. 1 and 2). In contrast, Al (500 mM) did not The level of secreted TNF-a increased after a 2-day treatment with LPS and a 6-day treatment further increased the amount of the protein present in the media. Al treatment also caused an increase in the amount of the cytokine but only after 6 days of treatment with the salt (Fig. 4). LPS also caused a substantial increase in the levels of IL-6 secretion at both time points assayed. In contrast, Al caused a significant reduction in the amount of secreted IL-6 after 6 days of treatment (Fig. 5).

Discussion
Inflammation appears to play an important role in the pathogenesis of Alzheimer's disease. Activated astrocytes cortex [28,33]. Treatment of rabbits chronically with    [39]. In the present study, the rate of proliferation greater than the value at 2 days of Al exposure. (P,0.05). Bars represent mean of five or six individual determinations6S.E. of human glioblastoma cells was increased by a 6-day exposure to either aluminum sulfate or LPS. This effect It was demonstrated here that Al and LPS cause a may have been due to increased activation of the cells.
time-dependent increase in NF-kB activation. While LPS Cytokines such as IL-1, IL-6 and TNF-a are generally caused an early acute response that declined after 6 days, synthesized by activated microglia and macrophages in Al caused a weak initial induction that increased after 6 response to pathogens and trauma [14]. Levels of cyto-days. There are three main categories of Al species in kines appear to be modulated in Alzheimer's disease.
solution. Monomeric complexes are formed immediately Detectable level of IL-6 was present in the temporal cortex upon addition of Al and hydroxylation of the metal in of AD patients but not in control brains [37]. Furthermore, solution. Polymeric species are formed through activation the serum levels of TNF-a was shown to be substantially of coordinated OH groups that become deprotonated and increased in AD patients compared to age-matched con-form bridges between metal centers. Further aging of the trols [15]. In an animal model of chronic inflammation, Al solution leads to the formation of metastable polynuinduced by intracranial infusion of lipopolysaccharide, clear Al complexes which grow in size and ultimately form there was astrogliosis as well as an increase in the levels of microcrystalline gibbsite [32]. The effect of Al on NF-kB IL-1, and TNF-a mRNA levels. This was followed by activation may be due to the precipitation of Al in solution hippocampal cell loss and impairment of spatial memory, and the formation of these colloidal species. We have all of which mirror changes seen in the AD brain [36]. In previously shown that the amount of positively charged the present study, a 6-day exposure of human glioblastoma Al-species capable of interacting with the Al-specific dye, cells to Al caused a significant increase in the levels of Morin, decreased as the solution aged [8]. This may be due TNF-a. This finding is in agreement with a study demon-to the formation of these aggregated Al species. In LPSstrating that levels of cerebral mRNA for this cytokine treated cells, although NF-kB activation was lower after substantially increased in mice chronically exposed to low the 6-day time point, it was still above control levels concentrations of Al in drinking water [35].
reflecting a sustained increase in pro-inflammatory cyto-Although Al caused a time-dependent increase in the kine secretion at the later time points. The significant level of TNF-a, the amount of IL-6 was unchanged after a LPS-induced induction of NF-kB after the 7-h treatment 2-day treatment and decreased after a 6-day treatment. It may underlie the decrease in viability after the 2-day has also been reported that Al does not effect the levels of treatment. IL-1b mRNA in mice cerebral tissue [35]. The two There are many similarities between the effect of cytokines appear to have different functions. While TNF-a aluminum salts and amyloid peptide fragments on cells of caused expression of the acute-phase protein, a -antich-glial origin. b-amyloid (Ab) can also activate microglial 1 ymotrypsin mRNA in human astrocytoma cells, IL-6 did cells [25,29]. The peptide increased NF-kB activation in not have the same effect [23]. This may underlie the primary rat astroglial cells [3,13], and Ab (25-35) fragdifference in the secretion profile of the two interleukins. ment dose-dependently caused release of TNF-a in mi-The selectivity and extent of the innate immune response, croglia [25]. In glial-derived cells, b-amyloid caused a corresponding to the type of pathogen encountered, is response that was similar to Al and in contrast to LPS, Ab controlled by the particular cytokine that is induced after (25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35) did not cause increased levels of IL-6 [26]. infection [27]. In the present study, the time-course While most studies showing parallels between Al and response of cells to Al or LPS was different and it would Ab have been largely performed in isolated systems, it be interesting to see if this may be due to activation of remains to be determined whether such similarities may alternate pathways. also apply to more complex in vivo systems. Furthermore, Cytokines such as IL-1b and TNF-a activate NF-kB it is not known whether co-exposure to Al and Ab may [21]. This transcription factor promotes the expression of modulate the inflammation induced by the peptide. These genes involved in inflammation, such as other cytokines, finding could then more solidly establish aluminum as a iNOS, and complement factors [2]. Thus, there is a cycle relevant contributory factor in the progression of AD. of cytokine secretion and NF-kB activation which if Furthermore, this study does not unequivocally prove prolonged, may cause neuronal cell death and further aluminum is an immunogen and further thorough invesproliferation of reactive glial cells. The activation of this tigation in both in vitro and in vivo settings are necessary transcription factor is involved in a variety of differing to reach such conclusion. However, this study does provide neuropathological conditions including experimental au-ample proof that aluminum does indeed activate cells of toimmune encephalomyelitis [19] and prion disease caused glial origin and by doing so, activates several parameters by the scrapie agent [20]. In temporal lobe neocortex involved in the inflammatory cascade. derived from aged or AD patients, NF-kB activation was significantly correlated to an elevation in a key inflammatory enzyme COX-2 [22]. The reason why so many diverse