EffEct of curcumin on accumulation in mononuclEar cElls and sEcrEtion in incubation medium of Аβ 40 and cytokinEs undEr local ExcEss of Аβ 42-homoaggrEgatEs

The aim of the work was to investigate accumulation of endogenous Aβ40 and cytokines (IL-1β, TNFα, IL-6, IL-10) in mononuclear cells and their secretion into incubation medium under Aβ42-aggregates’ toxicity and anti-inflammatory effects of curcumin. Mononuclear cells were isolated in Ficoll-Urografin density gradient from venous blood of healthy donors, resuspended and used for testing of homoaggregates of Aβ42 (15 nM), curcumin (54 pM) and their combinations on various timescales (0, 1, 2, 3, 6 and 24 hours). Endogenous Aβ40 and cytokines were detected in mononuclear cells and (separately) in incubation medium by ElISa. We demonstrated for the first time that homoaggregates of Aβ42 cause rapid accumulation of endogenous Aβ40 in mononuclear cells and accelerate its secretion into incubation medium. We found increased concentration of TNFα after 3 hours of incubation, and no changes in IL-1β concentration due to secretion of these pro-inflammatory factors into incubation medium. The concentrations of IL-6 in mononuclear cells were increased under effects of Aβ42 homoaggregates, and it was being secreted profoundly into incubation medium. Aβ42 did not affect Il-10 secretion, yet caused an increase in its intracellular concentration after 1 hour of incubation, which was subsequently suppressed. Curcumin prevented the increase in Aβ40 concentration in mononuclear cells and significantly decreased its secretion resulting from Aβ42 toxicity. Curcumin negated the activating effect of Aβ42 on pro-inflammatory cytokines, starting immediately for IL-1β and on 3-6 hours for TNFα, which resulted in decreased extracellular concentrations of these cytokines. The polyphenol also potentiated repleni shing of intracellular Il-6 and Il-10 concentrations and their secretion into incubation medium.

The aim of the work was to investigate accumulation of endogenous Aβ 40 and cytokines (IL-1β, TNFα, IL-6, IL -10) in mononuclear cells and their secretion into incubation medium under Aβ 42 -aggregates' toxicity and anti-inflammatory effects of curcumin.Mononuclear cells were isolated in Ficoll-Urografin density gradient from venous blood of healthy donors, resuspended and used for testing of homoaggregates of Aβ 42 (15 nM), curcumin (54 pM) and their combinations on various timescales (0, 1, 2, 3, 6 and 24 hours).Endogenous Aβ 40 and cytokines were detected in mononuclear cells and (separately) in incubation medium by ElISa.We demonstrated for the first time that homoaggregates of Aβ 42 cause rapid accumulation of endogenous Aβ 40 in mononuclear cells and accelerate its secretion into incubation medium.We found increased concentration of TNFα after 3 hours of incubation, and no changes in IL-1β concentration due to secretion of these pro-inflammatory factors into incubation medium.The concentrations of IL-6 in mononuclear cells were increased under effects of Aβ 42 homoaggregates, and it was being secreted profoundly into incubation medium.Aβ 42 did not affect Il-10 secretion, yet caused an increase in its intracellular concentration after 1 hour of incubation, which was subsequently suppressed.Curcumin prevented the increase in Aβ 40 concentration in mononuclear cells and significantly decreased its secretion resulting from Aβ 42 toxicity.Curcumin negated the activating effect of Aβ 42 on pro-inflammatory cytokines, starting immediately for IL-1β and on 3-6 hours for TNFα, which resulted in decreased extracellular concentrations of these cytokines.The polyphenol also potentiated repleni shing of intracellular Il-6 and Il-10 concentrations and their secretion into incubation medium.A lzheimer's disease is the primary cause of senile dementia associated with amyloid-β peptides (Aβ), oligomers and aggregates of which exert toxic and destructive effects upon neural tissue [1][2][3][4][5].Nevertheless, there is no clear evidence of the agent that provokes increased synthesis of amyloid-β precursor protein (AβPP) and the switch in its processing towards amyloid pathway followed by local accumulation of Aβ.Aβ is generally considered to be a toxic molecular waste product [6][7][8].Yet recent studies have demonstrated trophic [9] properties of AβPP (antimicrobial, in particular), that served as basis for its comparison with other antimicrobial peptides (defensins, histatins, and cathelicidins) and attribution to brain's innate immune system [10].We have proved [11] the role of chronic inflammation in initiation of amyloidosis, and the involvement of cytokines and Aβ in inflammatory response to toxic effects of Aβ-aggregates.
The variable dynamics of Aβ (Aβ 40 and Aβ 42 ) in blood serum and cerebrospinal fluid of patients with amyloidosis has been demonstrated [12,13].The Aβ-levels in peripheral blood flow were highly increased during initial asymptomatic stages of amy loid-associated pathology, on the other hand Aβ 40 and Aβ 42 levels in patients during neurodegenerative stages of the Alzheimer's disease were within or even below normal margins [14,15].The presen ce of these neuropeptides in biological fluids was attributed to Aβ accumulation in certain parts of brain (hippocampus, frontal cortex, and olfactory bulbs) and to increased permeability of blood-brain barrier due to inflammation [16,17].In contrast, we have established [18] that AβPP expression and its doi: http://dx.doi.org/10.15407/ubj88.03.083 amyloidogenic processing in human peripheral blood mono nuclear cells is activated in response to Aβ 42 effects and leads to Aβ 40 accumulation regardless of neural tissue.One of the aims of the present work was to establish the possibility of secretion of Aβ 40 produced by peripheral mononuclear cells into surrounding medium, which could be used as its indicator in biological fluids of patients with Alzheimer disease and of its function other than being an amyloidohenesis.
Another standing problem is finding approaches to eliminate causes for AβPP overexpression and amyloidogenic processing, and inhibiting the inflammation resulting from Aβ-aggregate toxicity.We have demonstrated curcumin efficiency as a regulator of cytokine-dependent inflammation in vivo and in vitro [18][19][20].Therefore it was sensible to investigate the effect of curcumin on accumulation and secretion of endogenous Aβ 40 and cytokines by mononuclear cells of human peripheral blood in vitro under Аβ 42 aggregates' toxicity.

materials and methods
The experiments were conducted in accordance with provisions of the Universal Declaration on Bioethics and Human Rights (UNESCO, 2005).
Peripheral blood mononuclear cells were isolated ex tempore in Ficoll-Urografin density gradient from venous blood samples of three healthy donors (separately).The cells were washed thrice with sterile normal saline at room temperature and resuspended in RPMI medium in aliquots of 2×10 6 cells/ ml.The resulting samples (n = 3) were subjected to Aβ 42 (15 nM), curcumin (54 pM) and their combination (with the same concentrations) at various timeframes.The ratio of volumes of effectors to cell suspension was 1:100.
Aβ 42 _Human (Human Amyloid β Protein Fragment 1-42, Sigma-Aldrich, USA) was dissolved in double-distilled water and aggregated for 24 h at 37 °C.Large crude Aβ 42 agglomerates were disintegrated by ultrasound and sterilized prior to application.
As curcumin is water-insoluble, the primary solution was first dissolved in 96% ethanol and then diluted to 0.7 g/l immediately prior to addition to cell suspension.
The effect of Aβ 42 and 0.9% NaCl on mononuclear cells was investigated in 0-, 1-, 2-, 3-, 6-, and 24-hour incubation experiments at 37 °C and 600 rpm mix.The effects of curcumin alone and curcumin combined with Aβ 42 were investigated in 2-, 3-, 6-, and 24-hour experiments under the same conditions; curcumin was added after 1 hour incubation with Aβ 42 or normal saline (0.9% NaCl).Cells were sampled at the mentioned time points (2×10 6 cells/ml), sedimented by centrifugation and disintegrated by ultrasound (MUSSON-1 ultrasound inhalator, 3 min treatment at 2.64 MHz wavelength and 0.25 W/cm 3 intensity) the samples were then centrifuged at 6000 rpm for 20 min.Cell supernatant and incubation medium aliquots were used for the enzyme-linked immunosorbent assay (ELISA).
Concentrations of endogenous Aβ 40 and cytokines were estimated by ELISA in accordance with kit manuals (Vektor-Best, Russia) for IL-1β, IL-6, IL-10, and TNFα, and ELISA Kit Human Аβ 40 (Invitrogen, USA).Absorptions were measured with GBG STAT-FAX 2100 (USA) at 450 nm and correction at 630 nm.The results were calculated against total protein concentration (ng/g of protein) measured by Lowry method [21].
Mean values and standard deviations were determined for the indicators of mononuclear cells suspension.The statistical analysis was performed with Student's t-test, the differences were considered significant at P < 0.05.

results and discussion
Study of Aβ 42 and curcumin's effects on inflammation dynamics in suspension of mononuclear cells include in vitro determination of intracellular accumulation of Aβ 40 and cytokines (IL-1β, TNFα, IL-6, and IL-10) and their secretion into incubation medium.The results (Fig. 1) demonstrate a 2-fold increase in endogenous Aβ 40 concentration under incubation of mononuclear cells with normal saline or curcumin for 6 h, which is non-specific and may be explained by a spontaneous flux in expression or processing of AβPP [18].Aβ 42 homoaggregates caused 7.7-fold increase in Aβ 40 concentration in mononuclear cells and activation of its secretion on the 1 st h of incubation.The described [18] early and rapid activation of AβPP processing in mononuclear cells under Aβ 42 toxicity led to notab le increase of intracellular Aβ 40 levels as well as to forced secretion of this mediator of inflammation into incubation medium.
The intracellular Aβ 40 concentration then gradual ly decreased for 6-24 hours, but did not recede to the starting levels (0 hour) or the levels in mononuclear cells incubated with normal saline for 24 h.Beginning at 6 th h of incubation with Aβ 42 , Aβ 40 secretion was increased 2-fold and remained ele vated afterwards (Fig. 1).
Curcumin had no effect upon Aβ 40 concentration in suspension of mononuclear cells in comparison with the dynamics displayed by cells incubated with normal saline.Nevertheless, it noticeably inhibi ted the increase in Aβ 40 intracellular concentration and secretion caused by Aβ 42 (Fig. 1).The inhibiting effect of curcumin upon Aβ 40 production is due to its ability to downregulate GSK-3β mediated activation of presenilin-1 (PS-1) [22].PS-1 participates in γ-secretase enzyme complex, which takes part in AβPP processing, and is a GSK-3β substrate.The latter modulates γ-secretase activity through phosphorylation of serine in PS-1 loop domain [23].Curcumin has been demonstrated to increase the proportion of inactivated (Ser9-phosphorylated) GSK-3β form depending on concentration and duration of exposure, and also to inhibit expression of PS1 and GSK-3β genes [22].These factors lead to diminished Aβ production.Our data are in accordance with the results by others who investigate curcumin effects on the model of Aβ-induced inflammation of primary astrocytes [24] and mouse cortical neurons culture [25].
Thus, Aβ 42 homoaggregates induced accumulation of endogenous Aβ 40 in mononuclear cells and stimulated secretion of this pro-inflammatory factor into incubation medium.Curcumin served to substantially prevent the Aβ 40 cellular concentration increase and significantly decreased its secretion associated with toxic effects of exogenous Aβ 42 .
We observed a noticeable increase in TNFα concentration in mononuclear cells and its secretion under effect of mere normal saline, with the maximum effect on 6 th h of exposure (Fig. 2).Unlike the TNFα dynamics, the levels of IL-1β under effects of normal saline fluctuated somewhat close to the baseline.The significant decrease was measured on the 1 st and 24 th h in cells, and the secretion was registered beginning with the 3 rd h of exposure (Fig. 3).We attribute these fluctuations in levels of cytokines to spontaneous activation of mononuclear cells due to isolation-associated stress.
We noted an increase of TNFα concentration in mononuclear cells (beginning with the 3 rd h of exposure) and incubation medium (beginning with the 6 th h of exposure) as a specific response to the effects of Aβ 42 homoaggregates (Fig. 2).Curcumin alleviated this effect of Aβ 42 within 3 to 6 hours.TNFα levels in mononuclear cells exposed to curcumin alone resembled those of normal saline-exposed cells.These results confirm that curcumin inhibits only the Aβ 42 -induced accumulation of TNFα, and does not affect its spontaneous production.
Curcumin effect on IL-1β concentration after 1-h incubation with Aβ 42 was apparent on the 1 st h of the polyphenol's effect (2 h, Fig. 3).Its total concentration was 33% lower.During subsequent incubation of mononuclear cells with Aβ 42 and curcumin the IL-1β levels did not differ from the basic ones (Fig. 3), which is probably due to the fact that curcumin has no effect on caspase 1 [31].
We observed increased secretion of the pro-inflammatory cytokines in incubation medium for the duration of incubation (up to 24 h) beginning from the 1 st h of exposure for IL-1β and from the 3 rd h for TNFα (Fig. 2 and 3) in response to all the effectors and their combination.
Thus, in vitro incubation of mononuclear cells with exogenous Aβ 42 leads to elevated intracellular concentration of TNFα (on the 3 rd h of incubation), but not to accumulation of IL-1β in cells, which served to potentiate the release of these cytokines into incubation medium.Curcumin addition alleviated this effect of Aβ 42 upon cellular concentrations of the pro-inflammatory cytokines, beginning immedia tely for Il-1β and from 3 to 6 hours for TNFα, which led to their diminished extracellular concentrations.
The dynamics of IL-6 content in mononuclear cells incubated with normal saline (Fig. 4) genera lly follows that of TNFα, with the exception that intracellular IL-6 level dropped 3.6-fold immediately and then gradually increased towards starting values for 6-24 h, while TNFα content in mononuclear cells increased twofold from the 1 st h of incubation and was 4.8 times higher than the starting value after 6 hours.IL-6 secretion into incubation medium was detected at 6-24 h (Fig 4), which differs substantially from the rapid excretion of the pro-inflammatory cytokines of the initial wave of cytokine system: TNFα (beginning with the 1 st h) and IL-1β (beginning with the 3 rd h).
Exposure to Aβ 42 homoaggregates, curcumin, or both caused immediate decrease in intracellular concentrations of IL-6 followed by restoration on the 3 rd h of incubation.Curcumin subsequently caused gradual increase in accumulation and secretion of the cytokine, and Aβ 42 homoaggregates caused its decreased accumulation and increased secretion (Fig. 4).Therefore, we established increase in intracellular levels of IL-6 and activation of its secretion into incubation medium under influence of all the tested effectors.The effect of curcumin was substantial on the 24 th h of exposure.These data are not in discrepancy with the evidence of the inhibitive effects of curcumin on the activation of pro-inflammatory cytokines [32,33].Aβ 42 did not affect IL-10 secretion, yet caused increase in its intracellular concentration after 1 h exposure, with subsequent inhibition of accumulation in cells (Fig. 5).Curcumin addition after 1-h exposure to Aβ 42 homoaggregates restored intracellular levels of the anti-inflammatory interleukin on 6-24 h of incubation in vitro.Curcumin alone caused gradual elevation in intracellular IL-10 content (1-3 h) with increased excretion in incubation medium on 6-24 h (Fig. 5).
These data corroborate our previous results indicating that IL-10 expression is not induced in mononuclear cells under effect of exogenous Aβ 42 , and that concentration of iRNA of IL-10 is increased under effect of curcumin [18].Others have proven the positive effects of curcumin upon IL-10 expres-sion [34,35], and attributed these in vivo effects to inhibition of p38 activity through suppression of its phosphorylation.
Therefore, we hereby establish the capability of mononuclear cells to produce and secrete endogenous Aβ 40 , which indicates its peripheral origin if detected in blood flow.We also demonstrate the non-amyloidogenic function of Aβ 40 as a pro-inflammatory messenger responsive to effects of Aβ 42 K e y w o r d s: curcumin, β-amyloid peptides 40 and 42 (Aβ 40 , Aβ 42 ), cytokines, secretion, human peripheral blood mononuclear cells.

Fig. 1 .
Fig. 1.Aβ 40 content in mononuclear cells and incubation medium under effect of normal saline, Aβ 42 , curcumin, and their combination.* denotes changes with P < 0.05 in comparison to normal saline effect; # denotes changes with P < 0.05 in comparison to Aβ 42 effect; & denotes changes with P < 0.05 in comparison to the preceding time point

Fig. 2 .Fig. 3 .
Fig. 2. TNFα content in mononuclear cells and incubation medium under effect of normal saline, Aβ 42 , curcumin, and their combination.*denotes changes with P < 0.05 in comparison to normal saline effect; # denotes changes with P < 0.05 in comparison to Aβ 42 effect; & denotes changes with P < 0.05 in comparison to the preceding time point

EffEct of curcumin on accumulation in mononuclEar cElls and sEcrEtion in incubation medium of Аβ 40 and cytokinEs undEr local ExcEss of Аβ 42 -homoaggrEgatEs
UDC 616.894-092:615.3