Suppression of Nuclear Factor-κB Activation and Inflammation in Microglia by Physically Modified Saline*

Background: Microglial activation plays an important role in the pathogenesis of neurodegenerative disorders. Results: Taylor-Couette-Poiseuille flow-modified saline (RNS60) inhibits microglial inflammation via type 1A phosphatidylinositol 3-kinase-Akt-CREB-mediated up-regulation of IκBα and inhibition of NF-κB activation. Conclusion: These results delineate a novel biological function of a physically modified saline. Significance: RNS60 may be of therapeutic benefit in neurodegenerative disorders. Chronic inflammation involving activated microglia and astroglia is becoming a hallmark of many human diseases, including neurodegenerative disorders. Although NF-κB is a multifunctional transcription factor, it is an important target for controlling inflammation as the transcription of many proinflammatory molecules depends on the activation of NF-κB. Here, we have undertaken a novel approach to attenuate NF-κB activation and associated inflammation in activated glial cells. RNS60 is a 0.9% saline solution containing charge-stabilized nanostructures that are generated by subjecting normal saline to Taylor-Couette-Poiseuille (TCP) flow under elevated oxygen pressure. RNS60, but not normal saline, RNS10.3 (TCP-modified saline without excess oxygen), and PNS60 (saline containing excess oxygen without TCP modification) were found to inhibit the production of nitric oxide (NO) and the expression of inducible NO synthase in activated microglia. Similarly, RNS60 also inhibited the expression of inducible NO synthase in activated astroglia. Inhibition of NF-κB activation by RNS60 suggests that RNS60 exerts its anti-inflammatory effect through the inhibition of NF-κB. Interestingly, RNS60 induced the activation of type IA phosphatidylinositol (PI) 3-kinase and Akt and rapidly up-regulated IκBα, a specific endogenous inhibitor of NF-κB. Inhibition of PI 3-kinase and Akt by either chemical inhibitors or dominant-negative mutants abrogated the RNS60-mediated up-regulation of IκBα. Furthermore, we demonstrate that RNS60 induced the activation of cAMP-response element-binding protein (CREB) via the PI 3-kinase-Akt pathway and that RNS60 up-regulated IκBα via CREB. These results describe a novel anti-inflammatory property of RNS60 via type IA PI 3-kinase-Akt-CREB-mediated up-regulation of IκBα, which may be of therapeutic benefit in neurodegenerative disorders.


Chronic inflammation involving activated microglia and astroglia is becoming a hallmark of many human diseases, including neurodegenerative disorders. Although NF-B is a multifunctional transcription factor, it is an important target for controlling inflammation as the transcription of many proinflammatory molecules depends on the activation of NF-B.
Here, we have undertaken a novel approach to attenuate NF-B activation and associated inflammation in activated glial cells. RNS60 is a 0.9% saline solution containing charge-stabilized nanostructures that are generated by subjecting normal saline to Taylor-Couette-Poiseuille (TCP) flow under elevated oxygen pressure. RNS60, but not normal saline, RNS10.3 (TCP-modified saline without excess oxygen), and PNS60 (saline containing excess oxygen without TCP modification) were found to inhibit the production of nitric oxide (NO) and the expression of inducible NO synthase in activated microglia. Similarly, RNS60 also inhibited the expression of inducible NO synthase in activated astroglia. Inhibition of NF-B activation by RNS60 suggests that RNS60 exerts its anti-inflammatory effect through the inhibition of NF-B. Interestingly, RNS60 induced the activation of type IA phosphatidylinositol (PI) 3-kinase and Akt and rapidly up-regulated IB␣, a specific endogenous inhibitor of NF-B. Inhibition of PI 3-kinase and Akt by either chemical inhibitors or dominant-negative mutants abrogated the RNS60mediated up-regulation of IB␣. Furthermore, we demonstrate that RNS60 induced the activation of cAMP-response elementbinding protein (CREB) via the PI 3-kinase-Akt pathway and that RNS60 up-regulated IB␣ via CREB. These results describe a novel anti-inflammatory property of RNS60 via type IA PI 3-kinase-Akt-CREB-mediated up-regulation of IB␣, which may be of therapeutic benefit in neurodegenerative disorders.
Although inflammation is a protective cellular response aimed at removing injurious stimuli and initiating the healing process, prolonged inflammation, known as chronic inflammation, goes beyond physiological control, and eventually destructive effects override the beneficial effects. Nowadays, persistent inflammation is considered as an underlying contributor to virtually every chronic disease. Recent studies demonstrate that chronic inflammation in the central nervous system (CNS) is also a hallmark of various neurodegenerative disorders in which progressive loss of structure and function of neurons and neuronal death are observed (1)(2)(3)(4)(5). For example, the concentration of NO 2 Ϫ (nitrite), a metabolite of nitric oxide (NO), increases in the cerebrospinal fluid of patients with Parkinson disease and Alzheimer disease in comparison with agematched controls (6). Consistently, the ablation of inducible nitric-oxide synthase (iNOS) 2 in mutant mice significantly protects dopaminergic neurons from MPTP neurotoxicity, indicating that iNOS is essential in MPTP-induced substantia nigra pars compacta dopaminergic neurodegeneration (7). A variety of pro-inflammatory cytokines, including tumor necrosis factor ␣ (TNF-␣), interleukin-1␤ (IL-1␤), IL-6, eicosanoids, and other immune neurotoxins, are found in either cerebrospinal fluid or affected brain regions of patients with neurodegenerative disorders (8). Finally, NF-B, a transcription factor required for the transcription of most proinflammatory molecules, is activated in the substantia nigra pars compacta of Parkinson disease patients and MPTP-intoxicated mice and monkeys, and selective inhibition of NF-B in mice and monkeys by NF-B essential modifier-binding domain peptides protects dopaminergic neurons from MPTP toxicity (9,10). Therefore, inflammation is an important target for neuronal protection in neurodegenerative disorders.
RNS60 is a physically modified saline that is generated by subjecting normal saline to Taylor-Couette-Poiseuille (TCP) flow under elevated oxygen pressure. TCP flow combines intense local mixing with limited axial dispersion (11), providing large turbulent energy dissipation rates and surface-to-volume mixing ratios (12). Gas/liquid interface and energy transfer were maximized by using a rotor with numerous cavities and by employing rotational speeds above 3,000 rpm. These conditions generate a strong shear layer at the interface between the vapor and liquid phases near the rotor cavities, which correlates with the generation of small bubbles from cavitation (13).
Although nanobubbles are used as contrast agents in biomedical imaging and have been explored for drug delivery, they have not been shown to have direct biological effects. Here, we delineate new beneficial properties of RNS60. RNS60 suppressed the expression of proinflammatory molecules and the activation of NF-B in activated glial cells via type 1A phosphatidylinositol-3 (PI-3) kinase-Akt-CREB-mediated up-regulation of IB␣, a specific endogenous inhibitor of the classical NF-B heterodimer p65:p50 (14,15). These results establish a novel mode of action of RNS60 and show promise for developing a new class of anti-inflammatory agents as primary or adjunct therapy for neuroinflammatory diseases.

EXPERIMENTAL PROCEDURES
Isolation of Primary Mouse Microglia and Astroglia-Microglia were isolated from mixed glial cultures according to the procedure of Giulian and Baker (16) with modifications as described previously by us (17). Briefly, on day 9, the mixed glial cultures were washed three times with Dulbecco's modified Eagle's medium/F-12 and subjected to shaking at 240 rpm for 2 h at 37°C on a rotary shaker. The floating cells were washed, seeded onto plastic tissue culture flasks, and incubated at 37°C for 1 h. The attached cells were seeded onto new plates for further studies. Ninety to 95% of cells were found to be positive for Mac-1. On day 11, mixed glial cultures were washed and shaken again on a rotary shaker at 180 rpm for 18 h. Floating cells were removed, and the remaining cells were trypsinized and seeded onto new plates for experiments. More than 96% of cells were positive for GFAP. Mouse BV-2 microglial cells (kind gift from Virginia Bocchini, University of Perugia) were also maintained and induced as indicated above.
Preparation of RNS60-RNS60 was generated at Revalesio (Tacoma, WA) using a rotor/stator device (18), which incorporates controlled turbulence and TCP flow. Briefly, sodium chloride (0.9%) for irrigation, USP (pH 5.6) (4.5-7.0, Hospira), was processed at 4°C, and a flow rate of 32 ml/s under 1 atm of oxygen back-pressure (7.8 ml/s gas flow rate) while maintaining a rotor speed of 3,450 rpm. The resulting fluid was immediately placed into glass bottles (KG-33 borosilicate glass, Kimble-Chase) and sealed using gray chlorobutyl rubber stoppers (USP class 6, West Pharmaceuticals) to maintain pressure and minimize leachables. When tested after 24 h, the oxygen content was 55 Ϯ 5 ppm. Chemically, RNS60 contains water, sodium chloride, 50 -60 ppm oxygen but no active pharmaceutical ingredients.
The following controls for RNS60 were also used in this study: (a) normal saline (NS) from the same manufacturing batch. This saline contacted the same device surfaces as RNS60 and was bottled in the same way; (b) RNS10.3, saline that was processed through the same device without adding excess oxygen, and (c) PNS60, saline with the same oxygen content (55 Ϯ 5 ppm) that was prepared inside the same device but was not processed with TCP flow. Careful analysis demonstrated that all three fluids were chemically identical.
Analysis of RNS60, NS, RNS10.3, and PNS60 by Liquid Chromatography Quadrupole Time-of-Flight Mass Spectrometry (LC-Q-TOF)-To test for compositional differences in RNS60, NS, and PNS60, the LC-Q-TOF system was configured with an electrospray ionization interface, and the analysis was performed in both positive and negative modes. The samples were introduced in triplicate via flow injection analysis into the system, and data acquisition was performed in TOF mode (MS only, scan) in the 100 -1,000 m/z mass range.
Atomic Force Microscopy (AFM)-Tapping mode AFM was performed at ambient temperature (20 Ϯ 1°C) and at pH 5.6 with a D3000 microscope, a DTFML-DD liquid cell, and DNPS silicon nitride cantilevers (all Bruker Instruments) with a quoted stiffness of 0.35 newtons/m, and height images were obtained. A scan rate of 1 Hz and amplitude set point of 90% were used for all images. The substrate is a silicon wafer spincoated with polystyrene to give a root mean square roughness of 0.44 nm and a static contact angle of 93°.
Assay for NO Synthesis-Synthesis of NO was determined by assay of culture supernatant for nitrite, a stable reaction product of NO with molecular oxygen, using "Griess" reagent as described earlier (17,19,20).
Assay of Type 1A and Type 1B PI 3-Kinases-After stimulation, cells were lysed with ice-cold lysis buffer containing 1% v/v Nonidet P-40, 100 mM NaCl, 20 mM Tris (pH 7.4), 10 mM iodoacetamide, 10 mM NaF, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl chloride, 1 g/ml leupeptin, 1 g/ml antipain, 1 g/ml aprotinin, and 1 g/ml pepstatin A. Lysates were incubated at 4°C for 15 min followed by centrifugation at 13,000 ϫ g for 15 min. The supernatant was precleared with protein G-Sepharose beads (Bio-Rad) for 1 h at 4°C followed by the addition of 1 g/ml p85␣, p101, or p84 monoclonal antibodies. After a 2-h incubation at 4°C, protein G-Sepharose beads were added, and the resulting mixture was further incubated for 1 h at 4°C. The immunoprecipitates were washed twice with lysis buffer, once with phosphate-buffered saline, once with 0.5 M LiCl and 100 mM Tris (pH 7.6), once in water, and once in kinase buffer (5 mM MgCl 2 , 0.25 mM EDTA, 20 mM HEPES (pH 7.4)). PI 3-kinase activity was determined as described earlier (21) using a lipid mixture of 100 l of 0.1 mg/ml phosphatidylinositol and 0.1 mg/ml phosphatidylserine dispersed by sonication in 20 mM HEPES (pH 7.0) and 1 mM EDTA. The reaction was initiated by the addition of 20 Ci of [␥-32 P]ATP (3,000 Ci/mmol; PerkinElmer Life Sciences) and 100 M ATP and terminated after 15 min by the addition of 80 l of 1 N HCl and 200 l of chloroform/methanol (1:1). Phospholipids were separated by TLC and visualized by exposure to iodine vapor and autoradiography. Similarly, to monitor p110␣-, p110␤-, and p110␦-associated PI 3-kinase activity, supernatants were immunoprecipitated with antibodies against p110␣, p110␤, and p110␦ followed by immunocomplex lipid kinase assay as described above.
Expression of Different Mutant Constructs of PI 3-Kinase-Class IA PI 3-kinase consists of a catalytic subunit (p110) of 110 kDa and a regulatory subunit (p85) of 85 kDa. In the dominantnegative form of p85␣, 35 amino acids in the inter-Src homology 2 region from residues 479 to 513 of wild type p85␣, important for binding the p110␣/␤ subunit of PI 3-kinase, were deleted, and two other amino acids (Ser-Arg) were inserted in this deleted position (22). In the constitutively active mutant of p110␣/␤ (p110*), the inter-Src homology 2 domain of p85 is ligated to the NH 2 terminus of p110, whereas in the kinasedeficient mutant of p110␣/␤ (p110-kd), the ATP-binding site was mutated (23). Microglial cells plated in 12-well plates were transfected with 0.2-0.25 g of different plasmids using Lipofectamine-Plus (Invitrogen) following the manufacturer's protocol as described previously (21).
Electrophoretic Mobility Shift Assay (EMSA) of NF-B and CREB-Nuclear extracts were prepared, and EMSA was carried out as described previously (20,24,25) using IRDye TM infrared dye end-labeled oligonucleotides containing the consensus binding sequence for NF-B and CREB (LI-COR Biosciences, Lincoln, NE). Six micrograms of nuclear extract were incubated with binding buffer and with IR-labeled probe for 20 min. Subsequently, samples were separated on a 6% polyacrylamide gel in 0.25ϫ TBE buffer (Tris borate/EDTA) and analyzed by Odyssey Infrared Imaging System (LI-COR Biosciences).
Assay of Transcriptional Activities of NF-B and CREB-Transcriptional activity was assayed as described previously (20,21,24). Briefly, cells plated at 50 -60% confluence in 12-well plates were transfected with 0.25 g of either pBIIX-Luc (a NF-B-dependent reporter construct) or pCRE-Luc (a CREBdependent reporter construct) and 12.5 ng of pRL-TK (a plasmid encoding Renilla luciferase, used as transfection efficiency control; Promega, Madison, WI) using Lipofectamine Plus (Invitrogen). After 24 h of transfection, cells were stimulated with MPP ϩ for an additional 6 h, and firefly and Renilla luciferase activities were recorded in a TD-20/20 Luminometer (Turner Designs, Sunnyvale, CA) by analyzing total cell extract according to standard instructions provided in the Dual-Luciferase kit (Promega). Relative luciferase activity of cell extracts was typically represented as (firefly luciferase value/ Renilla luciferase value) ϫ 10 Ϫ3 .
Immunoblot Analysis-Immunoblot analysis for IB␣, p65, p50, and iNOS was carried out as described earlier (24,26). Briefly, cell homogenates were electrophoresed, proteins were transferred onto a nitrocellulose membrane, and bands were visualized with an Odyssey infrared scanner after immunolabeling with respective primary antibodies followed by infrared fluorophore-tagged secondary antibody (Invitrogen).
Densitometric Analysis-Protein blots were analyzed using ImageJ (National Institutes of Health, Bethesda, MD), and bands were normalized to their respective ␤-actin loading controls. Data are representative of the average fold change with respect to control for three independent experiments.
Real Time PCR Analysis-DNase-digested RNA was analyzed by real time PCR in the ABI-Prism7700 sequence detection system (Applied Biosystems, Foster City, CA) as described earlier (20,21,27) using TaqMan universal master mix and optimized concentrations of FAM-labeled probes and primers. Data were processed using the ABI Sequence Detection System 1.6 software.
Statistics-All values are expressed as means Ϯ S.E. One-way analysis of variance was performed while analyzing dose-dependent effect of RNS60 on mRNA expression of IB␣ in activated microglial cells.
Immunofluorescence Analysis-It was performed as described earlier (9,17,24). Briefly, coverslips containing 100 -200 cells/ mm 2 were fixed with 4% paraformaldehyde followed by treatment with cold ethanol and two rinses in PBS. Samples were blocked with 3% BSA in PBS/Tween 20 (PBST) for 30 min and incubated in PBST containing 1% BSA and goat anti-CD11b (1:50), rabbit anti-iNOS (1:200), or goat anti-GFAP (1:50). After three washes in PBST (15 min each), slides were further incubated with Cy2 (Jackson ImmunoResearch). For negative controls, a set of culture slides was incubated under similar conditions without the primary antibodies. The samples were mounted and observed under a Bio-Rad MRC1024ES confocal laser-scanning microscope.

RNS60 Is Chemically Identical to Control Saline Solutions-
To detect any chemical contaminants, RNS60 was carefully compared with two control solutions as follows: (a) NS, unprocessed normal saline from the same manufacturing batch that contacted the same device surfaces and was bottled in the same way as RNS60; (b) PNS60, normal saline from the same manufacturing batch that was prepared inside the same device and bottled in the same way as RNS60, but not processed with the TCP flow. Oxygen pressure was applied above the solution until PNS60 contained a level of oxygen comparable with RNS60 (50 -60 ppm).
Careful ICP-MS testing for 26 metals and total organic carbon analysis revealed no differences between RNS60 and the control solutions within detection limits (supplemental Table  1). Similarly, triplicate MS-TOF analysis with and without chromatographic pretreatment, using either negative or positive ion modes, did not reveal chemical composition differences in the 100 -1000 m/e range (only a part of it has been presented in supplemental Fig. 1). In all cases, samples for RNS60 and saline source solution were found to be identical within ϳ1 ppb.
RNS60 Has a Nanobubble Composition Different from That of Control Saline-Small bubble formation occurs through hydrodynamic cavitation (28), and nanobubbles are produced within various types of mixing devices when sharp pressure variations are generated (29). Measurements of nanobubblebased structures in bulk solution are controversial (30), although recent molecular dynamics studies suggest that bulk nanobubbles exist and are stable for much longer periods of time than previously expected (31). These relationships are tied to size and charge and the distribution of the bubble populations within the solution. Observations on surfaces using AFM, however, are increasingly accepted (32) and may represent characteristics of the bubbles within the solution (33). Therefore, we studied the effects of RNS60 on nanobubble nucleation using AFM. We acquired AFM images of polystyrene-coated silicon wafers immersed in RNS60, PNS60, NS, and distilled water. As evident from AFM images, deionized water (Fig. 1A), NS (Fig. 1B), PNS60 (Fig. 1C), and RNS60 (Fig. 1D) displayed readily nucleated nanobubbles. Compared with control solutions (NS and PNS60), the AFM images for RNS60 reveal approximately half the number of nanobubbles, with an average diameter that increased from 26 to 33 nm. This same relative pattern of nanobubble number and size was observed when positive potentials were applied to AFM surfaces with the same control solutions, suggesting that charge is involved. These data suggest that RNS60 has a different bulk nanobubble composition than its control solutions.
RNS60 Attenuates the Expression of Proinflammatory Molecules in Activated Mouse Microglia and Astroglia-Activated microglia and astroglia are known to produce excessive amounts of NO and proinflammatory cytokines, which have the potential of damaging neurons in neurodegenerative disorders (2,4). LPS is a prototype inducer of various proinflammatory molecules in different cell types, including mouse microglia. Therefore, mouse BV-2 microglial cells, preincubated with different doses of RNS60 for 2 h, were stimulated with LPS under serum-free conditions. Although at lower concentrations (1 and 2% v/v), RNS60 was not effective in inhibiting the production of NO (data not shown), and at higher concentrations, RNS60 markedly suppressed LPS-induced production of NO in microglial cells as evident from the estimation of nitrite ( Fig. 2A). Conversely, normal saline (NS), TCP-modified saline without excess oxygen (RNS10.3), and saline containing excess oxygen in the absence of TCP modification (PNS60) had no such inhibitory effect on the production of NO ( Fig. 2A). To understand the mechanism further, we investigated the effect of RNS60 on protein and mRNA levels of iNOS in microglial cells. It is clearly evident from Western blot in Fig. 2B and real time PCR in Fig. 2C that RNS60, but not NS, RNS10.3, and PNS60, inhibited LPS-induced protein and mRNA expression of iNOS in microglial cells.
Next, to investigate whether RNS60 suppresses the expression of iNOS in primary cells, primary mouse microglia preincubated with RNS60 or NS for 2 h were stimulated with LPS. As expected, LPS induced the expression of iNOS protein in primary microglia. RNS60, but not NS, markedly suppressed LPSinduced expression of iNOS in microglia (Fig. 3A). In addition to microglia, there are other glial cells in the CNS. Earlier we have demonstrated that IL-1␤ is a potent inducer of iNOS in astroglia (34). Therefore, we examined the effect of RNS60 on the expression of iNOS in IL-1␤-stimulated primary astroglia. Similar to its effect on microglia, RNS60 markedly inhibited IL-1␤-stimulated iNOS in primary mouse astroglia, although NS had no such effect (Fig. 3B).

RNS60 Inhibits IL-1␤-, H 2 O 2 -, and Double-stranded RNA (Poly(IC))-induced Production of NO in Microglial
Cells-Activated microglia are considered to play an important role in different neurodegenerative disorders (2,35). Because RNS60 inhibited LPS-induced production of NO and the expression of iNOS in microglia, we were prompted to investigate whether RNS60 was also capable of suppressing the production of NO in microglial cells stimulated with etiological reagents of various neurological disorders. Oxidative stress is a hallmark of many human disorders, including neurodegenerative disorders. BV-2 microglial cells were challenged with H 2 O 2 (for oxidative stress), double-stranded RNA in the form of poly(IC) (one of the etiological reagents for viral encephalopathy), and IL-1␤ (one of the etiological reagents for different neurodegenerative disorders). It is clearly evident from Fig. 2, D, F, and G, that IL-1␤, H 2 O 2 , and poly(IC) induced the production of NO in microglial cells. However, RNS60 markedly knocked down IL-1␤-, poly(IC)-, and H 2 O 2 -induced production of NO in microglial cells (Fig. 2, D, F, and G). However, NS had no such inhibitory effect (Fig. 2, D, F, and G), suggesting the specificity of the effect. In contrast, under the same experimental condition, RNS60 had no effect on IFN-␥-induced microglial production of NO (Fig. 2E), suggesting that IFN-␥ induces NO production in microglia via a RNS60-insensitive pathway.
RNS60 Inhibits Microglial Activation of NF-B-LPS and other inflammatory stimuli are known to induce iNOS expression via activation of NF-B (17, 36 -39). Because RNS60 attenuated the expression of iNOS in glial cells, we examined its effect on NF-B activation. Activation of NF-B was monitored  by both DNA binding and transcriptional activities of NF-B. Both IL-1␤ (Fig. 4, A and C) and LPS (Fig. 4, B and D) induced the DNA binding (Fig. 4, A and B) and transcriptional (Fig. 4, C and D) activities of NF-B in BV-2 microglial cells. However, RNS60, but not NS, inhibited IL-1␤and LPS-induced DNA binding and transcriptional activities of NF-B in microglial cells (Fig. 4, A-D). These results suggest that RNS60 attenuates the expression of iNOS by suppressing the activation of NF-B.
RNS60 Up-regulates IB␣ in Glial Cells-Although there are multiple mechanisms by which an anti-inflammatory molecule may suppress the activation of NF-B, in resting cells the classical p65:p50 heterodimer is arrested in the cytoplasm as an BV-2 cells were treated with RNS60 or NS (10% v/v) under serum-free conditions for different minute intervals followed by monitoring protein levels of IB␣ and p65 by Western blot (E) and mRNA levels of IB␣ by real time PCR (F). a , p Ͻ 0.001 versus control. G, BV-2 cells were treated with different concentrations of RNS60 for 1 h followed by monitoring protein levels of IB␣, p65, and p50 by Western blot. Actin was run as a loading control. H, bands were scanned and results presented as protein expression relative to actin. a , p Ͻ 0.001 versus control. I, primary mouse microglia were treated with RNS60 or NS for 1 h followed by monitoring IB␣ protein by double-label immunofluorescence using CD11b as a marker for microglia. J, BV-2 cells were treated with 10% (v/v) RNS60, RNS10.3, or PNS60 for 1 h followed by monitoring the protein level of IB␣ by Western blot. K, bands were scanned, and results are presented as protein expression relative to actin. a , p Ͻ 0.001 versus control.
inactive complex by IB␣ (15). It has also been demonstrated that newly synthesized IB␣ proteins accumulate in the cytoplasm as well as in the nucleus, where it reduces NF-B binding (40). Therefore, we hypothesized that RNS60 might up-regulate IB␣ to suppress the activation of classical NF-B heterodimer. Indeed, RNS60, but not NS, markedly up-regulated IB␣ protein in microglial cells in a time-dependent fashion (Fig. 4E). Significant increase in IB␣ protein was observed within only 30 min of RNS60 treatment (Fig. 4E). This effect was specific as RNS60 did not up-regulate the expression of RelA p65 (Fig. 4E). Real time PCR data also show marked timedependent increase in IB␣ mRNA by RNS60 but not NS (Fig.  4F). Similarly, RNS60 also dose-dependently increased the expression of IB␣, but not of p65 or p50 (Fig. 4, G and H). In fact, we observed down-regulation of p65 and p50 by RNS60 (Fig. 4, E and G). To confirm this IB␣ up-regulation, we also performed double-label immunofluorescence. As evident from Fig. 4I, RNS60, but not NS, markedly increased the expression of IB␣ in CD11b-positive primary microglia. To confirm that the effect of RNS60 on IB␣ was specific, we used RNS10.3 and PNS60. In contrast to RNS60, RNS10.3 and PNS60 remained unable to up-regulate IB␣ in microglia (Fig. 4, J and K).
To understand whether this IB␣ up-regulation was cell type-specific, we included primary astroglia. RNS60, but not NS, also up-regulated IB␣ in GFAP-positive astroglia in a dose-dependent manner (Fig. 5, A-C), demonstrating that the effect was not specific for microglia only.
RNS60 Activates PI 3-Kinase in Microglia-Next, we investigated mechanisms by which RNS60 may transduce signals for the up-regulation of IB␣. As activation of PI 3-kinase, a dual protein, and lipid kinase occurs in close association with the cell membrane, we investigated the effect of RNS60 on the activation of PI 3-kinase. Class IA PI 3-kinase, which is regulated by receptor tyrosine kinases, consists of a heterodimer of a regulatory 85-kDa subunit and a catalytic 110-kDa subunit (p85: p110␣/␤/␦). Class IB PI 3-kinase, however, consists of a dimer of a 101-kDa regulatory subunit and a p110␥ catalytic subunit (p101/p110␥). In addition to p101, p84 has also been identified as a regulatory subunit of p110␥ (41). Therefore, we first examined the presence of different subunits of PI 3-kinase in BV-2 microglial cells. As evident from Fig. 6A, p110␣, p110␤, p110␦, p110␥, and the new regulatory subunit p84 are present in BV-2 cells. Earlier, we also demonstrated the presence of p85␣ in BV-2 cells (21). When we monitored lipid kinase activity, we observed that RNS60, but not NS, markedly induced the activation of p85␣-associated PI 3-kinase within 15 min of treatment (Fig. 6, B and C), suggesting that RNS60 induces the activation of class IA PI 3-kinase. Phosphatidylinositol (3,4,5)-triphosphate (PIP 3 ) is the product of class I PI 3-kinase. Therefore, to confirm the activation of class I PI 3-kinase from another angle, we monitored PIP 3 levels in RNS60-treated microglia by immunofluorescence. Although RNS60 was unable to increase the level of PIP 3 within 5 min of stimulation, marked increase in PIP 3 was observed at 15 min of stimulation (Fig. 6D). However, consistent with that observed in lipid kinase activity, NS remained unable to up-regulate the level of PIP 3 (Fig. 6D). RNS60 did not, however, activate PI 3-kinase activity associated with either p101 (Fig. 6E) or p84 (Fig. 6F), suggesting that it is unable to activate class IB PI 3-kinase. Next, we tried to identify the catalytic subunit of class IA PI 3-kinase that is involved in this lipid kinase activity. By immuno-complex lipid kinase assay, we observed that RNS60induced activation of PI 3-kinase was due to p110␣ (Fig. 6G) and p110␤ (Fig. 6H) but not p110␦ (Fig. 6I).
RNS60 Activates Akt in Microglia-PI 3-kinases have been linked to a diverse group of cellular functions, and many of these functions are related to its ability to activate protein kinase B (PKB, also known as Akt). Therefore, we examined whether RNS60 could activate Akt in microglial cells. RNS60, but not NS, markedly induced the activation of Akt in microglia as evident from immunoblot (Fig. 7, A and B) and immunofluorescence analysis (Fig. 7C) of phospho-Akt, without increasing the level of total Akt (Fig. 7D). Although Akt is downstream of PI 3-kinase, recently it has been shown that the serine/thre-FIGURE 5. RNS60 up-regulates IB␣ in mouse astroglia. A, primary mouse astroglia were treated with different concentrations of RNS60 or NS for 1 h followed by monitoring the level of IB␣ protein by Western blot. Rel., relative. B, bands were scanned, and results are presented as protein expression relative to actin. a , p Ͻ 0.001 versus control. C, astroglia were treated with RNS60 or NS for 1 h followed by monitoring IB␣ protein by double-label immunofluorescence using GFAP as a marker for astroglia. Results represent three independent experiments. onine kinase IKBKE activates Akt through a PI 3-kinase-independent pathway (42). Therefore, we examined whether RNS60 required PI 3-kinase for the activation of Akt. Because RNS60 induced the activation of p110␣ and p110␤, we used chemical inhibitors that are specific for these catalytic subunits. Although GDC-0941 blocks the activity of p110␣, TGX-221 specifically inhibits p110␤ (43,44). Strong inhibition of RNS60induced phosphorylation of Akt by both GDC-0941 and TGX-221 suggests that RNS60 employs class IA PI 3-kinase p110␣/␤ for the activation of Akt in microglial cells.
RNS60 Requires the PI 3-Kinase-Akt Signaling Pathway to Suppress the Expression of iNOS, Attenuate the Activation of NF-B, and Up-regulate IB␣-Because RNS60 induced the activation of PI 3-kinase-Akt signaling pathway, we investigated the possibility that RNS60 was utilizing this pathway to exhibit its anti-inflammatory activity. We examined the effect of LY294002, an inhibitor of PI 3-kinase, and AktI, an inhibitor of Akt, on the RNS60-mediated anti-inflammatory effect. Both LY294002 and AktI abrogated the inhibitory effect of RNS60 on the induction of NO production (Fig. 8A) and the expression of iNOS protein (Fig. 8, B and C) in LPS-stimulated microglial cells. Similarly, LY294002 and AktI also abolished the RNS60mediated up-regulation of IB␣ protein and mRNA (Fig. 8, D  and E). To identify the catalytic subunit of PI 3-kinase involved in RNS60-mediated up-regulation of IB␣, we used GDC-0941 (an inhibitor of p110␣) (43), TGX-221 (an inhibitor of p110␤) (44), IC-87114 (an inhibitor of p110␦) (45), and AS-605240 (an inhibitor of p110␥) (46). Suppression of RNS60-mediated upregulation of IB␣ by GDC-0941 (supplemental Fig. 2, A and B) and TGX-221 (supplemental Fig. 2, C and D), but not IC-87114 (supplemental Fig. 2, E and F) and AS-605240 (supplemental Fig. 2, G and H), suggests that p110␣ and p110␤, but not p110␦ and p110␥, are involved in RNS60-mediated increase in RNS60. This has been supported by siRNA knockdown studies as well.

FIGURE 6. Activation of PI 3-kinase by RNS60 in mouse microglial cells.
A, presence of different subunits of PI 3-kinase (p110␣, p110␤, p110␦, p110␥, and p84) was examined in BV-2 cells by Western blot. BV-2 microglial cells were treated with either 10% (v/v) RNS60 or NS in serum-free DMEM/F-12. After 15 min of treatment, cells were lysed, immunoprecipitated with antibodies against either p85␣ (B and C), p101 (E), or p84 (F), and the lipid kinase activity of immunoprecipitated PI 3-kinase was assayed. Lipids were detected by exposure to film at Ϫ70°C (B, E, and F) and quantitated by densitometry (C). Results represent mean Ϯ S.D. of three separate experiments. a , p Ͻ 0.001 versus control. D, cells were treated with RNS60 or NS and at different time points, and the level of PIP 3 was monitored by immunofluorescence. DAPI was used to visualize nucleus. Cells were treated with RNS60 for 15 min, and cell lysates were immunoprecipitated with antibodies against p110␣ (G), p110␤ (H), or p110␦ (I) followed by assaying PI 3-kinase activity in immunoprecipitates. Results represent three independent experiments.
Next we investigated whether PI 3-kinase-Akt pathway was also playing a role in RNS60-mediated inhibition of NF-B activation. Abrogation of the inhibitory effect of RNS60 on NF-B activation in LPS-stimulated microglial cells by LY and AktI (Fig. 8F) suggests that RNS60 affects NF-B activity via PI 3-kinase-Akt pathway.
To confirm this observation further, we transfected BV-2 microglial cells with a dominant-negative mutant of p85␣. As evidenced by Western blot analysis (Fig. 9A) and real time PCR (Fig. 9B), RNS60 induced the expression of IB␣ protein and mRNA in empty vector-transfected, but not p85␣-transfected, microglial cells. These results suggest that RNS60 up-regulates IB␣ and exerts its anti-inflammatory effect in microglial cells via the PI 3-kinase-Akt signaling pathway.
Activation of PI 3-Kinase Is Sufficient for the Up-regulation of IB␣ in Microglia-In addition to activating PI 3-kinase, RNS60 may activate other signaling pathways, which may cooperate with the PI 3-kinase-Akt pathway for the up-regulation of IB␣. Therefore, we next investigated if the activation of PI 3-kinase alone by a constitutively active mutant of p110␣/␤ (p110*) was sufficient to induce the expression of IB␣ in FIGURE 7. RNS60 induces the activation of Akt in mouse microglial cells via PI 3-kinase. A, cells were stimulated with RNS60 or NS for different time periods followed by monitoring the levels of phospho-Akt and total Akt by Western blot. B, bands were scanned, and results are presented as phospho-Akt/total Akt. Results represent means Ϯ S.D. of three separate experiments. a , p Ͻ 0.001 versus control. Levels of phospho-Akt (C) and total Akt (D) were also monitored by double-label immunofluorescence (C and D). E, cells preincubated with TGX-221 and GDC-0941 for 30 min were stimulated with RNS60 followed by monitoring the levels of phospho-Akt and total Akt by Western blot. F, bands were scanned and results presented as phospho-Akt/total Akt. Rel., relative. a , p Ͻ 0.001 versus control. microglia. It is evident from Fig. 9, C and D, that overexpression of p110* was capable of increasing mRNA and protein expression of IB␣. However, p110-kd (a kinase-dead mutant) had no effect on the expression of IB␣ (Fig. 9, C and D). These results clearly suggest that activation of PI 3-kinase alone is sufficient to induce microglial expression of IB␣.
How Does the PI 3-Kinase-Akt Signaling Pathway Couple to the Up-regulation of IB␣ in Microglia?-Upon analysis of the IB␣ promoter using MatInspector, we have observed binding sites for many transcription factors, including NF-B and CREB. Accordingly, several studies have delineated a role of NF-B activation in the transcription of IB␣ (14,40). However, in this case, RNS60 inhibits the activation of NF-B (Fig. 2,  A and B), and so it is not likely to utilize this transcription factor for inducing IB␣ up-regulation. Therefore, we were prompted to investigate if RNS60 was using CREB for the up-regulation of IB␣. RNS60, but not NS, induced the phosphorylation of CREB (Fig. 10, A and B). Consistently, RNS60 also induced the DNA binding activity (Fig. 10C) and transcriptional activity (Fig. 10D) of CREB in microglial cells, suggesting that RNS60 alone is sufficient for the activation of CREB. Furthermore, both LY294002 and AktI markedly suppressed RNS60-induced phosphorylation of CREB (Fig. 10E) and the transcriptional activation of CREB (Fig. 10F), suggesting the involvement of the PI 3-kinase-Akt pathway for the activation of CREB. Next, to examine if RNS60 required CREB for the up-regulation of IB␣ in microglia, we utilized siRNA to silence CREB. CREB siRNA, but not control siRNA, decreased the expression of CREB protein (Fig. 11A) and abrogated RNS60-mediated up-regulation of IB␣ mRNA (Fig. 11B) and protein (Fig. 11C). Taken together, these results clearly demonstrate that RNS60 induces the activation of CREB via PI 3-kinase-Akt signaling pathway and that CREB is required for increased transcription of IB␣.

DISCUSSION
Activation of glial cells (microglia and astroglia) has been implicated in the pathogenesis of a variety of neurodegenerative diseases, including Alzheimer disease, Parkinson disease, Creutzfeldt-Jacob disease, HIV-associated dementia, stroke, and multiple sclerosis (2,9,35). It has been found that activated microglia and astroglia accumulate at sites of injury or plaques in neurodegenerative CNS (35,47). Although activated microglia scavenge dead cells from the CNS and secrete different neurotrophic factors for neuronal survival and activated astrocytes may have important beneficial effects in the recovery of injured CNS by actively monitoring and controlling the extracellular water, pH, and ion homeostasis, it is believed that severe activation of these brain cells causes various autoimmune responses leading to neuronal death and brain injury (2,35,47). During activation, microglia and astroglia express various genes related to inflammation, such as proinflammatory cytokines, proinflammatory enzymes, and proinflammatory adhesion molecules. Therefore, characterization of signaling pathways required for the activation of glial cells is an active area of investigation because compounds capable of antagonizing such signaling steps may have a therapeutic effect in neurodegenerative disorders.  Although other drugs and approaches exist to combat glial inflammation, here we introduce an elegantly simple, yet uniquely innovative, saline-based agent to suppress glial activation. Nanobubbles have been suggested for potential use in biomedical imaging and drug delivery (48,49), but they have not demonstrated direct biological effects. Nanoparticles are known to interact with cells based upon their charge and size, and we propose that nanobubble-based structures within RNS60 act in a similar way. These stable and charged entities may possess the necessary energetics to affect the membrane and its voltage-sensitive elements. Because RNS60 does not contain any active pharmaceutical ingredient, these proposed "charge stabilized nanostructures" would be the effector of activity at the cell membrane level.
Several lines of evidence presented in this study clearly support the conclusion that RNS60 attenuates the activation of glial cells. Our conclusion is based on the following observations. First, LPS, a prototype inducer of inflammation in many cell types including microglia, and other proinflammatory molecules induced the expression of iNOS in mouse microglia. However, RNS60 attenuated the expression of iNOS in activated microglia without altering cell survival, suggesting that this attenuation is not due to any cell death. This inhibition was also specific as other saline preparations like NS (normal saline of the same batch), RNS10.3 (TCP-modified saline without excess oxygen), and PNS60 (saline containing oxygen in the absence of TCP modification) had no effect. Second, we extended the study beyond LPS and examined whether RNS60 was capable of suppressing microglial expression of iNOS induced by other neurotoxins. It is important that RNS60, but not NS, attenuated microglial production of NO induced by various neurotoxins such as H 2 O 2 , poly(IC), and IL-1␤. Third, IL-1␤ is critical for inducing iNOS and activating astrocytes (34). RNS60, but not NS, also suppressed the expression of iNOS in primary mouse astrocytes. Because excess NO has been implicated in the pathogenesis of demyelinating and neurodegenerative diseases, our results provide a potentially important mechanism whereby RNS60 may ameliorate neural injury.
The signaling events required for the transcription of iNOS and proinflammatory cytokines are becoming clear. Although many transcription factors such as NF-B, C/EBP␤, AP-1, STAT, IRF-1, etc. play a role, activation of NF-B seems essential for the transcription of most of the proinflammatory molecules (17, 36 -39). Therefore, for a drug to exhibit anti-inflammatory effects, it is almost mandatory to attenuate the activation of NF-B. Treatment of microglia with RNS60 led to the inhibition of NF-B activation at both DNA binding and transcriptional levels. Accordingly, we have seen that RNS60 suppresses the production of NO in response to LPS, IL-1␤, H 2 O 2 , and poly(IC) because these stimuli require NF-B for the expression of iNOS (21). In contrast, RNS60 remained unable to inhibit IFN-␥-induced NO production. This is consistent with a finding by Kleinert et al. (50) that delineates an important role of STAT1␣ in IFN-␥-induced expression of iNOS. Because RNS60 inhibits the activation of NF-B, this TCPmodified saline is capable of attenuating the expression of only those proinflammatory molecules whose expression depends on the activation of NF-B. However, IFN-␥ induces the FIGURE 11. RNS60 up-regulates the expression of IB␣ in mouse microglial cells via CREB. A, BV-2 microglial cells plated in 12-well plates were transfected with 0.25 g of either CREB siRNA or control (Cont) siRNA, and after 24 h of transfection, the level of CREB protein was monitored by Western blot. Actin was run as a control. Bands were scanned, and results are presented as protein expression relative to actin (B). Rel., relative. Results are mean Ϯ S.D. of three different experiments. a , p Ͻ 0.001 versus control siRNA. Cells were transfected with 0.25 g of either CREB siRNA or control siRNA, and after 24 h of transfection, cells were treated with different concentrations of RNS60. After 1 h of treatment, the expression of IB␣ was monitored by real time PCR (C) and Western blot (D). Bands were scanned, and results are presented as protein expression relative to actin (E). Results represent mean Ϯ S.D. of three separate experiments. a , p Ͻ 0.001 versus no RNS60-control siRNA; b , p Ͻ 0.001 versus RNS60 -30-min control siRNA; c , p Ͻ 0.001 versus RNS60 -60-min control siRNA. expression of iNOS independent of NF-B (50); therefore, RNS60 is unable to inhibit IFN-␥-induced microglial production of NO. In resting cells, the classical p65:p50 heterodimer is arrested in the cytoplasm as an inactive complex by IB␣ (15). Therefore, IB␣ is an important endogenous anti-inflammatory molecule that selectively attenuates the activation of classical NF-B. These data clearly demonstrate that RNS60 is a rapid inducer of IB␣ mRNA and protein in glial cells.
How RNS60 up-regulates the expression of IB␣ is an interesting question. Because of the facts that RNS60 may interact at the cell membrane, that activation of PI 3-kinase occurs in close association with cell membrane, and that PI 3-kinase activation induces the expression of IB␣ mRNA in microglia (21), we were prompted to investigate the role of PI 3-kinase in RNS60mediated up-regulation of IB␣. Here, we have discovered that RNS60 specifically induces the activation of class IA, but not class IB, PI 3-kinase in microglia. PI 3-kinase activation by RNS60 was due to p85␣-associated p110␣ and p110␤. Accordingly, chemical inhibitors of p110␣ and p110␤, and ⌬p85␣, a dominant-negative mutant of p85␣, blocked RNS60-mediated up-regulation of IB␣. Alternatively, overexpression of p110*, a catalytically active mutant of p110␣/p110␤, alone increased the expression of IB␣. In contrast, overexpression of p110-kd, a kinase-dead mutant of p110␣/p110␤, had no such effect. Because in most cases, activation of PI 3-kinase leads to Akt activation, we also examined the role of Akt in the up-regulation of IB␣. RNS60 alone induced the phosphorylation of Akt, and this was not due to increased expression of Akt.
We also went further downstream to describe how RNS60 coupled type IA PI 3-kinase-Akt pathway to the transcription of IB␣. Interestingly, we have delineated that RNS60 activates CREB via PI 3-kinase-Akt pathway and then CREB drives the transcription of IB␣ in response to RNS60. Rapid activation of the PI3K-Akt pathway suggests that RNS60 treatment may be beneficial for other important cellular pathologies involving this pathway. The exact mechanism by which RNS60 may interact with the cell membrane to activate PI3K is a challenging question that is currently under active investigation.
In summary, we have demonstrated a saline-based agent devoid of an active pharmaceutical ingredient to exhibit antiinflammatory effects via class IA PI-3K-Akt-CREB-mediated up-regulation of IB␣ and inhibition of NF-B activation. Our results highlight novel biological properties of a physically modified saline and indicate that such approaches may be developed for therapeutic intervention in neuroinflammatory or neurodegenerative disorders as primary or adjunct therapy.