Biochimica et Biophysica Acta 1830 (2013) 2981–2988

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbagen

Edible blue-green algae reduce the production of pro-inflammatorycytokines by inhibiting NF-κB pathway inmacrophages and splenocytes

Chai Siah Ku a, Tho X. Pham a, Youngki Park a, Bohkyung Kim a, Min Sun Shin b, Insoo Kang b, Jiyoung Lee a,⁎a Department of Nutritional Sciences, University of Connecticut, Storrs, CT 06269, USAb Department of Internal Medicine, Yale University School of Medicine, New Haven, CT 06520, USA

Abbreviations: apoE−/−, apolipoprotein E knock outbone marrow-derived macrophages; CBP, CREB-bindingelement-binding protein; C-PC, C-phycocyanin; CVD,histone acetyltransferase; HDAC, histone deacetylases;IL, interleukins; LPS, lipopolysaccharide; NLS, nuclear loccommune var. sphaeroides Kützing; NOE, NO lipid extractlipid extract; TNFα, tumor necrosis factor α; TSA, tricho⁎ Corresponding author: Tel.: +1 860 486 1827; fax

E-mail address: Ji-young.lee@uconn.edu (J. Lee).

0304-4165/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.bbagen.2013.01.018

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 24 July 2012Received in revised form 29 December 2012Accepted 15 January 2013Available online 26 January 2013

Keywords:Blue-green algaeAnti-inflammationRAW 264.7 macrophageNF-κBCytokine arrayHistone deacetylation

Background: Chronic inflammation contributes to the development of pathological disorders including insulinresistance and atherosclerosis. Identification of anti-inflammatory natural products can prevent the inflam-matory diseases.Methods: Anti-inflammatory effects of blue-green algae (BGA), i.e., Nostoc commune var. sphaeroides Kützing(NO) and Spirulina platensis (SP), were compared in RAW 264.7 and mouse bone marrow-derived macro-phages (BMM) as well as splenocytes from apolipoprotein E knockout (apoE−/−) mice fed BGA.Results: When macrophages pretreated with 100 μg/ml NO lipid extract (NOE) or SP lipid extract (SPE) wereactivated by lipopolysaccharide (LPS), expression and secretion of pro-inflammatory cytokines, such astumor necrosis factor α (TNFα), interleukin 1β (IL-1β), and IL-6, were significantly repressed. NOE andSPE also significantly repressed the expression of TNFα and IL-1β in BMM. LPS-induced secretion of IL-6was lower in splenocytes from apoE−/− fed an atherogenic diet containing 5% NO or SP for 12 weeks. InRAW 264.7 macrophages, NOE and SPE markedly decreased nuclear translocation of NF-κB. The degree of

repression of pro-inflammatory gene expression by algal extracts was much stronger than that of SN50, aninhibitor of NF-κB nuclear translocation. Trichostatin A, a pan histone deacetylase inhibitor, increased basalexpression of IL-1β and attenuated the repression of the gene expression by SPE. SPE significantlydown-regulated mRNA abundance of 11 HDAC isoforms, consequently increasing acetylated histone 3 levels.Conclusion: NOE and SPE repress pro-inflammatory cytokine expression and secretion in macrophages andsplenocytes via inhibition of NF-κB pathway. Histone acetylation state is likely involved in the inhibition.General significance: This study underscores natural products can exert anti-inflammatory effects by epige-netic modifications such as histone acetylation.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Chronic inflammation is an underlying cause of several pathologicaldisorders including insulin resistance, dementia, rheumatoid arthritis,atherosclerosis, and cancer [1,2]. Non-steroidal anti-inflammatorydrugs (NSAID) are generally used to treat acute and chronic inflam-matory conditions. However, due to their adverse side-effects andincreased cardiovascular disease (CVD) risk associated with chronicuse of several NSAID [3], there is a critical need to identify natural prod-ucts with anti-inflammatory properties. Several anti-inflammatory

; BGA, blue-green algae; BMM,protein; CREB, cAMP-responsecardiovascular disease; HAT,IκBα, inhibitor of kappa B α;alization sequence; NO, Nostoc; SP, Spirulina platensis; SPE, SPstatin A: +1 860 486 3674.

rights reserved.

bioactive compounds that have been extensively studied thus far in-clude curcumin, resveratrol, anthocyanin, and green tea polyphenols[4–8].

Nuclear factor κ B (NF-κB) is a major transcriptional factors respon-sible for pro-inflammatory cytokine production, such as tumor necrosisfactor α (TNFα), interleukins (ILs), inducible nitric oxide synthase andcyclooxygenase-2, under inflammatory conditions [9,10]. NF-κB existsas homo- or heterodimers consisting of five subunits of Rel family,i.e., p50, p52, p65, c-Rel and RelB [11]. In an unstimulated state, NF-κBis present in the cytoplasm bound with inhibitor of κB α (IκBα),which masks the nuclear localization sequence of p65 [12] and NF-κBactivation largely depends on IκB kinase β (IKKβ). Extracellular inflam-matory signals including lipopolysaccharide (LPS), pro-inflammatorycytokines, reactive oxygen species, advanced glycation end products,and oxidized low-density lipoprotein trigger the phosphorylation ofIKKβ, which in turn phosphorylates IκBα for ubiquitination and degra-dation by proteasome [13–18]. This event liberates NF-κB to enter thenucleus for the induction of pro-inflammatory gene expression [19–24].

Blue-green algae (BGA), also known as cyanobacteria, are one ofthe most primitive forms of photosynthetic prokaryotes. They have

2982 C.S. Ku et al. / Biochimica et Biophysica Acta 1830 (2013) 2981–2988

been consumed as food or medicine for centuries and human con-sumption of BGA was recorded in the 14th century during the Azteccivilization [25]. They are recognized for their protective effectsagainst viral and bacterial infections, cancer, allergy, diabetes, inflam-mation and hyperlipidemia [26–29]. At present, Spirulina platensis(SP) is the most commonly consumed and commercialized BGAspecies. We have previously reported that lipid extract of Nostoccommune var. sphaeroides Kützing (NO), another BGA species,inhibited NF-κB DNA binding activity and consequently repressedthe pro-inflammatory gene expression in RAW 264.7 macrophages[30]. In this study, we compared anti-inflammatory effects of twoBGA species using several model systems: RAW 264.7 macrophages,bone marrow-derived macrophages (BMM), and splenocytes isolatedfrom BGA-fed apolipoprotein E knockout (apoE−/−) mice. We foundthat in all of three systems, NO and SP either as lipid extract oras whole algae repressed pro-inflammatory gene expression andproduction.

2. Methods and materials

2.1. Preparation of BGA lipid extraction

BGA powder was kindly provided by Algaen Corporation (WinstonSalam, NC) for NO and by Earthrise Nutritionals (Irvine, CA) for SP.Lipid extracts of BGA into chloroform/methanol (1:2) were preparedas previously described [30,31]. Lipid extracts were stored under N2

gas at −20 °C for short term or at −80 °C for long term. The lipidextracts were dried down under N2 to remove solvents and thendissolved in cell medium by sonication.

2.2. Bone marrow isolation and macrophage differentiation

BMM were isolated from C57BL/6J mice (Jackson Laboratory, Barharbor, ME). Briefly, mouse legs were removed from the hip jointand cleaned, after which the femur and tibia were cut at the tip andsubsequently bone marrow was collected by centrifugation. Thebone marrow cells were differentiated into macrophages in low-glucose DMEM containing 2 mmol/L L-glutamine, 1 mmol/L sodiumpyruvate, 300 U/mL penicillin, 300 μg/mL streptomycin, 20% FBS,and 30% L929 cell conditionedmedia. L929 cells were generously pro-vided by Dr. John Parks (Wake Forest University School of Medicine,Winston-Salem, NC). BMM were incubated in a humidified chamberat 37 °C with 5% CO2 and cell culture medium was changed every3 days for 7 days until they became confluent.

2.3. Macrophage cell culture and treatment

RAW 264.7 macrophages (ATCC, Manasas, VA) and BMM wereincubated in RPMI-1640 containing 10% FBS, 100 U/mL penicillin,100 μg/mL streptomycin, 1 x vitamins and 2 mmol/L L-glutamine ina 37 °C humidified chamber with a supply of 5% CO2. Macrophageswere incubated with 0–100 μg/mL of NO lipid extract (NOE) or SPlipid extract (SPE) for 12 h and subsequently activated by LPS(Sigma-Aldrich, St. Louis, MO) at 100 ng/mL concentration for addi-tional 18 h. Cells and medium were collected for pro-inflammatorycytokine expression and secretion.

SN50, a cell permeable inhibitor specific for NF-κB nuclear translo-cation, was purchased from Enzo Life Science (Plymouth meeting, PA)and dissolved in sterile water. RAW 264.7 macrophages were incu-bated with 50 μg/mL of NOE or SPE for 11 h followed by 1 h incuba-tion with 50 μg/mL of SN50. Subsequently, the cells were treatedwith LPS (100 ng/mL) in the presence of algal lipid extract andSN50 for additional 3 h. For the experiment with trichostatin A(TSA), a pan histone deacetylase (HDAC) inhibitor, RAW 264.7macrophages were treated with algal lipid extract (100 μg/mL) and50 nmol/L of TSA for 12 h prior to LPS stimulation for 7 h. All cell

culture supplies were purchased from Thermo Scientific Hyclone(Logan, UT), unless stated otherwise.

For all cell culture experiments, cells without exposure to anyalgal extract were considered as control. Algal extracts were incorpo-rated into cell culture medium by sonication and therefore no solventwas used as a vehicle.

2.4. Splenocyte isolation and culture

Spleens were harvested frommale apoE−/− mice (Jackson Labora-tory) fed a high fat/high cholesterol (15% fat, 0.2% cholesterol by wt)containing 5% NO or SP by wt for 12 wk from 8 wk of age. After beinganesthetized with ketamine HCl (50 mg/kg)/xylazine (10 mg/kg)and subsequently euthanized by cardiac puncture and cervical dis-location, spleen of each mouse was excised aseptically and groundin RPMI-1640 medium containing 10% FBS, 100 U/mL penicillin and100 μg/mL streptomycin. After removal of red blood cells by apre-warmed RBC lysis buffer (eBioScience, San Diego, CA), the cellswere resuspended in the medium and centrifuged at 2000 rpm for5 min. Cell pellet was resuspended in PBS and filtered through40 μm strainer (BD Biosciences, San Jose, CA). After washing withPBS, cells were resuspended in RPMI-1640 complete media andplated at a density of 1×106/0.5 mL for experiments. Cells were in-cubated with LPS (500 ng/mL) for 20 h and IL-6 concentrations inthe media were measure by ELISA (eBioScience) according to themanufacturer's instruction. All procedures were approved by theInstitutional Animal Care and Use Committee of the University ofConnecticut.

2.5. Quantitative realtime PCR (qRT-PCR)

Total RNA was extracted using TRIzol reagent (Life technology,Carlsbad, CA) and cDNA synthesis was performed as previouslydescribed [30,31]. qRT-PCR was conducted using Bio-Rad CFX96Real-Time system (Bio-Rad, Hercules CA). Primers were designedaccording to GenBank database and the sequences were previouslypublished [30].

2.6. Cytokine array

RAW 264.7 macrophages were pretreated with 100 μg/mL of NOEor SPE for 12 h, after which they were incubated with LPS for 18 h.Conditioned medium was collected and centrifuged at 12,000 ×g for5 min to remove any cell debris or dead cells. Secretion of cytokineswas assessed by RayBio Mouse Cytokine Antibody Array (RayBiotech,Norcross, GA) according to the manufacturer's protocol. Signals werecaptured using ChemiDoc XRS+ system (Bio-Rad, Hercules, CA).

2.7. NF-κB translocation by Western blot analysis

RAW 264.7 macrophages incubated with 100 μg/mL of NOE or SPEfor 12 h were activated by LPS for 1 or 2 h. Nuclear and cytoplasmicfractions of the cells were separated by using Cayman nuclear extrac-tion kit (Ann Arbor, MI) and Western blot analysis was conduct as wepreviously described [30,31]. Polyclonal anti-p65 and anti-GAPDHantibodies were purchased from Santa Cruz Biotechnology (SantaCruz, CA). Mouse monoclonal antibody against TATA binding proteinwas obtained from Abcam (Cambridge, MA) and used as a loadingcontrol for nuclear fraction. Protein expression were detected usingWestpico horseradish peroxidase chemiluminescence (Pierce, Rock-ford, IL) and imaged using a Chemidoc XRS+ system (Bio-Rad) andImage Lab software (Bio-Rad).

Fig. 1. Repression of pro-inflammatory expression by NOE and SPE in RAW 264.7 macrophages. Cells were pre-treated with 100 μg/mL of NOE or SPE for 12 h followed byco-incubation of 100 ng/mL LPS and algal extracts for additional 18 h. qRT-PCR analysis was conducted to measure mRNA levels of TNFα, IL-1β and IL-6. Values are expressedas mean±SEM, Pb0.05, n=15–18.

2983C.S. Ku et al. / Biochimica et Biophysica Acta 1830 (2013) 2981–2988

2.8. Acetylation of histone 3 (H3)

After incubation with 100 μg/mL of SPE for 12 h and LPS for ad-ditional 18 h, RAW 264.7 macrophages were washed twice withice-cold PBS and cells were collected by centrifugation at 12,000 ×gfor 5 min. Cells were then resuspended in Triton X extraction buffercontaining 0.5% Trition X-100 (v/v), 0.02% NaN3 (w/v) in PBS and 1X protease inhibitor cocktail (Calbiochem, San Diego, CA) for 10 minon ice. Subsequently, the cell lysates were centrifuged for 10 min at4 °C to pellet nuclei and the pellets were then acid extracted in0.2 N HCl overnight at 4 °C. After centrifugation at 12,000 ×g for10 min, the supernatants were collected for Western blot analysisas described above. Antibodies against total H3 and acetyl-H3K9were purchased from Cell Signaling Technology (Danvers, MA).

2.9. Statistical analysis

One-way analysis of variance (ANOVA) and Newman-Keulspairwise comparisons to detect significant difference between groupswere performed using GraphPad InStat 5 (GraphPad Software, La

Fig. 2. Inhibition of pro-inflammatory cytokine secretion by NOE and SPE in RAW 264.7 macactivated by 100 ng/mL LPS for 18 h. Bottom chart indicates location of cytokines in thecolony-stimulating factor; IL, interleukins; IFNγ, interferon gamma; MCP-1, monocyte chcell factor; sTNFR1, soluble tumor necrosis factor receptor 1; TNFα, tumor necrosis factor a

Jolla, CA). P values less than 0.05 were considered significant. Valueswere expressed as mean±SEM.

3. Results

3.1. Pro-inflammatory cytokine expression and secretion were repressedby NOE and SPE in macrophages

We examined the effects of NOE and SPE on pro-inflammatorycytokine expression. In RAW 264.7 macrophages, both algal lipid ex-tracts significantly repressed LPS-induced TNFα, IL-1β and IL-6 mRNAexpression compared with control and their repressive effectsshowed a similar potency (Fig. 1). Significant reductions in the geneexpression were observed in a dose-dependent manner as low as25 μg/mL (data not shown).

Furthermore, NOE and SPE markedly decreased the secretion ofTNFα, IL-6, granulocyte colony-stimulating factor, granulocyte mac-rophage colony-stimulating factor, and monocyte chemoattractantprotein 1 (Fig. 2).

rophages. Cells were incubated with 50 μg/mL of NOE or SPE for 12 h and subsequentlyarray. GCSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte macrophageemoattractant protein 1; RANTES (CCL5), chemokine C-C motive ligand 5; SCF, stemlpha; TBP, TNF binding protein; VEGF, vascular endothelial growth factor.

Fig. 3. Inhibitory effects of TNFα and IL-1β expression byNOE and SPE in BMM. Bonemar-row derived macrophages isolated from C57BL/6 J mice were treated with 100 μg/mL ofNOE or SPE for 12 h, after which theywere co-incubatedwith 100 ng/mL for 18 h. Valuesare expressed as mean±SEM, Pb0.05, n=3.

Fig. 4. Decreased IL-6 secretion in spelnocytes from apoE−/− mice fed NO or SP.Splenocytes were isolated from apoE−/− mice fed an atherogenic diet supplementedwith 5% NO or SP for 12 weeks. The cells were treated with and without 500 ng/mLof LPS for 20 h. Values are expressed as mean±SEM, Pb0.05, n=5.

2984 C.S. Ku et al. / Biochimica et Biophysica Acta 1830 (2013) 2981–2988

Consistent with our observations in RAW 264.7 macrophages,BMM treated with NOE and SPE showed significantly lower expres-sion of TNFα and IL-1β compared with control BMM (Fig. 3).

3.2. IL-6 secretion was decreased from splenocytes of mice fed NO and SPsupplements

Fresh splenocytes were isolated from apoE−/− mice fed an athero-genic diet containing 5% of NO or SP for 12 wk. When the splenocyteswere challenged by LPS for 20 h, cells from NO or SP-fed mice showedsignificantly lower IL-6 levels in cell medium than control mice(Fig. 4).

Fig. 5. Attenuation of nuclear translocation of NF-κB p65 by NOE and SPE in RAW 264.7 macrthey were activated by 100 ng/mL LPS for 1 or 2 h. Cytoplasmic and nuclear fractions wereand GAPDH were used as a nuclear and cytoplasmic marker, respectively.

3.3. Algal lipid extracts inhibited NF-κB nuclear translocation inmacrophages

Translocation of NF-κB p50/p65 dimer to the nucleus is one ofthe major steps for its activation to increase transcription of pro-inflammatory cytokines. We previously reported that NOE signifi-cantly decreased NF-κB p65 DNA binding activity [30]. To address ifthe decrease in DNA binding activity is, at least in part, due to theinhibition of p65 nuclear translocation and to compare a potency ofthe repression between NOE and SPE, RAW 264.7 macrophagespre-treated with algal lipid extracts were stimulated by LPS for 1 or2 h. At 1 h LPS stimulation, control cells showed markedly increasedp65 in the nucleus with p65 being non-detectible in the cytoplasmicfraction (Fig. 5). In contrast, macrophages pretreated with NOE andSPE had less p65 translocation to the nucleus and detectible amountof p65 was present in the cytoplasm. The similar trend was shownat 2 h LPS stimulation.

3.4. Inhibition of NF-κB nuclear translocation could not solely account forthe repression of pro-inflammatory cytokine expression by NOE and SPEin macrophages

Although translocation of NF-κB to the nucleus is a major regula-tory step for its transcriptional activity, other posttranslational modi-fications in NF-κB have been recently recognized as importantregulatory mechanisms for NF-κB activity [32,33]. To evaluate contri-bution of nuclear translocation of p65 to the repressive effect of NOEand SPE on TNFα and IL-1β expression, we used SN50, an inhibitorspecific for NF-κB translocation to the nucleus. SN50 binds to the nu-clear localization sequence (NLS) of p65 and therefore p65 cannot berecognized by karyopherin α, resulting in cytoplasmic sequestration.With the treatment of SN50, LPS-induced nuclear translocation of p65was blocked (data not shown). Although SN50 significantly loweredmRNA levels of TNFα and IL-1β compared with control, repressionof the gene expression was markedly stronger in NOE or SPE-treated RAW 264.7 macrophages than SN50-treated cells (Fig. 6A).Interestingly, TSA significantly increased basal expression of IL-1βand TNFα and furthermore the repressive effect of SPE on the geneexpression was attenuated by TSA (Fig. 6B). NOE also elicited thesimilar results (data not shown).

3.5. Down regulation of HDAC expression and subsequent increases inacetylated histone 3 by SPE in macrophages

Inhibition of HDAC by TSA attenuated the repression of pro-inflammatory gene expression by SPE in LPS-stimulated macro-phages, suggesting HDAC may play a role in the anti-inflammatoryfunction of the algal extract. Therefore, we next evaluated if SPE canalter the expression of 11 HDAC isoforms in RAW 264.7 macrophages.mRNA abundance of all HDAC isoforms was significantly decreased bySPE in both unstimulated and LPS-stimulated macrophages (Fig. 7A).HDAC expression was also significantly lowered by LPS. Consistent

ophage. Cells were pre-treated with 50 μg/mL of NOE or SPE for 12 h and subsequentlyisolated and Western blot analysis for p65 was conducted. TATA binding protein (TBP)

A B

Fig. 6. Alternative mechanisms for anti-inflammatory effects of NOE and SPE in RAW 264.7 macrophages. A. Effect of SN50 on pro-inflammatory gene expression. Cells were incu-bated with NOE or SPE (50 μg/mL) for 12 h. Subsequently, they were incubated with 50 μg/mL of SN50, a NF-κB translocation inhibitor, for 1 h, after which they were activated with100 ng/mL LPS for 3 h. qRT-PCR analysis was conducted to measure mRNA levels of TNFα and IL-1β. Values are expressed as mean±SEM, Pb0.05, n=3. B. RAW 264.7 macrophageswere treated with NOE or SPE (100 μg/mL) and 50 nmol/L of TSA for 12 h and then they were activated by LPS (100 ng/mL) for 7 h. qRT-PCR analysis was conducted to measuremRNA levels of TNFα and IL-1β. Value are expressed as mean±SEM, Pb0.05, n=3.

2985C.S. Ku et al. / Biochimica et Biophysica Acta 1830 (2013) 2981–2988

with significant decreases in HDAC mRNA levels by SPE in bothunstimulated and LPS-stimulated macrophages, SPE increased thelevel of acetylated H3 by ~2-fold (Fig. 7B). Interestingly, althoughLPS repressed HDAC mRNA levels to a similar extent to SPE, therewas only ~40% increase in acetylated H3 in LPS-stimulated cellscompared with controls.

4. Discussion

Chronic low-grade inflammation is causally linked to the patho-genesis of several metabolic diseases, such as insulin resistance,metabolic syndrome, type 2 diabetes, CVD and non-alcoholic fattyliver disease. NSAID are commonly used to treat acute and chronicinflammatory conditions, but not without their adverse effects [3].

Fig. 7. Regulation of HDAC expression and levels of acetyl H3 by SPE in RAW 264.7 macrophsubsequently with 100 ng/mL LPS and algal extracts for additional 18 h. qRT-PCR analysimean±SEM, Pb0.05, n=6. B. Cells were treated with SPE and LPS as described above and Wtative blot of 2 separate experiments is shown. Densitometry analysis was conducted for thtreatment are listed in the table.

Therefore, identification of anti-inflammatory natural products withminimal side-effects, yet effective in blocking pathways leading tochronic inflammation is critically needed. We previously reportedthat NOE exerts ant-inflammatory effects by inhibiting NF-κB DNAbinding activity in macrophages [30]. In the present study, we furtherdemonstrated that SPE inhibits pro-inflammatory mediator expres-sion and secretion to a similar extent to NOE in macrophage cellline as well as primary mouse macrophages. Furthermore, the anti-inflammatory was confirmed in splenocytes isolated from NO orSP-fed apoE−/− mice, demonstrating the protection against inflam-mation by the two BGA species is likely to occur in vivo.

Humans have a long history of BGA consumption as food ormedicine for hundreds of years, yet scientific understanding of theirhealth-promoting properties only began ~30 years ago. Currently,

ages. A. RAW 264.7 macrophages were pre-treated with 100 μg/mL of SPE for 12 h ands was conducted to detect HDAC isoform mRNA abundance. Values are expressed asestern blot analysis was conducted to detect total and acetyl H3 levels. One represen-

e ratios of acetyl H3K9 to total H3. A ratio of each band and an averaged value of each

2986 C.S. Ku et al. / Biochimica et Biophysica Acta 1830 (2013) 2981–2988

SP is the most commonly consumed and commercialized BGA species[25]. NO is another edible BGA species with several purported healthbenefits. However, there has been a concern on safety of BGA forhuman consumption. Certain BGA species produce toxins, such asmicrocystins, nodularins, saxitoxins, anatoxins, and β-methylamino-L-alanine and therefore naturally harvested BGA products may con-tain the toxins due to contamination with toxin-producing algae spe-cies during harvest [34–37]. The two BGA species used in the presentstudy are cultivated in the controlled environment to minimize anytoxin contamination. We previously demonstrated that SP and NOare free of microcystins, the major algal toxin, and long-term sup-plementation of the two BGA did not show any adverse side-effectsin mice [37]. Furthermore, in 2011, The Dietary Supplements Infor-mation Expert Committee assigned a Class A safety rating for SP andpermitted the admission of quality monographs for SP in the UnitedStates Pharmacopeia and National Formulary [38]. As safety is prereq-uisite to investigation on health-promoting properties of any naturalproducts, we believe the two BGA species may be safe natural prod-ucts for human consumption to gain health benefits if their functionsare scientifically proven.

Studies have demonstrated that BGA consumption promotes immu-nity and protects against inflammatory diseases such as colitis, arthritis,and allergic rhinitis in animals and humans [39–42]. In particular,Rasool et al. [42] showed that Spirulina fusiformis supplementationmin-imized inflammatory response to near normal condition in miceinjected with inflammatory inducer. We previously showed that NOErepressed the expression and secretion of pro-inflammatory cytokinesuch as TNFα and IL-1β in RAW 264.7 macrophages [30]. Consistentwith our previous findings, NOE significantly decreased mRNA levelsof the pro-inflammatory cytokines not only in RAW264.7macrophagesbut also in BMM. SPE also elicited the same anti-inflammatory effectwith similar potency to NOE. Phycocyanin (PC) and γ-linoleic acid arethemost well-known anti-inflammatory components in BGA. In partic-ular, C-PC, a natural blue pigment (~14% of SP dry weight) [40,43,44],has shown to have a strong inhibitory property against inflammationin macrophages [45,46] and in animal models [47,48]. We found thatin NOE, unsaturated fatty acids, which account for ~15% of the extractmass, are partly responsible for the repression of pro-inflammatorycytokine production but other compounds are required to achieve fulldegree of repression in RAW 264.7 macrophages [30]. Studies onidentification of anti-inflammatory compounds in NOE and SPE arecurrently underway. As both algal extracts elicits marked repressionof inflammation, the identified compounds may be used as anti-inflammatory drugs alternative to NSAID.

In addition to the inhibitory role of NOE and SPE in the productionof pro-inflammatory mediators shown in vitro using RAW 264.7 mac-rophages and BMM, we sought to evaluate if the BGA supplementa-tion can accumulate enough amount of putative anti-inflammatorycompounds in other immune-modulating cells in mice to inhibit in-flammation. It has been shown that spleen is a monocyte reservoirand plays an important role not only in monocyte productionbut also in monocyte/macrophage clearance during inflammation[49,50]. Therefore, we isolated splenocytes from apoE−/− mice, acommon mouse model for atherosclerosis, that were fed an athero-genic diet supplemented with 5% of NO or SP for 12 wk. When thesplenocytes from mice fed a NO or SP-containing diet were stimulat-ed by LPS, IL-6 secretion was significantly lower than cells from con-trol mice. This result suggests that NO and SP supplementation can bebeneficial to inhibit inflammation in vivo.

Mounting evidence suggests that activation of NF-κB pathwayis a major underlying event for the development of inflammatorydiseases and its inhibition can decrease the disease progression[51–53]. Epidemiological and clinical studies have demonstrated the ef-fect of polyphenol-rich foods, such as green tea and extra virgin oliveoils, on the inhibition of NF-κB activation [54]. We previously demon-strated that NOE repressed pro-inflammatory cytokine production, at

least in part, by inhibiting NF-κB p65 DNA binding activity in RAW264.7 macrophages [30]. In the present study, both NOE and SPEinhibited nuclear translocation of NF-κB p65 in LPS-stimulated macro-phages. However, of interest is that blockage of p65 translocation tothe nucleus by a chemical inhibitor was significantly less effective inrepressing TNFα and IL-1β expression than NOE and SPE. This suggeststhat mechanisms other than NF-κB nuclear translocation may exist forthe algal extract to elicit the drastic reduction in the expression ofpro-inflammatory cytokine expression.

Although NF-κB nuclear translocation is the best-known mecha-nism for its activation [12], several lines of evidence indicate that post-translational modifications in p65 subunit, such as phosphorylationand acetylation, impact its activity by altering DNA binding, interac-tion with IκBα in the nucleus, and association with chromatin struc-ture modifiers such as histone acetyltransferases (HAT) and HDAC[32,33]. It is our particular interest that acetylation of p65 at Lys218,221 and 310 by p300 and cAMP response element binding (CREB)binding protein (CBP) regulates its transcriptional activity by increas-ing p65 DNA binding and by inhibiting its association with nuclearIκBα for the prevention of NF-κB relocation to the cytoplasm [55]. Incontrast, deacetylation at Lys310 by corepressor complexes inhibitsp65 activity [56,57]. In the present study, we observed that basalIL-1β and TNFα mRNA levels were significantly induced by TSA, apan HDAC inhibitor, in unstimulated macrophages. Bailey and Ghosh[58] suggested that NF-κB in association with a corepressor complexactively represses pro-inflammatory gene expression in unstimulatedmacrophages and the corepressor complex is composed of nuclearreceptor corepressor, HDAC3 and others. Therefore, the induction ofIL-1β and TNFα expression by TSA in the present study is well inline with their findings that active transcriptional repression of thepro-inflammatory genes occurs when macrophages are not stimulat-ed by inflammatory insults. Furthermore, we observed that repressionof the pro-inflammatory gene expression by NOE and SPE was signif-icantly ameliorated by TSA in LPS-activated RAW 264.7 macrophages.The data suggest that HDAC may be involved in basal transrepressionof IL-1β and TNFα when there are no extracellular inflammatory in-sults, while it mediates the repressive effect of the algal extracts onthe expression of the pro-inflammatory genes in the activated macro-phages.We found that themRNA expression of 11 HDAC isoformswassignificantly repressed by SPE in both unstimulated and activatedmacrophages. Among 11 HDAC isoforms, HDAC 1, 2, 3, 4, 5 and 9 aremajor isoforms in RAW 264.7 macrophages (data not shown). Highlevels of acetyl H3 observed in the SPE-treated cells are likely due tothe down-regulation of HDAC expression although we cannot ruleout possible alterations in HAT by SPE. LPS also repressed HDACmRNA levels to a similar extent to SPE. Although a degree of repres-sion of HDAC mRNA by SPE and LPS was similar, acetyl H3 levelswere markedly higher in SPE-treated cells than LPS-activated cells. Itcan be speculated that LPS and SPE may differentially regulate HDACexpression at the posttranscriptional levels and/or HAT activity mayalso be altered by the treatments to alter H3 acetylation. A conundrumis that both unstimulated cells and SPE treatment in LPS-treated cellsshowed low pro-inflammatory gene expression but their acetyl H3levels were very different. It seems important to understand the effectof SPE and LPS on the expression/activity of each HDAC isoform ratherthan overall changes in H3 acetylation state for gaining mechanisticinsights into critical roles of HDAC in inflammation. Future study iswarranted to identify specific HDAC isoforms that are involved in theregulation of the inflammatory gene expression and to determinehow their activity is regulated such as cellular localization, posttrans-lational modification and association with other regulatory proteins.The information can be used to develop effective therapeutic targetsto inhibit inflammation.

In conclusion, we found that NO and SP either as a whole or a lipidextract have an anti-inflammatory effect using several model sys-tems, i.e., RAW 264.7 macrophages, BMM and mouse splenocytes.

2987C.S. Ku et al. / Biochimica et Biophysica Acta 1830 (2013) 2981–2988

Repression of pro-inflammatory mediator expression by the BGA ismediated partly by inhibiting NF-κB nuclear translocation. In addi-tion, our results also support that HDAC and modulations in chroma-tin structure are likely involved in the anti-inflammatory function ofNO and SP. Dysregulation of pro-inflammatory mediator productionfrommacrophages is a major contributor to the development of path-ological conditions related to inflammation such as insulin resistanceand atherosclerosis. Therefore, the two BGA species tested in thepresent study may be used as a natural product to prevent inflamma-tory diseases.

Acknowledgements

C. S. Ku, Y. Park, B. Kim, T.X. Pham and M. Shin conducted experi-ments; C. S. Ku, and Y. Park contributed to manuscript preparation; I.Kang contributed to data analysis; and J. Lee designed the study,analyzed data and largely contributed to manuscript preparation. Allauthors read and approved the final manuscript. This work wassupported by National Institute Health grant R21AT005152 and fundfrom College of Agriculture and Natural Resources at the Universityof Connecticut to J. Lee.

References

[1] H.Y. Chung, M. Cesari, S. Anton, E. Marzetti, S. Giovannini, A.Y. Seo, C. Carter, B.P.Yu, C. Leeuwenburgh, Molecular inflammation: underpinnings of aging andage-related diseases, Ageing Res. Rev. 8 (2009) 18–30.

[2] A. Del Prete, P. Allavena, G. Santoro, R. Fumarulo, M.M. Corsi, A. Mantovani,Molecular pathways in cancer-related inflammation, Biochem. Med. (Zagreb)21 (2011) 264–275.

[3] B.I. Jugdutt, Cyclooxygenase inhibition and adverse remodeling during healingafter myocardial infarction, Circulation 115 (2007) 288–291.

[4] A. Lubbad, M.A. Oriowo, I. Khan, Curcumin attenuates inflammation through inhibi-tion of TLR-4 receptor in experimental colitis,Mol. Cell. Biochem. 322 (2009) 127–135.

[5] S.H. Tsai, S.Y. Lin-Shiau, J.K. Lin, Suppression of nitric oxide synthase and thedown-regulation of the activation of NFkappaB in macrophages by resveratrol,Br. J. Pharmacol. 126 (1999) 673–680.

[6] X. Terra, J. Valls, X. Vitrac, J.M. Merrillon, L. Arola, A. Ardevol, C. Blade, J.Fernandez-Larrea, G. Pujadas, J. Salvado, M. Blay, Grape-seed procyanidins actas antiinflammatory agents in endotoxin-stimulated RAW 264.7 macrophagesby inhibiting NFkB signaling pathway, J. Agric. Food Chem. 55 (2007) 4357–4365.

[7] B. Kumar, S.K. Gupta, T.C. Nag, S. Srivastava, R. Saxena, Green tea preventshyperglycemia-induced retinal oxidative stress and inflammation in streptozotocin-induced diabetic rats, Ophthalmic Res. 47 (2012) 103–108.

[8] M.C. Recio, I. Andujar, J.L. Rios, Anti-inflammatory agents from plants: progressand potential, Curr. Med. Chem. 19 (2012) 2088–2103.

[9] Y. Zou, K.J. Jung, J.W. Kim, B.P. Yu, H.Y. Chung, Alteration of soluble adhesionmolecules during aging and their modulation by calorie restriction, FASEB J. 18(2004) 320–322.

[10] H.Y. Chung, B. Sung, K.J. Jung, Y. Zou, B.P. Yu, The molecular inflammatory processin aging, Antioxid. Redox Signal. 8 (2006) 572–581.

[11] P.A. Baeuerle, D. Baltimore, NF-kappa B: ten years after, Cell 87 (1996) 13–20.[12] M.S. Hayden, S. Ghosh, Signaling to NF-kappaB, Genes Dev. 18 (2004) 2195–2224.[13] L.F. Chen, S.A. Williams, Y. Mu, H. Nakano, J.M. Duerr, L. Buckbinder, W.C. Greene,

NF-kappaB RelA phosphorylation regulates RelA acetylation, Mol. Cell. Biol. 25(2005) 7966–7975.

[14] Z.J. Chen, V. Bhoj, R.B. Seth, Ubiquitin, TAK1 and IKK: is there a connection? CellDeath Differ. 13 (2006) 687–692.

[15] D. Krappmann, C. Scheidereit, A pervasive role of ubiquitin conjugation in activa-tion and termination of IkappaB kinase pathways, EMBO Rep. 6 (2005) 321–326.

[16] B. Pamukcu, G.Y. Lip, E. Shantsila, The nuclear factor-kappa B pathway in athero-sclerosis: a potential therapeutic target for atherothrombotic vascular disease,Thromb. Res. 128 (2011) 117–123.

[17] S.F. Yan, R. Ramasamy, Y. Naka, A.M. Schmidt, Glycation, inflammation, and RAGE:a scaffold for the macrovascular complications of diabetes and beyond, Circ. Res.93 (2003) 1159–1169.

[18] R. Foncea, C. Carvajal, C. Almarza, F. Leighton, Endothelial cell oxidative stress andsignal transduction, Biol. Res. 33 (2000) 89–96.

[19] J. DiDonato, F. Mercurio, C. Rosette, J. Wu-Li, H. Suyang, S. Ghosh, M. Karin,Mapping of the inducible IkappaB phosphorylation sites that signal its ubiquitina-tion and degradation, Mol. Cell. Biol. 16 (1996) 1295–1304.

[20] C.C. Chang, J. Chen, M.A. Thomas, D. Cheng, V.A. Del Priore, R.S. Newton, M.E. Pape,T.Y. Chang, Regulation and immunolocalization of acyl-coenzyme A: cholesterolacyltransferase in mammalian cells as studied with specific antibodies, J. Biol.Chem. 270 (1995) 29532–29540.

[21] K. Brown, S. Gerstberger, L. Carlson, G. Franzoso, U. Siebenlist, Control of I kappaB-alpha proteolysis by site-specific, signal-induced phosphorylation, Science 267(1995) 1485–1488.

[22] J.A. DiDonato, F. Mercurio, M. Karin, Phosphorylation of I kappa B alpha precedesbut is not sufficient for its dissociation from NF-kappa B, Mol. Cell. Biol. 15 (1995)1302–1311.

[23] I. Alkalay, A. Yaron, A. Hatzubai, S. Jung, A. Avraham, O. Gerlitz, I. Pashut-Lavon, Y.Ben Neriah, In vivo stimulation of I kappa B phosphorylation is not sufficient toactivate NF-kappa B, Mol. Cell. Biol. 15 (1995) 1294–1301.

[24] M. Karin, The beginning of the end: IkappaB kinase (IKK) and NF-kappaB activa-tion, J. Biol. Chem. 274 (1999) 27339–27342.

[25] O. Ciferri, Spirulina, the edible microorganism, Microbiol. Rev. 47 (1983) 551–578.[26] R. Samuels, U.V. Mani, U.M. Iyer, U.S. Nayak, Hypocholesterolemic effect of spirulina

in patients with hyperlipidemic nephrotic syndrome, J. Med. Food 5 (2002) 91–96.[27] A. Rodriguez-Hernandez, J.L. Ble-Castillo, M.A. Juarez-Oropeza, J.C. Diaz-Zagoya,

Spirulina maxima prevents fatty liver formation in CD-1 male and female micewith experimental diabetes, Life Sci. 69 (2001) 1029–1037.

[28] R.I. Kushak, C. Drapeau, E.M. Van Cott, Favorable effects of blue-green algaeAphanizomenon flos-aquae on rat plasma lipids, J. Am. Nutraceutical Assoc. 2(2000) 59–65.

[29] K. Hori, G. Ishibashi, T. Okita, Hypocholesterolemic effect of blue-green alga,ishikurage (Nostoc commune) in rats fed atherogenic diet, Plant Foods Hum.Nutr. 45 (1994) 63–70.

[30] Y.K. Park, H.E. Rasmussen, S.J. Ehler, K.R. Blobaum, F. Lu, V.L. Schlegel, T.P. Carr, J.Y.Lee, Repression of proinflammatory gene expression by lipid extract of Nostoccommune var sphaeroides Kutzing, a blue-green alga, via inhibition of nuclearfactor-kappa B in RAW 264.7 macrophages, Nutr. Res. 28 (2008) 83–92.

[31] H.E. Rasmussen, K.R. Blobaum, Y.K. Park, S.J. Ehlers, F. Lu, J.Y. Lee, Lipid extractof Nostoc commune var. sphaeroides Kutzing, a blue-green alga, inhibits the activa-tion of sterol regulatory element binding proteins in HepG2 cells, J. Nutr. 138(2008) 476–481.

[32] N.D. Perkins, Post-translational modifications regulating the activity and functionof the nuclear factor kappa B pathway, Oncogene 25 (2006) 6717–6730.

[33] M. Neumann, M. Naumann, Beyond IkappaBs: alternative regulation ofNF-kappaB activity, FASEB J. 21 (2007) 2642–2654.

[34] W.W. Carmichael, The toxins of cyanobacteria, Sci. Am. 270 (1994) 78–86.[35] P.A. Cox, S.A. Banack, S.J. Murch, U. Rasmussen, G. Tien, R.R. Bidigare, J.S. Metcalf,

L.F. Morrison, G.A. Codd, B. Bergman, Diverse taxa of cyanobacteria producebeta-N-methylamino-L-alanine, a neurotoxic amino acid, Proc. Natl. Acad. Sci.U. S. A. 102 (2005) 5074–5078.

[36] H.E. Johnson, S.R. King, S.A. Banack, C. Webster, W.J. Callanaupa, P.A. Cox,Cyanobacteria (Nostoc commune) used as a dietary item in the Peruvian highlandsproduce the neurotoxic amino acid BMAA, J. Ethnopharmacol. 118 (2008) 159–165.

[37] Y. Yang, Y. Park, D.A. Cassada, D.D. Snow, D.G. Rogers, J. Lee, In vitro and in vivosafety assessment of edible blue-green algae, Nostoc commune var. sphaeroidesKutzing and Spirulina platensis, Food Chem. Toxicol. 49 (2011) 1560–1564.

[38] P. Klivenyi, O.A. Andreassen, R.J. Ferrante, A. Dedeoglu, G. Mueller, E. Lancelot, M.Bogdanov, J.K. Andersen, D. Jiang, M.F. Beal, Mice deficient in cellular glutathioneperoxidase show increased vulnerability to malonate, 3-nitropropionic acid, and1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine, J. Neurosci. 20 (2000) 1–7.

[39] C. Selmi, P.S. Leung, L. Fischer, B. German, C.Y. Yang, T.P. Kenny, G.R. Cysewski,M.E. Gershwin, The effects of Spirulina on anemia and immune function in seniorcitizens, Cell. Mol. Immunol. 8 (2011) 248–254.

[40] T.K. Mao, J. Van de Water, M.E. Gershwin, Effects of a Spirulina-based dietary sup-plement on cytokine production from allergic rhinitis patients, J. Med. Food8 (2005) 27–30.

[41] Z.K. Coskun, M. Kerem, N. Gurbuz, S. Omeroglu, H. Pasaoglu, C. Demirtas, N.Lortlar, B. Salman, O.T. Pasaoglu, H.B. Turgut, The study of biochemical and histo-pathological effects of Spirulina in rats with TNBS-induced colitis, Bratisl. Lek.Listy 112 (2011) 235–243.

[42] M. Rasool, E.P. Sabina, B. Lavanya, Anti-inflammatory effect of Spirulina fusiformison adjuvant-induced arthritis in mice, Biol. Pharm. Bull. 29 (2006) 2483–2487.

[43] R. Deng, T.J. Chow, Hypolipidemic, antioxidant, and antiinflammatory activities ofmicroalgae Spirulina, Cardiovasc. Ther. 28 (2010) e33–e45.

[44] C. Romay, R. Gonzalez, N. Ledon, D. Remirez, V. Rimbau, C-phycocyanin: abiliprotein with antioxidant, anti-inflammatory and neuroprotective effects,Curr. Protein Pept. Sci. 4 (2003) 207–216.

[45] S.C. Cherng, S.N. Cheng, A. Tarn, T.C. Chou, Anti-inflammatory activity ofc-phycocyanin in lipopolysaccharide-stimulated RAW 264.7 macrophages, LifeSci. 81 (2007) 1431–1435.

[46] D. Remirez, V. Fernandez, G. Tapia, R. Gonzalez, L.A. Videla, Influence ofC-phycocyanin on hepatocellular parameters related to liver oxidative stressand Kupffer cell functioning, Inflamm. Res. 51 (2002) 351–356.

[47] A.J. Romay, D. Remirez, R. González, N. Ledon, I. García, Antioxidant andanti-inflammatory properties of C-phycocyanin from blue-green algae, Inflamm.Res. 47 (1998) 36–41.

[48] R. Gonzalez, S. Rodriguez, C. Romay, O. Ancheta, A. Gonzalez, J. Armesto, D.Remirez, N. Merino, Anti-inflammatory activity of phycocyanin extract in aceticacid-induced colitis in rats, Pharmacol. Res. Commun. 39 (1999) 55–59.

[49] F.K. Swirski, M. Wildgruber, T. Ueno, J.L. Figueiredo, P. Panizzi, Y. Iwamoto, E.Zhang, J.R. Stone, E. Rodriguez, J.W. Chen, M.J. Pittet, R. Weissleder, M.Nahrendorf, Myeloperoxidase-rich Ly-6C+ myeloid cells infiltrate allograftsand contribute to an imaging signature of organ rejection in mice, J. Clin. Invest.120 (2010) 2627–2634.

[50] F. Leuschner, P.J. Rauch, T. Ueno, R. Gorbatov, B. Marinelli, W.W. Lee, P. Dutta, Y.Wei, C. Robbins, Y. Iwamoto, B. Sena, A. Chudnovskiy, P. Panizzi, E. Keliher, J.M.Higgins, P. Libby, M.A. Moskowitz, M.J. Pittet, F.K. Swirski, R. Weissleder, M.Nahrendorf, Rapid monocyte kinetics in acute myocardial infarction are sustainedby extramedullary monocytopoiesis, J. Exp. Med. 209 (2012) 123–137.

2988 C.S. Ku et al. / Biochimica et Biophysica Acta 1830 (2013) 2981–2988

[51] N.S. Rial, K. Choi, T. Nguyen, B. Snyder, M.J. Slepian, Nuclear factor kappa B(NF-kappaB): a novel cause for diabetes, coronary artery disease and cancerinitiation and promotion? Med. Hypotheses 78 (2012) 29–32.

[52] R. Madonna, R. De Caterina, Relevance of new drug discovery to reduce NF-κBactivation in cardiovascular disease, Vasc. Pharmacol. 57 (2012) 41–47.

[53] M. Karin, Y. Yamamoto, Q.M. Wang, The IKK NF-kappa B system: a treasure trovefor drug development, Nat. Rev. Drug Discov. 3 (2004) 17–26.

[54] C. Santangelo, R. Vari, B. Scazzocchio, R. Di Benedetto, C. Filesi, R. Masella,Polyphenols, intracellular signalling and inflammation, Ann. Ist. Super. Sanita 43(2007) 394–405.

[55] L.F. Chen, Y. Mu, W.C. Greene, Acetylation of RelA at discrete sites regulatesdistinct nuclear functions of NF-kappaB, EMBO J. 21 (2002) 6539–6548.

[56] J.E. Hoberg, F. Yeung, M.W. Mayo, SMRT derepression by the IkappaB kinasealpha: a prerequisite to NF-kappaB transcription and survival, Mol. Cell 16(2004) 245–255.

[57] F. Yeung, J.E. Hoberg, C.S. Ramsey, M.D. Keller, D.R. Jones, R.A. Frye, M.W. Mayo,Modulation of NF-kappaB-dependent transcription and cell survival by theSIRT1 deacetylase, EMBO J. 23 (2004) 2369–2380.

[58] S.T. Bailey, S. Ghosh, ‘PPAR’ting ways with inflammation, Nat. Immunol. 6 (2005)966–967.