En

Da

b

a

ARRAA

KIiMNNNp

I

somisic

irnt

t6

h0

Immunobiology 220 (2015) 83–92

Contents lists available at ScienceDirect

Immunobiology

jo ur nal ho me page: www.elsev ier .com/ locate / imbio

3 ubiquitin ligase NKLAM positively regulates macrophage inducibleitric oxide synthase expression

onald W. Lawrencea, Gail Gullicksona, Jacki Kornblutha,b,∗

Department of Pathology, Saint Louis University School of Medicine, St. Louis, MO 63104, United StatesVeterans Administration Medical Center, St. Louis, MO 63106, United States

r t i c l e i n f o

rticle history:eceived 23 January 2014eceived in revised form 15 August 2014ccepted 15 August 2014vailable online 23 August 2014

eywords:nnate immunityNOS

acrophageKLAMF�Bitric oxide65

a b s t r a c t

Stimulated macrophages generate potent anti-microbial reactive oxygen and nitrogen species withintheir phagosomes. Previous studies have shown that the E3 ubiquitin ligase natural killer lytic-associatedmolecule (NKLAM) is a macrophage phagosomal protein that plays a role in macrophage anti-bacterialactivity. In vivo, NKLAM-knockout (KO) mice produce less nitric oxide (NO) upon exposure to lipopolysac-charide (LPS) than wild type (WT) mice. In vitro, we found that NO production and inducible nitricoxide synthase (iNOS) protein were diminished in LPS-stimulated NKLAM-KO bone marrow-derived andsplenic macrophages. Additionally, LPS-stimulated NKLAM-KO macrophages displayed defects in STAT1tyrosine phosphorylation and production of interferon beta (IFN�). The JAK/STAT pathway is critical forthe production of IFN�, which augments iNOS protein expression in mice. iNOS protein expression isalso regulated by the transcription factor NF�B, thus we investigated whether NKLAM influences NF�Bfunction. LPS-stimulated NKLAM-KO macrophages showed evidence of delayed nuclear translocation ofthe NF�B subunit p65. This was associated with a reduction in p65/DNA colocalization. The defect in

p65 translocation was independent of IKB� degradation. NKLAM-KO macrophages also expressed lessp65 and showed evidence of defective p65 phosphorylation at serine 536. Importantly, LPS-stimulatedNKLAM-KO macrophages have diminished NF�B transcriptional activity as assessed by transfection of aluciferase reporter plasmid. Collectively, our data implicate NKLAM as a novel modulator of macrophageiNOS expression.

Published by Elsevier GmbH.

ntroduction

Macrophages are a first line of defense against pathogens anderve as a critical bridge between the innate and adaptive armsf the immune system. Macrophages employ several bactericidalechanisms to eradicate foreign pathogens. One such mechanism

s the generation of nitric oxide (NO) from l-arginine via the expres-

ion and activation of inducible nitric oxide synthase (iNOS; NOS2).NOS is a dimeric protein that is expressed in several immuneells including macrophages, dendritic cells, and neutrophils (Cruz

Abbreviations: BMDM, bone marrow-derived macrophage; IFN, interferon;NOS, inducible nitric oxide synthase; KO, NKLAM-knockout; LPS, lipopolysaccha-ide; NK, natural killer cell; NKLAM, natural killer lytic-associated molecule; NO,itric oxide; RBR, RING in between RING; STAT, signal transducer and activator ofranscription; TLR, Toll-like receptor; WT, wild type; GFP, green fluorescent protein.∗ Corresponding author at: Department of Pathology, Edward Doisy Research Cen-

er, Saint Louis University School of Medicine, 1100 South Grand Blvd., St. Louis, MO3104, United States. Tel.: +1 314 977 9296; fax: +1 314 977 6910.

E-mail address: kornblut@slu.edu (J. Kornbluth).

ttp://dx.doi.org/10.1016/j.imbio.2014.08.016171-2985/Published by Elsevier GmbH.

et al., 2001; Ratajczak-Wrona et al., 2013). The production of NO byiNOS in response to infection is a critical defense mechanism. Stud-ies have shown that iNOS-knockout mice are unable to mount anefficient immune response (Alayan et al., 2006). Stimulation withmicrobial products such as lipopolysaccharide (LPS), lipoteichoicacid or pro-inflammatory cytokines induces the transcriptionalupregulation of iNOS protein (Zhang et al., 2013). Previous studieshave shown that treatment with exogenous IFN� enhances iNOSprotein expression and NO production (Jacobs and Ignarro, 2001;Yao et al., 2001). Once activated, iNOS is capable to producing largeamounts of NO. Nitric oxide can combine with the simultaneouslyproduced super oxide anion resulting in peroxynitrite, a potent bac-tericidal free radical that induces oxidative damage to proteins,lipids, and nucleic acids (Habib and Ali, 2011; Tecder-Unal et al.,2008). The mouse iNOS promoter contains several transcriptionfactor binding sites such as interferon-stimulated response ele-

ments (ISRE), IFN� activation sites, Oct-1 and NF�B binding sites(Korhonen et al., 2005; Xie et al., 1993).

The regulation of iNOS transcription by transcription factorNF�B has been well studied. There are two NF�B consensus binding

8 unob

stNspsltl2minNst

mRctwlmimwcah2(dWeEd(itailN

M

M

AtNmfD(Lcbtmalw

4 D.W. Lawrence et al. / Imm

ites in the iNOS promoter that are critical for the regulation of iNOSranscription (Kim et al., 1997). In the canonical NF�B pathway,F�B is sequestered in the cytoplasm by its inhibitor IKB�. Upon

timulation with bacterial products such as LPS, IKB� is rapidlyhosphorylated, ubiquitinated and degraded in the 26S protea-ome. The binding of IKB� to NF�B subunit p65 masks a nuclearocalization signal in p65. Once IKB� is degraded, NF�B is freed andranslocates to the nucleus where it transcriptionally regulates aarge number of immunologically important genes (Zerfaoui et al.,008). NF�B p65 is subject to several forms of post-translationalodifications including phosphorylation, acetylation, and ubiquit-

nation (Hoesel and Schmid, 2013). These modifications affect theuclear translocation, DNA-binding and transcriptional activity ofF�B (Perkins, 2006). According to the NF�B barcode hypothe-

is, it is the unique combination of these modifications that directranscription in a gene-specific pattern (Moreno et al., 2010).

Natural killer lytic-associated molecule (NKLAM) is aembrane-bound E3 ubiquitin ligase and a member of the

ING1-in between RING-RING2 (RBR) family of proteins. Threeysteine-rich regions in the N-terminus are critical for the ubiqui-in ligase activity of NKLAM (Fortier and Kornbluth, 2006). NKLAMas originally discovered in our laboratory as a protein that co-

ocalized with granzyme B in natural killer (NK) cytolytic granuleembranes (Kozlowski et al., 1999). In addition to NK cells, NKLAM

s expressed in other mononuclear cells such as monocytes andacrophages (Kozlowski et al., 1999; Portis et al., 2000). NKLAM iseakly expressed under resting conditions but is upregulated by

ytokines (e.g. IFN�) and bacterial products such as LPS (Lawrencend Kornbluth, 2012). NK cells from NKLAM-deficient (KO) miceave diminished anti-tumor cytolytic activity (Hoover et al.,009). In vivo, NKLAM plays a role in controlling tumor metastasisHoover et al., 2012). Recent studies from our laboratory have alsoemonstrated a role for NKLAM in macrophage bacterial killing.e found that macrophages (bone marrow-derived and periton-

al) from NKLAM-KO mice are significantly defective in killingscherichia coli. Despite its localization to phagosomes, NKLAMoes not regulate bacteria uptake or phagosome acidificationLawrence and Kornbluth, 2012). Thus, NKLAM is likely involvedn regulating some other macrophage killing mechanism withinhe phagosome. One such mechanism is iNOS protein expressionnd NO production. Results presented here indicate that NKLAMs a positive regulator of iNOS protein expression, and does so, ateast in part, by modulating the activity of the transcription factorF�B.

aterials and methods

acrophage cultures

All experiments on mice were approved by the Institutionalnimal Care and Use Committees at Saint Louis University and

he VA. Wild type (WT) C57BL/6 and corresponding age-matchedKLAM-KO mice were used in all studies. For isolation of bonearrow, euthanized mice were sprayed with 70% ethanol and the

emurs and tibias were dissected. The bones were flushed withMEM and the collected marrow was resuspended in BM20 media

DMEM supplemented with 20% fetal bovine serum (FBS), 20%929-cell conditioned media, 2 mM l-glutamine, 100 U/ml peni-illin, 100 U/ml streptomycin, and 1 mM sodium pyruvate). Theone marrow cells were cultured for 7 days in non-tissue cul-ure petri dishes with a partial media change on day 3. For splenic

acrophage isolation, spleens were harvested from healthy micend the red blood cells were removed by Lympholyte-M (Cedar-ane) density gradient centrifugation. The mononuclear cell layer

as washed, resuspended in RPMI 1640 supplemented with 10%

iology 220 (2015) 83–92

FBS and plated in tissue culture plastic dishes for 7 days. Cellsobtained using this method are predominantly mature F4/80+,CD11b+ macrophages (Van Ginderachter et al., 2006).

Nitric oxide measurement

Adherent bone marrow-derived macrophages (BMDM) wereincubated with various concentrations of LPS for times indicated.Culture supernatants were assayed for nitrite using the Greiss reac-tion. For the in vivo studies, mice were intraperitoneally injectedwith sterile PBS or 25 �g of LPS and blood was obtained at 1, 2 and6 h post-injection. The blood was centrifuged at 4000 × g for 4 minand the total concentration of nitrite present in the plasma wasdetermined using the Griess reaction following the plasma prepa-ration protocol of Moshage et al. (Moshage et al., 1995). Briefly,plasma samples were diluted and deproteinized with the additionof 1/20th volume of zinc sulfate for a final concentration of 15 g/L.The samples were centrifuged and the resulting supernatants wereassayed for total nitrite using a commercial kit according the man-ufacturer’s protocol (Cayman Chemical Company).

Macrophage transfection and luciferase activity measurement

Adherent BMDM were harvested from dishes with 1.5 mMEDTA, washed once with cold PBS, and then resuspended in DMEMplus 10% FBS. Wild type and NKLAM-KO macrophages (4 × 106)were mixed with 3 �g of the luciferase reporter plasmid pNF�B-luc(Clontech) and suspended in nucleofection solution T. Cells werenucleofected using program T-20 with the nucleofector I (AmaxaBiosystems). The nucleofected macrophages were resuspended inDMEM plus 10% FBS, transferred to 12-well plates and stimulatedwith 100 ng/ml LPS for the times indicated. At the desired time,the cells were collected, lysed in 1X luciferase reporter lysis bufferand snap frozen at −80 ◦C to aid in cell disruption. The total fire-fly luciferase activity was measured using the Promega LuciferaseAssay System (Promega). Protein concentrations were determinedusing bicinchoninic acid (BCA) protein assay reagents (Pierce).Transfection efficiency was assessed by flow cytometry after nucle-ofection of cells with the green fluorescence protein (GFP) reporterplasmid pmaxGFP (Amaxa).

Immunoblotting

Whole cell protein lysates were separated using SDS-PAGEand then transferred to PVDF membrane. Membranes wereblocked with 1% (wt/vol) BSA in Tris-buffered saline plus 0.1%Tween-20 (TBS-T) and then incubated in primary antibody withrocking overnight at 4 ◦C. The antibodies for iNOS (BD Transduc-tion Laboratories), NF�B p65 (Santa Cruz Biotechnologies), IKB�(Cell Signaling), phospho-p65 Ser536 (Cell Signaling), STAT1 (CellSignaling), poly ADP ribose polymerase (PARP; Cell Signaling),and phospho-STAT1 (701) (Cell Signaling) were used at 1:1000.Anti �-actin antibody (Sigma–Aldrich) was used at 1:4000. Afterthree washes in TBS-T, the membranes were probed with HRP-conjugated secondary antibodies and the proteins were visualizedwith Bio-Rad Immun-Star Western C chemiluminescence kit.Images were captured and analyzed using a Bio-Rad ChemidocXRS+ imager (BioRad).

Preparation of cytosolic and nuclear fractions

WT and NKLAM-KO BMDM were treated with 100 ng/ml LPS for

15, 30 or 60 min. Cytoplasmic and nuclear protein fractions wereisolated using NE-PER nuclear and cytoplasmic extraction reagents(Thermo Scientific). Equal protein amounts were immunoblottedfor NF�B p65. The membranes with nuclear and cytoplasmic cell

unobiology 220 (2015) 83–92 85

ed

I

cwmetb(icpwTpLwp

Q

(attRtpTSRhS

S

uss

R

Lo

maca(sA1NolTom

Fig. 1. Nitric oxide production in WT and NKLAM-KO BMDM. (A) WT (black col-umn) and NKLAM-KO (white column) BMDM were treated with 100 ng/ml LPS for18 h. The culture supernatants were assayed for nitrite using the Griess reaction.Data represent the mean ± s.e.m. of three experiments; *p ≤ 0.05, compared to LPS-stimulated WT BMDM. (B) LPS (25 �g) or sterile PBS was injected into the peritonealcavity of WT and NKLAM-KO mice. Blood was obtained at the times indicated and

D.W. Lawrence et al. / Imm

xtracts were reprobed for PARP or beta actin, respectively, toemonstrate purity of the fractions and equivalent protein loading.

mmunofluorescence

Adherent BMDM were plated on acid-cleaned 18 mm glassoverslips and stimulated with 100 ng/ml LPS. The monolayersere fixed and permeabilized by the addition of chilled (−20 ◦C)ethanol:acetone (1:1) for 30 s. After washing in PBS, the cov-

rslips were blocked with 3% BSA in PBS for 60 min at roomemperature. The coverslips were placed in a humidified cham-er and incubated with 1:100 anti-NF�B p65 or anti-phospho-p65Ser536) overnight at 4 ◦C. The coverslips were washed 3 × 10 minn PBS at room temperature and incubated with Alexa Fluor 594hicken anti-rabbit IgG (Life Technologies) for 60 min at room tem-erature. The wash was repeated and the coverslips were incubatedith 1 �g/ml 4′,6-diamidino-2-phenylindole (DAPI) for 10 min.

he coverslips were washed in PBS a final time and mounted inolyvinyl alcohol mounting media. Images were collected with aeica DMI 4000B microscope. All image processing was performedith ImageJ (NIH, Bethesda, MD) or Photoshop. All images wererocessed identically.

uantitative real-time PCR

Unstimulated and LPS-stimulated adherent macrophages4 × 105) in 12-well plates were rinsed quickly with ice-cold PBS,nd then 350 �L of Qiagen RLT buffer was added per well andhe plates were snap frozen at −80 ◦C. RNA was prepared withhe RNeasy Mini Kit (Qiagen) and cDNA was made using TaqManeverse Transcription Gene Expression Assay kit (Applied Biosys-ems). Quantitative PCR analysis of iNOS, 18S rRNA, and IFN� waserformed using qPCR primers purchased from Integrated DNAechnologies. Samples were run on an ABI 7500 Real-Time PCRystem (Applied Biosystems) using iTaq Sybr green supermix (Bio-ad). The change in Ct (�Ct) was calculated with 18S rRNA as aousekeeping gene. iNOS data are presented as 2−(�Ct) (Livak andchmittgen, 2001) and IFN� are presented as 2−(��Ct).

tatistical analysis

Statistical differences were assessed using a two-tailed,npaired Student’s t-test or one-way ANOVA with Microsoft Exceloftware. For both tests, a p value of 0.05 or lower was consideredtatistically significant. Data represent the mean ± s.e.m.

esults

ack of NKLAM expression is associated with diminished nitricxide production in response to LPS in vitro and in vivo

Our previous studies demonstrated that NKLAM-KOacrophages are inherently defective in killing E. coli (Lawrence

nd Kornbluth, 2012). Macrophages employ a number of bacteri-idal mechanisms to combat pathogens; these include phagosomecid hydrolases, phagosomal pH reduction and NO productionFlannagan et al., 2009). We therefore examined the effect of LPStimulation on NO production in WT and NKLAM-KO BMDM.dherent macrophages were treated with 100 ng/ml LPS for8 h and supernatants were assayed for nitrite concentration.KLAM-KO macrophages showed evidence of defective nitricxide production. At 18 h, NKLAM-KO macrophages had accumu-

ated less than half the nitrite levels of WT macrophages (Fig. 1A).o support these initial in vitro studies, we injected 25 �g of LPSr sterile PBS into the peritoneal cavity of WT and NKLAM-KOice. Blood was collected at 1, 2 and 6 h post-injection and total

the concentration of total plasma nitrite was determined using the Griess assay afterconversion of nitrate to nitrite; *p ≤ 0.05, compared to WT mice injected with sterilePBS; #p ≤ 0.05, compared to NKLAM-KO mice injected with sterile PBS.

plasma nitrite levels were measured using the Griess reaction afterconversion of nitrate to nitrite. As shown in Fig. 1B, the plasmanitrite levels in LPS-injected WT mice were significantly increasedat 1, 2 and 6 h post-injection, compared to PBS-injected WT mice.In contrast, the plasma nitrite levels in NKLAM-KO mice were notsignificantly elevated above PBS-injected control mice until 6 hpost-injection, and did not reach the levels observed in controlWT mice. These results suggest that NKLAM plays a positive rolein regulating the production of NO in response to LPS stimulationin vitro and in vivo.

Lack of NKLAM expression is associated with diminished iNOSmRNA and protein expression in response to LPS

Based upon our results that demonstrate defective NO produc-tion in NKLAM-KO macrophages, we next examined the expression

of iNOS protein in LPS-stimulated macrophages. As shown in Fig. 2Aand B, LPS-treated WT BMDM expressed greater than twice theamount of iNOS protein as compared to NKLAM-KO BMDM. Addi-tionally, we found that splenic macrophages respond similarly to

86 D.W. Lawrence et al. / Immunobiology 220 (2015) 83–92

Fig. 2. iNOS expression in WT and NKLAM-KO BMDM stimulated with LPS. (A) WT and NKLAM-KO BMDM were treated with 100 ng/ml LPS for 18 h and whole cell lysates weremade and immunoblotted for iNOS. (B) Densitometric analysis for Fig. 2A; *p ≤ 0.05, n = 3. (C) Splenic macrophages isolated from WT and NKLAM-KO mice were untreated ort iNOSN sion w*

LLKwa

Np

u1Naswusp

SL

JJsami

LPS for the times indicated. Whole cell lysates were preparedand immunoblotted for phospho-STAT1 (701) and total STAT1. Asshown in Fig. 4A and B, LPS stimulation for 2 h induced maxi-mal STAT1 phosphorylation at tyrosine 701. Although the kinetics

Fig. 3. Interferon beta mRNA expression in WT and NKLAM-KO splenic

reated with 100 ng/ml LPS for 18 h and whole cell lysates were immunoblotted forKLAM-KO BMDM were stimulated with 100 ng/ml LPS for 2 h. iNOS mRNA expres

p ≤ 0.03, n = 4.

PS with respect to iNOS expression. iNOS expression was higher inPS-treated (100 ng/ml) WT splenic macrophages than in NKLAM-O splenic macrophages (Fig. 2C). We also found that iNOS mRNAas significantly greater in WT BMDM than in NKLAM-KO BMDM

fter 2 h of LPS stimulation (Fig. 2D).

KLAM-KO macrophages have defective interferon betaroduction in response to LPS stimulation

Murine macrophages secrete IFN� in response to LPS stim-lation and IFN� augments the expression of iNOS (Gao et al.,998). For our next set of experiments, we stimulated WT andKLAM-KO BMDM and spleen macrophages with 100 ng/ml LPSnd quantitated the levels of IFN� mRNA by real time PCR. Ashown in 3 independent experiments we found that IFN� mRNAas 38–67% less in NKLAM-KO BMDM than in WT BMDM stim-lated with LPS. Supportively, IFN� mRNA from LPS-stimulatedplenic macrophages from NKLAM-KO mice was ∼50% less com-ared to WT splenic macrophages (Fig. 3).

TAT1 phosphorylation at tyrosine 701 is defective inPS-stimulated NKLAM-KO macrophages

Interferons regulate the induction of target genes via theAK/STAT pathway. Following ligation of cell surface receptors,AK phosphorylates STAT (signal transducer and activator of tran-

cription) proteins that dimerize before entering the nucleusnd initiate target gene transcription (Rauch et al., 2013). Inurine macrophages, STAT1 is phosphorylated on tyrosine 701

n response to LPS stimulation (Jacobs and Ignarro, 2001; Rhee

protein. For all immunoblots, beta actin was used as a loading control. (D) WT andas determined by quantitative real-time PCR and the data are expressed as 2−�Ct;

et al., 2003). For our next set of experiments we explored the effectof NKLAM expression on STAT1 phosphorylation in response toLPS. WT and NKLAM-KO BMDM were stimulated with 100 ng/ml

macrophages and BMDM following LPS stimulation. WT (black column) and NKLAM-KO (white column) BMDM and splenic macrophages were treated with 100 ng/mlLPS for 2 h. Total RNA was isolated and IFN� mRNA expression was determinedby quantitative real-time PCR. Three individual experiments are presented. Graphsdepict the fold change (2−��Ct) in IFN� mRNA levels at 2 h LPS.

D.W. Lawrence et al. / Immunob

Fig. 4. STAT1 phosphorylation induction by LPS in WT and NKLAM-KO BMDM.WT and NKLAM-KO BMDM were treated with 100 ng/ml LPS for the times indi-cated. (A) Whole cell lysates were prepared and total STAT1 and phospho-STAT1(701) were detected by immunoblot. Immunoblots are representative of 3 indepen-dent experiments. (B) Quantitation by densitometric analysis of the fold increase inpoN

osNSrp

LN

(ttqpdwlmawcmdintIdLwt

hospho-STAT1 (701) relative to untreated cells. Data represent the mean ± s.e.m.f three experiments; *p < 0.05, comparing phosphorylation of STAT1 in WT andKLAM-KO BMDM after 2 h of LPS stimulation.

f STAT1 phosphorylation at tyrosine 701 was similar, there wasignificantly less induction of STAT1 phosphorylation by LPS inKLAM-KO BMDM than in WT macrophages. The levels of totalTAT1 were comparable in NKLAM-KO and WT macrophages. Theseesults suggest that NKLAM plays a role in the signal transductionathway that regulates LPS-induced STAT1 phosphorylation.

PS-induced NF�B p65 nuclear translocation is delayed inKLAM-KO macrophages

NF�B regulates nearly 250 different genes, including iNOSDhruv et al., 2013). In macrophages, stimulation with LPS induceshe translocation of NF�B p65 subunit from the cytoplasm intohe nucleus. NF�B translocation is regulated in part by the ubi-uitination of multiple components of the Toll-like receptor (TLR)athway that are triggered by exposure to LPS. Based on theiminished iNOS expression observed in NKLAM-KO macrophages,e investigated the possible involvement of NF�B p65 in regu-

ating iNOS expression in WT and NKLAM-KO BMDM. Adherentacrophages were stimulated with 100 ng/ml LPS and the nuclear

nd cytosolic fractions were isolated. Equal amounts of proteinere immunoblotted for p65. In unstimulated cells, the levels of

ytosolic p65 were higher in WT macrophages than in NKLAM-KOacrophages (Fig. 5A and B). In WT BMDM, p65 cytosolic levels

ecreased sharply after 15 min of LPS stimulation. This decreasen cytoplasmic p65 was associated with a concomitant rise inuclear p65 expression, consistent with translocation. Followinghis initial decrease, WT cytosolic p65 levels remained constant.n contrast, the cytosolic p65 levels in NKLAM-KO macrophages

id not change as drastically as in WT macrophages followingPS stimulation. The kinetics of p65 expression in the nucleusas much more dynamic (Fig. 5A and C). In WT macrophages,

here was a large (∼3.2-fold) increase in nuclear p65 after 15 min

iology 220 (2015) 83–92 87

of LPS stimulation, which remained elevated out to 60 min. Thelevels of nuclear p65 increased more modestly (∼2-fold) in NKLAM-KO macrophages after 15 min of LPS stimulation and continuedto gradually increase over time. Overall, p65 expression in thenucleus of NKLAM-KO macrophages was significantly less than inWT macrophages (Fig. 5C). Total p65 expression was also less inNKLAM-KO than WT BMDM (Fig. 5D).

To reveal the intracellular expression pattern of p65 during LPSstimulation in WT and NKLAM-KO BMDM, we localized p65 byimmunofluorescence. As translocation of p65 into the nucleus israpid, we stimulated macrophage monolayers with LPS (100 ng/ml)for 15 min before fixation and p65 immunostaining (Fig. 5E). Thenuclei were counterstained with DAPI. Analysis of LPS-stimulatedmacrophages showed that in WT macrophages, p65 was highlyexpressed in the nucleus and perinuclear area. In NKLAM-KOmacrophages, there was less overall p65 and very little within thenucleus (Fig. 5E, lower left panel). Using the NIH ImageJ Colocal-ization Finder plugin we were able to identify specific areas ofp65/DNA colocalization, denoted in white, in WT macrophages butnot in NKLAM-KO macrophages (Fig. 5E, right panels). Therefore,by both immunoblotting and immunostaining, we observed thatNKLAM-KO macrophages have a defect in p65 translocation intothe nucleus after LPS stimulation. Presumably, diminished translo-cation would result in less interaction between p65 and its targetDNA sequences.

LPS-stimulated IKB ̨ degradation is unaffected by the lack ofNKLAM expression

In the canonical NF�B pathway, NF�B is sequestered in thecytoplasm by IKB�. Phosphorylation and ubiquitination of IKB�leads to its proteasomal degradation and release of NF�B whichthen translocates into the nucleus. The prominent role of IKB�ubiquitination in facilitating NF�B nuclear translocation coupledwith our observation that p65 nuclear translocation is attenuatedin NKLAM-KO macrophages led us to examine the degradationand rebound expression of IKB� during LPS stimulation. Adher-ent macrophages were stimulated with 100 ng/ml LPS for 15, 30 or60 min. Cytoplasmic fractions were prepared and immunoblottedfor IKB�. As shown in Fig. 6A, IKB� was nearly completely degradedafter 15 min of LPS stimulation. IKB� degradation was followedby rapid rebound expression. After 30 min of LPS stimulation, theIKB� expression levels increased substantially and were ∼50% ofthe resting IKB� levels. The kinetics and expression levels of IKB�were nearly identical in WT and NKLAM-KO BMDM, suggesting thatNKLAM does not affect IKB� degradation.

Phosphorylation of NF�B p65 at serine 536 is defective inLPS-stimulated NKLAM-KO macrophages

Post-translational modification events such as phosphoryla-tion and acetylation are critical regulators of p65 transcriptionalactivity, and are believed to aid in guiding specific target gene tran-scription (Perkins, 2006). Phosphorylation of NF�B p65 at serine536 has been shown to be associated with nuclear translocation andtranscriptional activity (Hu et al., 2005); thus our next set of experi-ments focused on the kinetics of p65 serine 536 phosphorylation inresponse to LPS. We stimulated WT and NKLAM-KO macrophageswith 100 ng/ml LPS for 15, 30 and 60 min and examined theexpression of serine 536-phosphorylated p65 by both immunofluo-rescence and immunoblotting. In unstimulated WT and NKLAM-KOcells, low levels of serine 536 phosphorylated p65 were seen in

the cytoplasm and displayed a punctate staining pattern, con-sistent with previous reports (Moreno et al., 2010). After 15 minof LPS stimulation, the intensity of serine 536 phosphorylatedp65 staining increased dramatically in WT but not in NKLAM-KO

88 D.W. Lawrence et al. / Immunobiology 220 (2015) 83–92

Fig. 5. NF�B p65 expression and nuclear translocation in WT and NKLAM-KO BMDM. (A–D) WT and NKLAM-KO BMDM were treated with 100 ng/ml LPS for the timesindicated. Isolated cytosolic and nuclear protein fractions were immunoblotted for p65. Immunoblotting for PARP (nuclear lysates) and actin (cytoplasmic lysates) wasdone to show purity of the fractions and equivalent protein loading. (A) Immunoblots represent one of 3 identical experiments. (B–D) Densitometric analysis of p65 incytoplasmic, nuclear and total cellular lysates from WT and NKLAM-KO macrophages treated with 100 ng/ml LPS for the times indicated. Data represent the mean ± s.e.m. oft M-KOL ned wb olor i

mlcoawlsa

hree experiments; p ≤ 0.05, one-way ANOVA, comparing response of WT and NKLAPS for 15 min and then immunostained for p65 (red). The nuclei were counterstaiy the ImageJ Colocalization Finder plugin. (For interpretation of the references to c

acrophages (Fig. 7A). In addition, this increase in phosphory-ated p65 was localized to an area close to the nucleus. In WTells only, the increase in serine 536 phosphorylated p65 was stillbservable 30 min after LPS stimulation. In order to more preciselynd quantitatively examine the kinetics of p65 phosphorylation,

e made lysates from WT and NKLAM-KO macrophages stimu-

ated with 100 ng/ml LPS and performed Western blot analysis. Ashown in Fig. 7B and C, unstimulated macrophages expressed smallmounts of p65 phosphorylated at serine 536. In WT macrophages,

BMDM. (E) Monolayers of WT and NKLAM-KO BMDM were treated with 100 ng/mlith DAPI (blue). Arrows denote areas of p65/DNA colocalization (white) as definedn this figure legend, the reader is referred to the web version of this article.)

treatment with LPS induced an increase in the amount of phospho-rylated p65 at 15 min that returned to baseline levels by 60 min.The kinetics of serine 536 phosphorylation of p65 in NKLAM-KOwas similar to WT cells; however, the amount of phosphorylatedprotein was significantly less in comparison. Beta actin was used as

a loading control and was similar at all time points. The lower levelsof phosphorylated p65 in LPS-stimulated NKLAM-KO macrophagesmay account for the delay in NF�B p65 nuclear translocationobserved in NKLAM-KO macrophages.

D.W. Lawrence et al. / Immunobiology 220 (2015) 83–92 89

Fig. 6. IKB� expression kinetics in WT and NKLAM-KO macrophages stimulatedwith LPS. WT and NKLAM-KO BMDM were incubated with 100 ng/ml LPS for thetimes indicated. (A) Cytoplasmic protein fractions were isolated from each timepie

NB

dtWa1etoap2UtegdwptetsN6tKwrN

Fig. 7. LPS-induced serine 536 phosphorylation of NF�B p65 in WT and NKLAM-KO BMDM. (A) WT and NKLAM-KO BMDM were stimulated with 100 ng/ml LPSfor the times indicated. Phosphorylated p65 was visualized by immunostaining themonolayers with anti-phospho-p65 Ser536 antibody (red). The nuclei were coun-terstained with DAPI (blue). (B) WT and NKLAM-KO macrophages were treated with100 ng/ml LPS for the times indicated and cytosolic fractions were immunoblottedfor p65 phosphorylated at serine 536. Beta actin was used as a loading control. (C)Densitometric analysis of phosphorylated NF�B p65 induction by LPS. Data repre-

oint and immunoblotted for IKB�. Immunoblots represent one of 3 identical exper-ments. (B) Densitometric analysis of IKB�. Data represent the mean ± s.e.m. of threexperiments.

F�B transcriptional activity is lower in NKLAM-KO than in WTMDM

Our observation that NF�B translocation into the nucleus iselayed in NKLAM-KO macrophages prompted us to examine NF�Branscriptional activity in response to LPS stimulation. To this end,

T and NKLAM-KO BMDM were transfected by nucleofection withn NF�B firefly luciferase reporter plasmid and stimulated with00 ng/ml LPS for 6, 12 and 24 h. NF�B proteins bind to the kappanhancer element within the reporter plasmid, inducing luciferaseranscription. The amount of luminescence is therefore a measuref NF�B transcriptional activity. Transfection efficiencies of WTnd NKLAM-KO BMDM were monitored by nucleofection of themaxGFP plasmid. Cells were evaluated for GFP expression after4 h by flow cytometry in a manner similar to Sims et al. (2003).sing this approach allowed us to examine the percentage of cells

hat are transfected and producing GFP protein. The transfectionfficiency of WT and NKLAM-KO macrophages was identical, ran-ing from 22 to 24% GFP positive cells per experiment. Since theifference in nucleofection efficiency between WT and NKLAM-KOas insignificant, we normalized the relative luciferase light unitser mg protein to compare WT and NKLAM-KO NF�B transcrip-ional activities. To ensure that the nucleofection efficiency withinach group (unstimulated and LPS-stimulated) was identical, all ofhe nucleofected cells were aliquoted into 12-well plates from aingle suspension prior to LPS stimulation. As shown in Fig. 8A, theF�B transcriptional activity in LPS-stimulated WT cells peaked at

h (∼7-fold compared to untreated cells), and then decreased withime. NF�B transcriptional activity from LPS-stimulated NKLAM-

O macrophages followed a similar pattern except the valuesere significantly lower than WT macrophages (p ≤ 0.05). These

esults indicate that NKLAM-KO macrophages have attenuatedF�B transcriptional activity in response to LPS stimulation.

sent the mean ± s.e.m. of three experiments; *p ≤ 0.05; one-way ANOVA, comparingresponses of WT and NKLAM-KO BMDM. (For interpretation of the references tocolor in this figure legend, the reader is referred to the web version of this article.)

Discussion

A growing body of literature has implicated members of the RBRfamily of ubiquitin ligases as important regulators of the immuneresponse to foreign pathogens (de Leseleuc et al., 2013; Manzanilloet al., 2013). Recent studies from our laboratory demonstratedthat the RBR family member NKLAM plays an important role inregulating macrophage killing of E. coli (Lawrence and Kornbluth,2012). However, precisely how NKLAM is involved in the murinemacrophage killing response has not been elucidated. This studywas designed to address potential mechanisms by which NKLAM

regulates the macrophage anti-bacterial response.

In this present study we show that LPS-stimulated BMDM fromNKLAM-KO mice produce less NO than BMDM from WT mice(Fig. 1). To support this initial in vitro observation we employed

90 D.W. Lawrence et al. / Immunobiology 220 (2015) 83–92

Fig. 8. NF�B p65 transcriptional activity in WT and NKLAM-KO BMDM and a model of iNOS regulation in BMDM. (A) WT and NKLAM-KO BMDM (4 × 106) were nucleofectedwith 3 �g of the NF�B luciferase reporter plasmid pNF�B-luc and then stimulated with 100 ng/ml LPS for the times indicated. Firefly luciferase activity was measured andexpressed as relative light units per mg protein. Data represent the mean ± s.e.m. of five experiments; p ≤ 0.05; one-way ANOVA. (B) Proposed model of iNOS regulationin LPS-stimulated BMDM. Our data suggest that macrophages from NKLAM-KO mice have lower NF�B p65 expression, less p65 serine 536 phosphorylation (dotted circle)and delayed p65 nuclear translocation (dotted arrow). As there are defined NF�B binding sites (�B) within the promoters of iNOS and IFN�, we propose that diminishedtranslocation is likely to attenuate the transcription of both genes. This observation is supported by our findings of reduced levels of iNOS and IFN� mRNA in LPS-stimulatedN ay aca ated r

aNwdmpKnuie

rtew

KLAM-KO cells compared to WT cells. Furthermore, diminished IFN� production mffect STAT1-mediated iNOS upregulation through binding to the interferon-stimul

n in vivo peritonitis model to examine systemic NO production inKLAM-KO mice. In agreement with previously reported studies,e found that i.p. injection of LPS induced a significant time-ependent increase in plasma nitrite levels (Yeh et al., 2011). In WTice, plasma nitrite levels were significantly elevated at 1, 2 and 6 h

ost-injection, and the highest levels were seen at 6 h. In NKLAM-O mice, there was a lag in NO production. The levels of plasmaitrite were not significantly elevated over PBS-injected controlsntil six hour post-injection and never approached the levels seen

n WT mice. This suggests that NKLAM may play a predominant rolearly in the regulation of NO production after LPS exposure in vivo.

In macrophages, iNOS is responsible for NO production in

esponse to LPS; thus we examined the production of iNOS pro-ein and mRNA in WT and NKLAM-KO macrophages. iNOS proteinxpression was both time and LPS concentration dependent ande found that overall NKLAM-KO BMDM produced less iNOS

count for the observed reduction in STAT1 phosphorylation which may ultimatelyesponse element (ISRE).

protein than WT BMDM. Transcription of iNOS was also lowerin NKLAM-KO than WT BMDM after stimulation with LPS. Weassessed whether differences in iNOS expression could be observedin resident splenic macrophages from WT and NKLAM-KO mice. Inresponse to LPS stimulation, splenic macrophages from NKLAM-KOmice expressed less iNOS than WT macrophages (Fig. 2C).

In macrophages, Toll-like receptor 4 (TLR 4) stimulation by LPSinduces iNOS expression. However, maximal iNOS expression isachieved through synergistic induction by inflammatory mediatorssuch as IFN� (Jacobs and Ignarro, 2001). Interestingly, we found lessIFN� mRNA expression after two h of LPS stimulation in NKLAM-KO spleen macrophages and BMDM than in WT macrophages, thus

implicating NKLAM as a positive regulator, at least in part, of typeI interferon expression. A contrasting role for RBR ubiquitin ligasesas regulators of IFN� was demonstrated by Inn et al. The authorsshowed that RBR family members HOIL-1 and HOIP, as part of the

unob

llsq

FptloLtmiikfiap

ngrNNmtmttfmOcsifeaopralkeWllLiMbccsu(ta

tl2pw

D.W. Lawrence et al. / Imm

inear ubiquitin chain assembly complex (LUBAC), negatively regu-ated type I interferon production (Inn et al., 2011). Clearly, furthertudies are needed to fully elucidate the multiple roles of RBR ubi-uitin ligases in interferon production.

Interferon signaling is mediated by the JAK/STAT pathways.ollowing interferon receptor ligation, JAK phosphorylates STATroteins, which then dimerize and enter the nucleus to initiateranscription. Ligation of TLR 4 also activates STAT phosphory-ation. Rhee et al. demonstrated that STAT1 is phosphorylatedn serine 727 and tyrosine 701 in macrophages stimulated withPS (Rhee et al., 2003). We found less STAT1 phosphorylation atyrosine 701 in LPS-stimulated NKLAM-KO BMDM than in WT

acrophages (Fig. 4). These results suggest NKLAM may play a rolen the tyrosine phosphorylation state of STAT1. Due to the lack ofmmunologically-related substrates for NKLAM and the absence ofnown protein-protein interaction motifs within NKLAM, it is dif-cult to predict what proteins NKLAM may interact with. Studiesre now in progress to determine how NKLAM may influence thishosphorylation event.

The NF�B family of transcription factors is crucial for coordi-ating the expression of a number of immunologically importantenes. Consensus NF�B binding sites are found within the promoteregions of IFN� (Doyle et al., 2002) and iNOS. The central role ofF�B in iNOS and IFN� protein expression prompted us to examineF�B expression and translocation kinetics in WT and NKLAM-KOacrophages stimulated with LPS. We found that translocation of

he NF�B p65 subunit into the nucleus was delayed in NKLAM-KOacrophages. Other studies have shown that delayed NF�B p65

ranslocation has a negative effect on DNA binding and transcrip-ion (Loscher et al., 2005; Park and James, 2005). Indeed, we alsoound less p65/DNA colocalization in LPS-stimulated NKLAM-KO

acrophages than in WT macrophages under similar conditions.ne potential explanation for the observed delay in p65 translo-ation is that IKB� degradation or rebound expression during LPStimulation is influenced by NKLAM. We tested this possibility bymmunoblotting for IKB� in macrophages treated with LPS andound no difference in the kinetics of IKB� degradation or reboundxpression between WT and NKLAM-KO macrophages. Addition-lly, WT and NKLAM-KO macrophages expressed similar amountsf IKB� protein. Thus, it is unlikely that NKLAM affects signalingroteins upstream of IKB� degradation. Interestingly, we observededuced levels of NF�B p65 in unstimulated NKLAM-KO cells. In WTnd NKLAM-KO macrophages stimulated with LPS, p65 expressionevels increased over time. Similar LPS-stimulated p65 expressioninetics has been observed in other experimental systems (Sundayt al., 2007). The p65 expression kinetics is nearly identical inT and NKLAM-KO BMDM; however, total p65 expression was

ower in NKLAM-KO than in WT macrophages (Fig. 5B). Thus, theack of NKLAM expression is associated with both lower basal andPS-stimulated p65 expression in macrophages. Other RBR fam-ly members have been shown to play a role in NF�B activation.

uller-Rischart et al. recently demonstrated that RBR family mem-er parkin, in association with the linear ubiquitin chain assemblyomplex, activates NEMO (IKK�) and thus facilitates the canoni-al NF�B signaling pathway (Muller-Rischart et al., 2013). Previoustudies have demonstrated that NF�B subunit p65 is a target forbiquitin ligases, with polyubiquitination resulting in degradationHou et al., 2012; Wang et al., 2013). Further studies are neededo determine if NF�B p65 is a target for NKLAM ubiquitin ligasectivity.

Phosphorylation is a key regulatory mechanism that governshe movement of NF�B subunits into the nucleus. There are at

east ten phosphorylation sites in the NF�B p65 subunit (Perkins,006). One of the most well studied sites is serine 536. NF�B65 subunit phosphorylation at serine 536 has been associatedith p65 translocation and transcriptional activation (Hu et al.,

iology 220 (2015) 83–92 91

2005; Viatour et al., 2005). By immunofluorescence, we observedless phospho-p65 Ser536 in NKLAM-KO macrophages stimulatedwith LPS than in WT macrophages (Fig. 7). The lack of phos-phorylated p65 may reflect the reduced total p65 expression inNKLAM-KO macrophages both before and after LPS stimulation.We hypothesize that the lower p65 levels may reduce LPS-inducedtranscriptional activation of NF�B in NKLAM-KO macrophages.To test this hypothesis, we nucleofected WT and NKLAM-KOmacrophages with an NF�B luciferase reporter plasmid and inves-tigated the effect of NKLAM expression on NF�B-dependenttranscriptional activity. As shown in Fig. 8A, NF�B transcriptionalactivity in response to LPS is significantly attenuated in NKLAM-KOmacrophages.

In conclusion, we present evidence that the E3 ubiquitin ligaseNKLAM is an important and positive regulator of the macrophageimmune response. Our data identify NKLAM as part of an addi-tional regulatory control point for synthesis of immunologicalproteins such as iNOS. Our data suggest that E3 ubiquitin ligaseNKLAM modulates the transcriptional activity of NF�B. Fig. 8Bdepicts a model of the proposed role of NKLAM in this pathway. InNKLAM-KO macrophages, NF�B p65 expression, serine 536 phos-phorylation, and nuclear translocation are significantly reduced.We propose that these deficiencies significantly contribute to theobserved reduction in iNOS protein via the attenuation of NF�Band IFN�-mediated transcription of the iNOS gene. By controllingNF�B transcriptional activity, NKLAM has the potential to affect thetranscriptional regulation of hundreds of immunologically impor-tant genes. Our data further strengthen the emerging concept thatRBR ubiquitin ligases such as NKLAM play important roles in fun-damental immune functions.

Conflicts of interest

The authors have no conflicts of interest.

Acknowledgments

These studies were supported in part by NIH grant R56AI089758and by a Merit Award (1 IO1BX000705) from the Departmentof Veterans Affairs, Veterans Health Administration, Office ofResearch and Development.

References

Alayan, J., Ivanovski, S., Gemmell, E., Ford, P., Hamlet, S., Farah, C.S., 2006. Deficiencyof iNOS contributes to Porphyromonas gingivalis-induced tissue damage. OralMicrobiol. Immunol. 21, 360.

Cruz, M.T., Duarte, C.B., Goncalo, M., Carvalho, A.P., Lopes, M.C., 2001. LPS inductionof I kappa B-alpha degradation and iNOS expression in a skin dendritic cell lineis prevented by the janus kinase 2 inhibitor, Tyrphostin b42. Nitric Oxide 5, 53.

de Leseleuc, L., Orlova, M., Cobat, A., Girard, M., Huong, N.T., Ba, N.N., Thuc, N.V.,Truman, R., Spencer, J.S., Adams, L., Thai, V.H., Alcais, A., Schurr, E., 2013. PARK2mediates interleukin 6 and monocyte chemoattractant protein 1 production byhuman macrophages. PLoS Negl. Trop. Dis. 7, e2015.

Dhruv, H.D., McDonough Winslow, W.S., Armstrong, B., Tuncali, S., Eschbacher, J.,Kislin, K., Loftus, J.C., Tran, N.L., Berens, M.E., 2013. Reciprocal activation of tran-scription factors underlies the dichotomy between proliferation and invasion ofglioma cells. PLOS ONE 8, e72134.

Doyle, S.E., Vaidya, S.A., O’Connell, R., Dadgostar, H., Dempsey, P.W., Wu, T.-T.,Rao, G., Sun, R., Haberland, M.E., Modlin, R.L., Cheng, G., 2002. IRF3 mediatesa TLR3/TLR4-specific antiviral gene program. Immunity 17, 251.

Flannagan, R.S., Cosio, G., Grinstein, S., 2009. Antimicrobial mechanisms of phago-cytes and bacterial evasion strategies. Nat. Rev. Microbiol. 7, 355.

Fortier, J.M., Kornbluth, J., 2006. NK lytic-associated molecule, involved in NK cyto-toxic function, is an E3 ligase. J. Immunol. 176, 6454.

Gao, J.J., Filla, M.B., Fultz, M.J., Vogel, S.N., Russell, S.W., Murphy, W.J., 1998.

Autocrine/paracrine IFN-alphabeta mediates the lipopolysaccharide-inducedactivation of transcription factor Stat1alpha in mouse macrophages: pivotal roleof Stat1alpha in induction of the inducible nitric oxide synthase gene. J. Immunol.161, 4803.

Habib, S., Ali, A., 2011. Biochemistry of nitric oxide. Indian J. Clin. Biochem. 26, 3.

9 unob

H

H

H

H

H

I

J

K

K

K

L

L

L

M

M

M

M

P

2 D.W. Lawrence et al. / Imm

oesel, B., Schmid, J.A., 2013. The complexity of NF-kappaB signaling in inflamma-tion and cancer. Mol. Cancer 12, 86.

oover, R.G., Gullickson, G., Kornbluth, J., 2009. Impaired NK cytolytic activityand enhanced tumor growth in NK lytic-associated molecule-deficient mice.J. Immunol. 183, 6913.

oover, R.G., Gullickson, G., Kornbluth, J., 2012. Natural killer lytic-associatedmolecule plays a role in controlling tumor dissemination and metastasis. Front.Immunol. 3, 393.

ou, Y., Moreau, F., Chadee, K., 2012. PPARgamma is an E3 ligase that induces thedegradation of NFkappaB/p65. Nat. Commun. 3, 1300.

u, J., Haseebuddin, M., Young, M., Colburn, N.H., 2005. Suppression of p65 phos-phorylation coincides with inhibition of IkappaBalpha polyubiquitination anddegradation. Mol. Carcinog. 44, 274.

nn, K.-S., Gack, M.U., Tokunaga, F., Shi, M., Wong, L.-Y., Iwai, K., Jung, J.U., 2011. Linearubiquitin assembly complex negatively regulates RIG-I- and TRIM25-mediatedtype I interferon induction. Mol. Cell 41, 354.

acobs, A.T., Ignarro, L.J., 2001. Lipopolysaccharide-induced expression of interferon-beta mediates the timing of inducible nitric-oxide synthase induction in RAW264.7 macrophages. J. Biol. Chem. 276, 47950.

im, Y.M., Lee, B.S., Yi, K.Y., Paik, S.G., 1997. Upstream NF-kappaB site is requiredfor the maximal expression of mouse inducible nitric oxide synthase gene ininterferon-gamma plus lipopolysaccharide-induced RAW 264.7 macrophages.Biochem. Biophys. Res. Commun. 236, 655.

orhonen, R., Lahti, A., Kankaanranta, H., Moilanen, E., 2005. Nitric oxide pro-duction and signaling in inflammation. Curr. Drug Targets Inflamm. Allergy 4,471.

ozlowski, M., Schorey, J., Portis, T., Grigoriev, V., Kornbluth, J., 1999. NK lytic-associated molecule: a novel gene selectively expressed in cells with cytolyticfunction. J. Immunol. 163, 1775.

awrence, D.W., Kornbluth, J., 2012. E3 ubiquitin ligase NKLAM is a macrophagephagosome protein and plays a role in bacterial killing. Cell. Immunol. 279, 46.

ivak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data usingreal-time quantitative PCR and the 2(−Delta Delta C(T)) Method. Methods 25,402.

oscher, C.E., Draper, E., Leavy, O., Kelleher, D., Mills, K.H., Roche, H.M., 2005. Con-jugated linoleic acid suppresses NF-kappa B activation and IL-12 production indendritic cells through ERK-mediated IL-10 induction. J. Immunol. 175, 4990.

anzanillo, P.S., Ayres, J.S., Watson, R.O., Collins, A.C., Souza, G., Rae, C.S., Schnei-der, D.S., Nakamura, K., Shiloh, M.U., Cox, J.S., 2013. The ubiquitin ligase parkinmediates resistance to intracellular pathogens. Nature 501, 512.

oreno, R., Sobotzik, J.M., Schultz, C., Schmitz, M.L., 2010. Specification of theNF-kappaB transcriptional response by p65 phosphorylation and TNF-inducednuclear translocation of IKK epsilon. Nucleic Acids Res. 38, 6029.

oshage, H., Kok, B., Huizenga, J.R., Jansen, P.L., 1995. Nitrite and nitrate determi-nations in plasma: a critical evaluation. Clin. Chem. 41, 892.

uller-Rischart, A.K., Pilsl, A., Beaudette, P., Patra, M., Hadian, K., Funke, M., Peis,R., Deinlein, A., Schweimer, C., Kuhn, P.H., Lichtenthaler, S.F., Motori, E., Hrelia,

S., Wurst, W., Trumbach, D., Langer, T., Krappmann, D., Dittmar, G., Tatzelt, J.,Winklhofer, K.F., 2013. The E3 ligase parkin maintains mitochondrial integrityby increasing linear ubiquitination of NEMO. Mol. Cell 49, 908.

ark, S., James, C.D., 2005. ECop (EGFR-coamplified and overexpressed pro-tein), a novel protein, regulates NF-kappaB transcriptional activity and

iology 220 (2015) 83–92

associated apoptotic response in an IkappaBalpha-dependent manner. Onco-gene 24, 2495.

Perkins, N.D., 2006. Post-translational modifications regulating the activity andfunction of the nuclear factor kappa B pathway. Oncogene 25, 6717.

Portis, T., Anderson, J., Esposito, A., Kornbluth, J., 2000. Gene structure of human andmouse NKLAM, a gene associated with cellular cytotoxicity. Immunogenetics51, 546.

Ratajczak-Wrona, W., Jablonska, E., Garley, M., Jablonski, J., Radziwon, P., Iwa-niuk, A., Grubczak, K., 2013. PI3K-Akt/PKB signaling pathway in neutrophils andmononuclear cells exposed to N-nitrosodimethylamine. J. Immunotoxicol. 11,231.

Rauch, I., Muller, M., Decker, T., 2013. The regulation of inflammation by interferonsand their STATs. JAKSTAT 2, 1.

Rhee, S.H., Jones, B.W., Toshchakov, V., Vogel, S.N., Fenton, M.J., 2003. Toll-like recep-tors 2 and 4 activate STAT1 serine phosphorylation by distinct mechanisms inmacrophages. J. Biol. Chem. 278, 22506.

Sims 3rd, R.J., Liss, A.S., Gottlieb, P.D., 2003. Normalization of luciferase reporterassays under conditions that alter internal controls. Biotechniques 34, 938.

Sunday, L., Osuna, C., Krause, D.N., Duckles, S.P., 2007. Age alters cerebrovascularinflammation and effects of estrogen. Am. J. Physiol. Heart Circ. Physiol. 292,H2333.

Tecder-Unal, M., Can, F., Demirbilek, M., Karabay, G., Tufan, H., Arslan, H., 2008. Thebactericidal and morphological effects of peroxynitrite on Helicobacter pylori.Helicobacter 13, 42.

Van Ginderachter, J.A., Meerschaut, S., Liu, Y., Brys, L., De Groeve, K., HassanzadehGhassabeh, G., Raes, G., De Baetselier, P., 2006. Peroxisome proliferator-activatedreceptor gamma (PPARgamma) ligands reverse CTL suppression by alternativelyactivated (M2) macrophages in cancer. Blood 108, 525.

Viatour, P., Merville, M.P., Bours, V., Chariot, A., 2005. Phosphorylation of NF-kappaBand IkappaB proteins: implications in cancer and inflammation. Trends Biochem.Sci. 30, 43.

Wang, Y., Ren, F., Wang, Y., Feng, Y., Wang, D., Jia, B., Qiu, Y., Wang, S., Yu, J., Sung, J.J.,Xu, J., Zeps, N., Chang, Z., 2013. CHIP/Stub1 functions as a tumor suppressor andrepresses NF-kappaB-mediated signaling in colorectal cancer. Carcinogenesis35, 983.

Xie, Q.W., Whisnant, R., Nathan, C., 1993. Promoter of the mouse gene encodingcalcium-independent nitric oxide synthase confers inducibility by interferongamma and bacterial lipopolysaccharide. J. Exp. Med. 177, 1779.

Yao, S.Y., Ljunggren-Rose, A., Stratton, C.W., Mitchell, W.M., Sriram, S., 2001. Regu-lation by IFN-beta of inducible nitric oxide synthase and interleukin-12/p40 inmurine macrophages cultured in the presence of Chlamydia pneumoniae anti-gens. J. Interferon Cytokine Res. 21, 137.

Yeh, C.-H., Chou, W., Chu, C.-C., So, E.C., Chang, H.-C., Wang, J.-J., Hsing, C.-H., 2011.Anticancer agent 2-methoxyestradiol improves survival in septic mice by reduc-ing the production of cytokines and nitric oxide. Shock 36, 510.

Zerfaoui, M., Suzuki, Y., Naura, A.S., Hans, C.P., Nichols, C., Boulares, A.H., 2008.Nuclear translocation of p65 NF-kappaB is sufficient for VCAM-1, but not ICAM-

1, expression in TNF-stimulated smooth muscle cells: differential requirementfor PARP-1 expression and interaction. Cell. Signal. 20, 186.

Zhang, L., Liu, Q., Yuan, X., Wang, T., Luo, S., Lei, H., Xia, Y., 2013. Requirement of heatshock protein 70 for inducible nitric oxide synthase induction. Cell. Signal. 25,1310.