Modulation of nuclear factor b signaling attenuates allergic airway inflammation 4

Figure 4.1 Sequence-dependent inhibition of Rip-2 mRNA and protein expression by S2 in mouse cell lines A Rip-2 mRNA level after transfection in RAW264.7 and NIH/3T3 cells.. 5.1 Results

4 Rip-2 gene silencing attenuates allergic airway inflammation in mice

4.1 Results

4.1.1 In vitro characterization of Rip-2 siRNA

We have screened the gene silencing effects of three Rip-2 siRNA sequences (S1 – S3) targeted at different sites of the coding region of mouse Rip-2 mRNA, in both RAW264.7 and NIH/3T3 cell lines The screening was performed by observing for Rip-2 mRNA reduction, using agarose gel electrophoresis analysis, 24 h after transfection siRNA was transfected into the cells using Lipofectamine 2000 Transfection without siRNA or with control siRNA served as controls S1, S2 and S3 markedly silenced Rip-2 mRNA expression in both cell lines 24 h after transfection by about 70%, as compared to the control siRNA (Figure 4.1A) Results from immnuoblots suggest that the same three sequences produced equivalent knockdown of Rip-2 protein expression by at least 80% in RAW264.7 cells and NIH/3T3 cells at 48 h and 72 h, respectively (Fig 4.1B) The intensity of bands from gel electrophoresis and western blots were quantified using ImageJ as described in Materials and methods Among the 3 sequences, S2 consistently knocked down Rip-2 with the least variability and

was chosen as the lead siRNA for subsequent in vivo experiments

4.1.2 Rip-2 silencing in vivo

In order to verify that 30 µl of solution administered intratracheally to a mouse could reach a substantial portion of the mouse lungs, 30 µl of Evan’s blue was administered to the mice intratracheally Figure 4.2A shows that nearly all parts of the lungs were stained by Evan’s blue Therefore, in subsequent experiments, 30 µl was administered to the mice for each dose Daily intratracheal administration of 5 nmol S2 to nạve BALB/c mice for 3 consecutive days was able to knock-down Rip-2 protein lung level for up to 72 h after the last siRNA dose (Figure 4.2B) In OVA mouse asthma model, we observed for the first time that Rip-2 lung level was markedly elevated (Figure 4.2C) To ensure lung Rip-2 protein knock-down in asthma, we compared the gene silencing

Figure 4.1 Sequence-dependent inhibition of Rip-2 mRNA and protein expression by S2 in mouse cell lines

(A) Rip-2 mRNA level after transfection in RAW264.7 and NIH/3T3 cells RNA evaluation was performed 24 h after transfection β-actin was used as an internal control Agarose gel band intensities were analysed using ImageJ software and normalized to β-actin control (B) Rip-2 protein level after transfection in RAW264.7 and NIH/3T3 cells Protein evaluation was performed 24, 48 and 72 h after transfection β-actin was used as an internal control Immunoblot band intensities were analysed using ImageJ software and normalized to β-actin control * Significant difference from Con, P<0.05

Rip-2

β-actin

RAW264.7

Rip-2 β-actin

SiRNA - - Con S1 S2 S3 SiRNA - - Con S1 S2 S3

*

* * * * *

* * * * * * * * *

* * *

Figure 4.2: Inhibition of Rip-2 protein expression and mRNA by S2 in mouse lungs

(A) Single dose intratracheal administration of 30 µl of 0.1 % Evan’s blue (B) Rip-2 protein level in lung tissues of nạve mice not treated with S2 and Rip-2 protein level in lungs of nạve mice 24 h, 48

h and 72 h after intratracheal administration of 3 doses of S2 for 3 consecutive days (C) Rip-2 protein levels in Saline mice (OVA sensitized and saline challenged) and OVA mice (OVA sensitized and OVA challenged) (D) Rip-2 protein level in lung tissues of OVA sensitized and challenged mice after intratracheal administration of indicated doses of S2 or Con (E) Effect of S2 on Rip-2, TLR-3, and Znhit-3 mRNA in mouse lungs of OVA sensitized and challenged mice, 24 h after intratracheal

Left lobe Superior lobe Middle lobe Inferior lobe Postcava lobe

0 1 2 3

*

*

Rip-2 β-actin

OVA Saline

OVA Con S2

0 0.4 0.8 1.2

independent experiments β-actin was used as loading or internal control for immunoblot and RT-PCR Immunoblot band intensities were analysed using ImageJ and normalized to β-actin control RQ values shown were ratios of various treatments to OVA group * Significant difference from Con, P<0.05

capacities of 3 daily-dose regime of S2 before starting OVA aerosol challenge and 6 daily-dose regime of S2 (3 doses before OVA challenge plus 3 additional doses during OVA aerosol challenges) Only the 6 daily-dose regime of S2 was able to down-regulate Rip-2 lung level in mouse asthma model (Figure 4.2D) Based on Figure 4.2C and D, S2 reduces OVA-stimulated increase in Rip-2 back to saline level Control siRNA did not result in down-regulation of Rip-2 level To confirm S2 specificity for Rip-2 mRNA, a BLAST search was conducted and has revealed 2 closest complementary sequences to S2 siRNA, which are from the Toll-like receptor 3 (TLR3) and Zn finger HIT-domain containing protein 3 (Znhit-3), a thyroid receptor-interacting protein The blast score of Rip-2 was 38.2; while that of TLR3 and Znhit were 28.2 bits and 26.3 bits respectively S2 did not show any silencing effect on TLR3 and Znhit-3 gene expression in lung tissues from mouse asthma model (Figure 4.2E)

4.1.3 Rip-2 siRNA suppresses OVA-induced inflammatory cell recruitment and mucus production

BALF was collected 24 h after the last OVA aerosol challenge, and total differential cell counts were performed OVA inhalation markedly increased total cell, eosinophil and macrophage counts, and slightly yet significantly increased lymphocyte and neutrophil counts, as compared with the saline aerosol control (Figure 4.3A) Intratracheal S2 (1 and 5 nmol) drastically decreased the total cell and eosinophil counts in BALF in a dose-dependent manner as compared with the control siRNA (Figure 4.3A) In addition, S2 at 5 nmol significantly reduced BALF neutrophil count (Figure 4.3A) OVA aerosol challenge induced marked infiltration of inflammatory cells into the peribronchial and perivascular connective tissues as compared with saline challenge S2 (5 nmol) significantly attenuated the eosinophil-rich leukocyte infiltration as compared with control siRNA (Figure 4.3B)

On the other hand, OVA-challenged mice, but not saline-challenged mice, developed marked mucus hypersecretion in the bronchi S2 (5 nmol) substantially reduced mucus hypersecretion as compared to control siRNA (Figure 4.3C)

Figure 4.3 Effects of S2 on OVA-induced inflammatory cell recruitment and mucus hypersecretion (A) Inflammatory cell counts in BALF obtained from sensitized mice 24 h after the last saline aerosol (n = 6 mice) or OVA aerosol (n = 7 mice) challenge S2 dose dependently reduced OVA-induced inflammatory cell counts in BALF from sensitized mice 24 h after the last OVA aerosol challenge (Con, n = 7; 1 nanomole S2, n = 7; 5 nanomoles S2 n = 8 mice) Differential cell counts were performed on a minimum of 500 cells to identify eosinophil (Eos), macrophage (Mac), neutrophil (Neu), and lymphocyte (Lym) Histological examination of lung tissue eosinophilia (B, magnification

× 200) and mucus secretion (C, magnification × 1000) 24 h after the last challenge of saline aerosol,

OVA + Con

peribronchial cell counts were performed blind based on a 5-point scoring system: 0, no cells; 1, a few cells; 2, a ring of cells 1 cell layer deep; 3, a ring of cells 2-4 cells deep; 4, a ring of cells of >4 cells deep To determine the extent of mucous secretion, percentage of airway epithelium covered with mucous was quantified blind using a 5-point grading system: 0, no mucous; 1, <25 % of airway covered with mucous; 2, 25-50% of airway covered with mucous; 3, 50-75% of airway covered with mucous; 4, >75% of airway covered with mucous Scoring of inflammatory cells and mucus secretion was performed in at least 3 different fields for each lung section Mean scores were obtained from 4 animals *Significant difference from Con, P < 0.05

Total IgE OVA –specific IgE

Saline OVA Con 1 5

Figure 4.4 Effects of S2 on OVA-induced BALF cytokine and chemokine levels and serum Ig production

(A) BALF were collected 24 h after the last OVA aerosol challenge Levels of IL-4, IL-5, IL-13, eotaxin, and IL-33 were analyzed using ELISA (n = 6-9 mice per group) (B) Mouse serum was collected 24 h after the last OVA aerosol challenge The levels of total IgE and OVA-specific IgE, were analyzed using ELISA (n = 5 - 9 mice per group) Values shown are the mean ± SEM

*Significant difference from Con, P < 0.05

4.1.4 Rip-2 siRNA reduces OVA-induced BALF cytokines and serum IgE

To determine the levels of cytokines in vivo, BALF samples were collected 24 h after the last OVA

challenge OVA aerosol challenge produced a notable increase in IL-4, IL-5, IL-13, IL-1β, eotaxin and, IL-33 levels in BALF as compared with saline aerosol controls Rip-2 siRNA dose-dependently suppressed IL-4, IL-5, IL-13, IL-1β, eotaxin, and IL-33 as compared with control siRNA (Figure 4.4A) To evaluate whether Rip-2 gene silencing could modify an ongoing OVA-specific Th2

response in vivo, serum levels of total IgE and OVA-specific IgE were determined.Marked elevations

in serum total IgE and OVA-specific IgE levels were observed in OVA-challenged mice as compared with saline-challenged mice Rip-2 gene silencing strongly suppressed OVA-specific IgE levels in a dose-dependent manner, and, to a lesser extent, the serum level of total IgE (Figure 4.4B)

4.1.5 Rip-2 siRNA suppresses OVA-induced inflammatory gene expression in lungs

OVA aerosol challenge markedly up-regulated mRNA levels of lung adhesion molecules (ICAM-1, VCAM-1 and E-selectin), chemokine (RANTES), pro-inflammatory cytokines (TSLP, IL-17, IL-33 and TNF-α), and inflammatory mediators (iNOS and Muc5ac) Intratracheal S2 (5 nmol) demonstrated strong suppression of all these pro-inflammatory mediators in the allergic airways as compared with control siRNA treatment (Figure 4.5)

4.1.6 Rip-2 siRNA reduces OVA-induced AHR

To investigate the effect of Rip-2 gene silencing on AHR, we measured both Rl and Cdyn in mechanically ventilated mice Rl is defined as the pressure driving respiration divided by flow Cdyn refers to the distensibility of the lung and is defined as the change in volume of the lung produced by

a change in pressure across the lung OVA-challenged mice developed AHR to increasing concentrations of methacholine, which is typically reflected by high RI and low Cdyn Intratracheal S2 (5 nmol) significantly reduced RI and restored Cdyn in OVA- challenged mice in response

Figure 4.5 Effects of S2 on inflammatory gene expression in allergic airway inflammation

Lung tissues were collected 24 h after the last OVA aerosol challenge Total mRNA was extracted using TriZol reagent and the quantitative real time PCR was performed All reactions were run in triplicate and three independent experiments for each target The relative quantity of target gene expression was automatically normalized by β-actin as an internal control and values shown were ratios of various treatments to saline group The experiments were repeated for three times (n = 3 mice per group) with similar pattern of results RQ values shown were ratios of various treatments to saline group *Significant difference from Con, P < 0.05

Figure 4.5

ICAM-1 VCAM-1 E-selectin

RANTES IL-17 IL-33

TSLP TNF-α Muc5ac

iNOS

Figure 4.5

to methacholine, suggesting that immune-mediated airway pathology in vivo was modified

(Figure 4.6)

4.17 Rip2 gene silencing disrupts NF-κB signaling pathway

To verify that Rip-2 is a positive regulator of NF-κB pathway, we examined Rip-2 gene silencing on IκBα protein level and NF-κB p65 subunit nuclear translocation and transactivation in lung tissues obtained 24 h after the last OVA or saline aerosol challenge Under resting condition, IκBα binds to NF-κB and prevents NF-κB from entering the nucleus Upon stimulation, IκBα is phosphorylated and degraded Degradation of IκBα allows for NF-κB to translocate into the nucleus In the nucleus, NF-

κB binds to NF-κB response element to mediate transcription of its target genes (Oeckinghaus and Ghosh, 2009) OVA challenge markedly decreased the cytosolic level of IκBα, and promoted p65 nuclear translocation as evidenced by a drastic drop in cytosolic p65 level and a corresponding surge

in nuclear p65 level in lung tissues as compared to those from saline aerosol control (Figure 4.7A) In addition, OVA challenge substantially enhanced nuclear p65 DNA-binding activity (Figure 4.7B) Intratracheal S2 (5 nmol) significantly (P < 0.05) maintained cytosolic IκBα level, retained p65 in the cytosol, and halted p65 nuclear translocation and DNA-binding activity in OVA-challenged lungs as compared to control siRNA (Fig 4.7)

4.2 Discussion

Persistent NF-κB activation has been observed in allergic airway inflammation (Edwards et al., 2009; Gagliardo et al., 2003) Various strategies targeted at the NF-κB signaling pathway such as NF-κB-specific decoy oligonucleotide (Desmet et al., 2004), p65-specific antisense oligonucleotide (Choi et al., 2004) and IKKβ-selective small molecule inhibitor (Broide et al., 2005) have demonstrated beneficial effects in experimental asthma models Rip-2 is not only a transcriptional product of NF-κB

Figure 4.6 Effects of S2 on OVA-induced AHR

Airway responsiveness of mechanically ventilated mice in response to intravenous methacholine was measured 24 h after the last saline aerosol or OVA aerosol with pretreatment of either control or 5 nanomoles S2 AHR is expressed as percentage change from the baseline level of (A) lung resistance (Rl, n = 5 mice per treatment group) and (B) dynamic compliance (Cdyn, n = 5 mice per treatment group) Rl is defined as the pressure driving respiration divided by flow Cdyn refers to the distensibility of the lung and is defined as the change in volume of the lung produced by a change in pressure across the lung *Significant difference from Con, P < 0.05

Figure 4.7 Effects of S2 on NF-κB activity

Immunoblotting of cytosolic extract of lung tissues and p65 NF-κB in nuclear extract of lung tissues isolated from mice 24 h after the last saline aerosol or OVA aerosol challenge pretreated with either Con or 5 nanomoles S2 Cytosolic and nuclear proteins were separated by 10% SDS-PAGE, probed with anti-IκBα, anti-p65, anti-β-actin, or anti-TBP antibody and developed by enhanced chemiluminescence reagent β-actin cytosolic protein and TBP nuclear protein were used as internal control The experiments were repeated for three times (n = 3 mice per group) with similar pattern of results (B) Nuclear p65 DNA-binding activity was determined using a TransAM p65 transcription factor ELISA kit Values shown are the mean ± SEM of three separate experiments *Significant difference from Con, P < 0.05

Nuclear p65 TBP

*

first time that lung Rip-2 protein level was markedly elevated in experimental asthma, which was accompanied by increased NF-κB nuclear translocation and DNA-binding It has been reported that over-expression of Rip-2 in HEK293T cells promotes NF-κB activation (Yin et al., 2010) On the contrary, transfection of plasmid expressing Rip-2 alternative splice variant, without CARD domain, failed to stimulate NF-κB activation (Krieg et al., 2009) Stimulated T cells from Rip-2 knock-out mice exhibited diminished NF-κB activity (Zhen et al., 2007) Likewise, Rip-2-deficient macrophages were defective in NF-κB activation and cytokine production in response to LPS stimulation (Lu et al., 2005) In line with this, suppression of Rip-2 expression resulted in abrogation of NF-κB activation (Kersse et al., 2011), and reduction of TSLP and IL-1β expression in mast cells (Moon et al., 2011; Moon and Kim, 2011), as well as abrogation of caspase-1-mediated NF-κB activation (Humke et al., 2000; Kersse et al., 2011; Lamkanfi et al., 2004; Wang et al., 2005)

We have characterized a Rip-2-specific siRNA (S2) capable of knocking down Rip-2 mRNA and protein levels by at least 70-80 % in two cell lines, without non-specific gene silencing effect on TLR3 and Znhit-3 which encode closely resembling mRNA sequence complementary to S2 siRNA It has been demonstrated that Rip-2 serves as a scaffolding structure directly interacting with IKKγ (NEMO), rendering the IKK complex functional and leading to NF-κB activation (Hasegawa et al., 2008; Inohara et al., 2000) Notably, Rip-2-mediated NF-κB activation does not require its serine/threonine kinase activity (Kobayashi et al., 2002; Ruefli-Brasse et al., 2004; Zhang et al., 2010)

In this study, intratracheal administration of Rip-2 siRNA to experimental asthma mice resulted in a marked lung Rip-2 knockdown, and mitigation of OVA-induced inflammatory cell infiltration, airway mucus hypersecretion, cytokine, chemokine and pro-inflammatory mediator productions, and AHR These observations were also accompanied by an inhibition of p65 nuclear translocation and κB

DNA-binding activity in OVA-challenged lungs in vivo

OVA challenge of mice with disrupted NF-κB function such as conditional knockout of IKKβ or transgenic IκBα mutant expression selectively in airway epithelium recovered significantly less IL-4, IL-5 and IL-13 in the BALF (Broide et al., 2005; Poynter et al., 2004) Th2 cytokines including IL-4, IL-5, IL-13 and IL-33 play an essential role in the pathogenesis of asthma (Besnard et al., 2012; Galli

et al., 2008a) We observed a major drop in BALF levels of IL-4, IL-5, IL-13 and IL-33 in siRNA-treated OVA challenged mice There is increasing evidence supporting the role of IL-1β, expressed in an NF-κB-dependent manner and activated by NLRP3 inflammasome in allergic airways,

Rip-in Rip-inducRip-ing IL-5, IL-13, IL-17, IL-33 and TSLP production Rip-in asthma (Besnard et al., 2011a; Besnard

et al., 2012) Rip-2 gene silencing strongly halted the rise of 1β BALF level as well as 17,

IL-33 and TSLP expression in lung tissues IL-IL-33 is capable of enhancing the IL-5 and IL-13 production through NF-κB activation by Th2 cells (Schmitz et al., 2005) IL-17 plays a critical role in neutrophil and eosinophil recruitment to the lungs in severe asthma (Besnard et al., 2011b) TSLP can promote the effector functions of Th2 cells in asthma and its expression is mediated by NF-κB pathway (Moon and Kim, 2011; Ziegler and Artis, 2010) Notably, while the level of IL-5, IL-13, and IL-33 dropped dramatically to baseline levels following Rip2 siRNA (5 nmole) treatment, the level of IL-1 remains 4-5 fold higher than saline control We speculate that the dramatic drop in IL5, IL-13, and IL-33 could

be mediated by the drop in IL-4 together with IL-1β IL-4 has been shown to mediate the production

of IL-5 and IL-13 (Lloyd and Hessel, 2010) Taken together, Rip-2 gene silencing can suppress a wide spectrum of pro-inflammatory cytokines in experimental mouse asthma and is likely via disruption of NF-κB pathway

Our findings show that Rip-2 gene silencing prevented inflammatory cell infiltration into the airways

as evidenced by a significant drop in total and eosinophil counts in BALF, and in tissue eosinophilia

in lung sections Eosinophil transmigration into the airways is orchestrated by Th2 cytokines IL-5 and

IL-13 and IL-17 have been shown to induce eotaxin production from airway epithelial and smooth muscle cells, respectively (Cosmi et al., 2011; Matsukura et al., 2001; Price et al., 2010; Rahman et al., 2006) TNF-α upregulates the epithelial expression of ICAM-1 and VCAM-1(Babu et al., 2011; Berry

et al., 2007) Besides, combination of IL-17 and TNF-α synergistically induced RANTES, E-selectin and ICAM-1 expressions in human endothelial cells (Hot et al., 2012) IL-33 has been shown to enhance the differentiation and survival of eosinophils (Stolarski et al., 2010) Intratracheal Rip-2 siRNA strongly suppressed eotaxin and RANTES, TNF-α, and ICAM-1, VCAM-1 and E-selectin in OVA-challenged lungs These results are likely due to interruption of NF-κB transcriptional activity

by Rip-2 knock-down, as the genes for eotaxin, TNF-α, RANTES, VCAM-1 and E-selectin contain the κB site for NF-κB within their promoters (Edwards et al., 2009)

There are concrete evidence that goblet cell hyperplasia and MUC5AC production requires IL-4, IL-5 and IL-13 (Thai et al., 2008) MUC5AC gene expression is dependent on the transcriptional activity

of NF-κB (Lai and Rogers, 2010b; Thai et al., 2008) Selective ablation of NF-κB function in airway epithelium has been shown to reduce OVA-induced mucus production in mice (Broide et al., 2005) NF-κB has been reported to regulate an array of genes such as MUC genes MUC5AC and MUC2 promoters contain κB site (Edwards et al., 2009) Recently studies revealed the important role of IL-1β and TNF-α in mucus production in experimental asthma and airway epithelial cells (Thai et al., 2008) As such, the marked decrease in mucus production and MUC5AC expression in the lungs of Rip-2 siRNA-treated mice may be attributable to a significant reduction of those pro-inflammatory cytokines, and a direct disruption of NF-κB pathway in airway epithelium

Elevated serum IgE levels are collectively a hallmark of the Th2 immune response NF-κB plays a crucial role in B cell proliferation and development (Schulze-Luehrmann and Ghosh, 2006) In addition, IL-4 and IL-13 are important in directing B cell growth, differentiation and secretion of IgE

The biological activities of IgE are mediated through its interaction with the FcεRI on mast cells and basophils (Galli and Tsai, 2012) Cross-linking of FcεRI initiates multiple signaling cascades leading

to NF-κB activation and production of proinflammatory lipid mediators, cytokines and chemokines (Metcalfe, 2008; Schulze-Luehrmann and Ghosh, 2006) Our data showed that Rip-2 siRNA given at the later stage of OVA sensitization and during OVA challenge substantially reduced serum levels of total IgE and OVA-specific IgE The observed reduction may be contributed by its interference action

on NF-κB activation in B cells, and on IL-4- and IL-13-mediated class switching to IgE

Increased exhaled NO is associated with increased iNOS expression in the lung epithelium of asthma patients (Lane et al., 2004; Malinovschi et al., 2011) IL-13, IL-1β and TNF-α have been shown to induce iNOS expression in human bronchial epithelial cells leading to elevate NO production (Chibana et al., 2008; Jiang et al., 2007) In addition, iNOS gene expression is regulated by the NF-κB pathway (Edwards et al., 2009; Kumar et al., 2004) Our results show that Rip-2 gene silencing markedly suppressed the OVA-induced iNOS expression in the lungs, which may be due to the direct interruption of NF-κB signaling and the reduced level of IL-13, IL-1β and TNF-α in the allergic airways

Inhibition of NF-κB has been shown to attenuate AHR development (Pasparakis, 2009; Vallabhapurapu and Karin, 2009) Inflammatory mediators released during the allergic inflammation are key contributors to AHR development OVA-induced AHR to increasing concentration of methacholine is significantly suppressed by Rip-2 It is known that IL-5, TNF-α, IL-17, and IL-33 play critical role in AHR by mobilizing and activating esoinophils, leading to the release of pro-inflammatory products such as MBP and cysteinyl-leukotrienes which are closely associated with AHR (Fujisawa et al., 2011; Kim et al., 2010; Long, 2009) In addition, IL-4 and IL-13 have been

(Kim et al., 2010) and that blockade of Rip-2 results in reduction of TSLP expression (Moon et al., 2011; Moon and Kim, 2011) Studies have also revealed that IgE can mediate mast cell activation which may contribute to AHR by producing a wide array of inflammatory mediators and cytokines (Rahman et al., 2006; Ruefli-Brasse et al., 2004) Thus, the observed reduction of AHR by Rip-2 siRNA may be associated with the reduction in Th2 cytokine, TSLP, pro-inflammatory cytokines such

as IL-17 and IL-33 release, tissue eosinophilia and serum IgE

Although results from this study suggest that Rip-2 down-regulation may attenuate OVA-induced Th2

cytokines – IL4, IL5, and IL-13, Chin et al (2002) demonstrated that Rip-2 is required to establish

Th1 response rather than Th2 response They showed that T cells from OVA-sensitised Rip-2 out mice were able secrete IL-4 In addition, serum IgG1 level from OVA-sensitised Rip-2 knock-out mice is similar to that of the wildtype mice (Chin et al., 2002) However, in the same study, the mice were only subjected to OVA sensitisation, the mice were not OVA challenged Therefore, in the study

knock-by Chin et al.(2002), it is expected that Rip-2 level was not markedly increased in the mice On the

other hand, Rip-2 level was markedly increased in our mouse asthma model The differences in the experimental conditions possible account for the discrepancies in our observations

Although results from this study strongly suggest that down-regulation of Rip-2 could afford protective effect against allergic airway inflammation through negative-regulation of NF-κB signaling cascade, we have yet to determine the Rip-2 associated receptors that are the central players of our OVA-asthma model Studies have confirmed that Rip-2 is essential for NOD1/2 mediated NF-κB activation (Inohara et al., 2000) However, currently the association between OVA mouse asthma model and NOD1/2 mediated NF-κB activation is still unclear On the other hand, TNFR is an important receptor in pathogenesis of allergic airway inflammation but Rip-2 was reported to be despensible in TNFR-mediated NF-κB activation (Ruefli-Brasse et al., 2004) Nonetheless, a recent study has shown that Rip-2 interacts with NRLP-10, which is an NF-κB activating receptor (Lautz et

al., 2012) In addition, a recently published study, using OVA-asthma model, show that NRLP-10 knock-out mice fail to develop Th2 allergic airway inflammation (Eisenbarth et al., 2012) In another

study by Meng et al.(2011), it was shown that TLR-3 siRNA suppresses OVA induced-IL-4, IgE and

IgG1 in their rat asthma model Their results suggest the roles of an endogenous TLR-3 ligand or constitutive TLR-3 activity in OVA-induced inflammatory response (Meng et al, 2011) As shown by Kobayashi et al (2002), TLR-3 mediated NF-κB signaling pathway activation requires Rip-2 These taken together could suggest that Rip-2 down-regulation possibly affected NLRP-10 and TLR-3 signaling pathway and thus NF-κB activation in the OVA-mouse model More studies would need to

be conducted to verify this postulation

siRNA was chosen to down-regulate Rip-2 expression because it is deemed as one of the most recent techniques for powerful and effective gene silencing (Hung et al., 2006) Multiple copies of a specific protein can be synthesized from a single mRNA molecule In addition, siRNA specifically suppresses the expression of disease causing genes Such specificity reduces potential side-effect (Darcan-Nicolaisen et al., 2009) Moreover, kinase domain of Rip-2 has been reported to be dispensable for NF-κB activation (Ruefli-Brasse et al., 2004) Therefore, targeting the kinase domain of Rip-2 by small molecule inhibitor may not be an effective approach in suppression of NF-κB activity

siRNA and oligodeoxynucleotide (ASOs) both result in degradation of mRNA by targeting RNA through Watson-Crick base pairing However, results from studies demonstrated that siRNA are more stable in mammalian cells and physiological fluids than ASOs Therefore, SiRNA presents as a better tool to down-regulate gene expression (Bertrand et al., 2002) siRNA drugs in phase II clinical trial are as follows: ALN-RSV01 in respiratory syncytial virus infection; PF-655 for wet age related macular degeneration (AMD) and diabetic macular oedema; QPI-1002 for acute renal failure, and Excellair for asthma (Watts and Corey, 2012)

siRNA in this study was administered intratracheally Intratracheal delivery of siRNA helps to avoid problems such as rapid filtration through kidney and serum degradation by RNase activity (Agu et al., 2001) In addition, intracheal route minimizes the oropharynx deposition and reduces drug loss Therefore, intratracheal delivery ensures high delivery with little drug lost The success of intratracheal siRNA delivery to the lungs was first demonstrated by Perl et al (Perl et al., 2005) Perl

et al reported that by using mice overexpressing GFP, intratracheal administration of GFP resulted in reduced fluorescence intensity in the lungs More importantly, no reduction inflorescence intensity was observed in liver and spleen The absence of such non-specific distribution of siRNA in the liver

is important in our study because Rip-2 siRNA results in significant suppression of NF-κB signaling pathway As mentioned in section 1.2.1 (Introduction of the NF-κB pathway), NF-κB pathway plays important role in liver development and immune response Furthermore, the absence of such non-specific distribution of siRNA throughout the body means less siRNA is needed to achieve its gene silencing effect

After the siRNA reaches the target organ, it has to pass through tissue barriers and reach its target cells (Darcan-Nicolaisen et al., 2009) To facilitate the uptake of siRNA by the cells, complexed siRNA has been used in various studies to down-regulate the target genes However, these modifications may result in off-target effects after non-specific distribution and interferon response (Sledz et al., 2003) Furthermore, viral vectors create the risk of toxicity, rendering them unsuitable

for human use In this study, vector or transfection reagent was not used in vivo siRNA administration

On the other hand, several groups of researchers have also managed to knockdown target gene without the use of vectors or transfection reagents (Darcan-Nicolaisen et al., 2009; Lai et al., 2009)

Although intratracheal route is commonly used for drug delivery in in vivo studies, it has limited

clinical application This route of administration is an extremely uncomfortable technique and is

non-physiological (Driscoll et al., 2000) Nonetheless, it is an excellent delivery route for proof-of-concept

in vivo study Therefore, Rip-2 siRNA was administered to the mice intratracheally

Allergic airway inflammation and AHR development involve multiple inflammatory cells and a wide array of mediators We report here for the first time that Rip-2 siRNA effectively reduces OVA-induced inflammatory cell recruitment into BALF — IL-4, IL-5, IL-13, IL-1β, and eotaxin production, pulmonary eosinophilia, mucus hypersecretion and AHR in a mouse asthma model These findings support a therapeutic value for Rip-2 siRNA in the treatment of asthma

5 Anti-inflammatory effects of

ribosomal protein S3 (RPS3) siRNA in vitro

5.1 Results

5.1.1 RPS-3 gene silencing in vitro

We have screened the gene silencing effect of RPS-3 in two human lung cell lines, namely NCI-H292 (lung mucoepidermoid carcinoma) and BEAS-2B (normal bronchial epithelial cells, virus transformed) The efficiency was determined by observing for RPS-3 mRNA and protein 24, 48, and

72 h after transfection siRNA was transfected into the cells using Lipofectamine 2000 Transfection without siRNA served as vehicular control while transfection with control siRNA served as negative control RPS-3 siRNA showed substantial knockdown efficiency 24 h after transfection, RPS-3 mRNA reduced by 60 – 80 % Also, significant down-regulation of RPS-3 protein was observed in both cell lines 48 h after transfection; protein level of RPS-3 reduced by 60 – 80 % (Figure 5.1 and 5.2) In NCI-H292 and BEAS-2B cells, lowest protein level of RPS-3 was detected 48 h after transfection

5.1.2 Effects of RPS-3 siRNA on TNF-α-induced MUC5AC production

To determine the effect of RPS-3 siRNA on TNF-α-induced MUC5AC production from

mucoepidemoid cell line NCI-H292, the RPS-3 siRNA transfected NCI-H292 cells were stimulated with TNF-α as performed in the earlier experiments In NCI-H292, TNF- α stimulation markedly increased MUC5AC at mRNA and protein level (Figure 5.3) MUC5AC mRNA increased by about two folds following TNF-α stimulation MUC5AC produced in TNF-α stimulated cells was about 1.75 folds that of the unstimulated cells RPS-3 siRNA transfection before TNF-α stimulation significantly reduced MUC5AC at mRNA and protein levels as compared to cells not transfected with RPS-3 siRNA (Figure 5.3)

Figure 5.1 Inhibition of RPS-3 mRNA by RPS-3 siRNA in human cell line NCI-H292 (lung mucoepidermoid carcinoma)

Top panel: RPS-3 mRNA level after transfection in NCI-H292 RNA evaluation was performed 24 h after transfection (n = 3) β-actin was used as loading control Middle panel: RPS-3 protein level after transfection in NCI-H292 Protein evaluation was performed 24, 48 and 72 h after transfection (n = 3 per time point) β-actin was used as loading control Bottom panel: Immunoblot intensities were analysed using ImageJ software and normalized to β-actin control Data shown are representative of three independent experiments Values shown are mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: V or Veh, vehicle control; C or Con, control siRNA

0 0.4 0.8 1.2

*

0 0.2

Figure 5.2 Inhibition of RPS-3 mRNA by RPS-3 siRNA in human cell line BEAS-2B (lung epithelial cell line)

Top panel: RPS-3 mRNA level after transfection in BEAS-2B RNA evaluation was performed 24 h after transfection (n = 3) β-actin was used as loading control Middle panel: RPS-3 protein level after transfection in BEAS-2B Protein evaluation was performed 24, 48 and 72 h after transfection (n = 3 per time point) β-actin was used as loading control Bottom panel: Immunoblot intensities were analysed using ImageJ software and normalized to β-actin control Data shown are representative of three independent experiments Values shown are mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: V or Veh, vehicle control; C or Con, control siRNA

0 0.2 0.4 0.6 0.8 1

M V C RPS-3

0 0.4 0.8 1.2

Figure 5.3 Effects of RPS-3 siRNA on MUC5AC expression in NCI-H292 cells NCI-H292 were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α (50

ng/ml) was added to the cells for 24 h After 24 h, the cells were harvested and lysed Top panel:

MUC5AC mRNA level was evaluated (n = 5) Bottom panel: MUC5AC level in cell lysate was

analysed using ELISA (n = 5) Values shown are mean ± SEM * Significant difference from control

siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

012345

UnstimulatedStimulated

Media

siRNA

Figure 5.3 A

UnstimulatedStimulated

Media

siRNA

*B

5.1.3 Effect of RPS-3 siRNA on TNF-α-induced inflammatory cytokines and mediators

To determine the effect of RPS-3 siRNA on TNF-α-induced inflammatory cytokines and mediators in

lung cell lines, we transfected NCI-H292 and normal human bronchial epithelial cell line (BEAS-2B) with RPS-3 siRNA then stimulated the cells with TNF-α for 24 h (IL-6 and IL-8) or 6 h (TSLP) TNF-α stimulation drastically increased IL-8 and IL-6 expression at mRNA level and IL-8 and IL-6 secreted in both cell lines (Figure 5.4 – 5.7) 6 h of TNF-α stimulation significantly increased TSLP mRNA in both cell lines (Figure 5.8) Transfection of RPS-3 siRNA prior to stimulation significantly suppressed TNF-α-induced up-regulation of IL-8 production in both cell lines (Figure 5.4 – 5.5) However, RPS-3 siRNA transfection failed to suppress TNF-α-induced IL-6 and TSLP expression up-regulation (Figure 5.6 – 5.8)

5.1.4 Effects of RPS-3 siRNA on NF-κB activity

We verified the NF-κB inhibitory effect of RPS-3 siRNA in NCI-H292 and BEAS-2B by examining p65 DNA-binding activity after TNF-α (50 ng/ml) stimulation NCI-H292 and BEAS-2B were transfected with RPS-3 siRNA 48 h after transfection, when the protein levels of RPS-3 are at the lowest in NCI H292 and BEAS-2B, the cells were stimulated with TNF-α for 24 h Following stimulation, the cells were harvested and p65 DNA binding activity and RPS-3 mRNA levels were measured TNF-α stimulation markedly promoted p65 DNA-binding activity RPS-3 siRNA transfection significantly suppressed TNF-α-induced p65 DNA-binding activity (P < 0.05) (Figure 5.9) RPS-3 siRNA transfection also resulted in persistent down-regulation of RPS-3 expression throughout the stimulation (Figure 5.9) Our results suggest that suppressed RPS-3 expression was significantly associated with reduced TNF-α-induced p65 DNA-binding activity

In addition, we also demonstrated the NF-κB inhibitory effect of RPS-3 siRNA using NF-κB gene reporter assay NF-κB/SEAP stable HEK 293 cells were transfected with RPS-3 siRNA and

Figure 5.4 Effects of RPS-3 siRNA on IL-8 expression in NCI-H292 cells NCI-H292 were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α (50

ng/ml) was added to the cells for 24 h After 24 h, the cells were harvested and lysed The cell culture

media were collected to measure IL-8 secreted by the cells Top panel: IL-8 mRNA level was

evaluated Bottom panel: IL-8 secreted was analysed using ELISA (n = 6) Values shown are mean ±

SEM * Significant difference from control siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

050100150200

UnstimulatedStimulated

Media

siRNA

Figure 5.4 A

Unstimulated Stimulated

TNF-α (-) TNF-α (+)

Figure 5.5

Figure 5.5 Effects of RPS-3 siRNA on IL-8 expression in BEAS-2B cells BEAS-2B cells were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α

(50 ng/ml) was added to the cells for 24 h After 24 h, the cells were harvested and lysed The cell

culture media were collected to measure IL-8 secreted by the cells Top panel: IL-8 mRNA level was

evaluated (n = 6) Bottom panel: IL-8 secreted was analysed using ELISA (n = 6) Values shown are

mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

0 10 20 30 40 50

Unstimulated Stimulated

0200400600800

Media Vehicle Ctrl siRNA Rps3 siRNA

UnstimulatedStimulated

TNF-α (-) TNF-α (+)

IL-8 protein

Figure 5.6

Figure 5.6 Effects of RPS-3 siRNA on IL-6 expression in NCI-H292 cells

NCI-H292 were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α (50 ng/ml) was added to the cells for 24 h After 24 h, the cells were harvested and lysed The cell culture media were collected to measure IL-6 secreted by the cells Top panel: IL-6 mRNA level was evaluated (n = 6) Bottom panel: IL-6 secreted was analysed using ELISA (n = 6) Values shown are

mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

0 2 4 6 8

Unstimulated Stimulated

0246810

UnstimulatedStimulated

TNF-α (-) TNF-α (+)

IL-6 protein

Figure 5.7

Figure 5.7 Effects of RPS-3 siRNA on IL-6 expression in BEAS-2B cells

BEAS-2B cells were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α (50 ng/ml) was added to the cells for 24 h After 24 h, the cells were harvested and lysed The cell culture media were collected to measure IL-6 secreted by the cells Top panel: IL-6 mRNA level was evaluated (n = 6) Bottom panel: IL-6 secreted was analysed using ELISA (n = 6) Values shown are

mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

0369

UnstimulatedStimulated

0 2 4 6 8 10

Unstimulated Stimulated

TNF-α (-) TNF-α (+)

Figure 5.8

Figure 5.8 Effects of RPS-3 siRNA on TSLP expression

(A) NCI-H292 and (B) BEAS-2B were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α (50 ng/ml) was added to the cells for 6 h After 6 h, the cells were harvested and lysed mRNA level of TSLP was evaluated in both cell lines (n = 4 per cell lines) Values shown are

mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

0123

UnstimulatedStimulated

UnstimulatedStimulated

TNF-α (-) TNF-α (+)

Figure 5.9

Figure 5.9 Effects of RPS-3 siRNA on p65 DNA-binding activity

(A) NCI-H292 cells and (B) BEAS-2B were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α (50 ng/ml) was added to the cells for 24 h After 24 h, the cells were harvested and lysed (n = 6 per cell line) Nuclear p65 DNA-binding activity of the lysed cells was determined using a TransAM™ p65 transcription factor ELISA kit RPS-3 mRNA levels of the cells harvested were evaluated (n = 6) Values shown are mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

0 0.4 0.8 1.2

0 0.4 0.8 1.2

RPS-3 mRNA p65 DNA binding activity

TNF-α (-) TNF-α (+)

TNF-α (-) TNF-α (+) 0.0

alkaline phosphatase was quantified The amount of alkaline phosphatase secreted correlates positively with the NF-κB activity As compared to non-TNF-α-stimulated cell, TNF-α stimulation drastically increased the amount of alkaline phosphatase secreted RPS-3 siRNA transfected cells have significantly reduced TNF-α-induced alkaline phosphatase production and RPS-3 expression (Figure 5.10) This significant reduction indicates reduced NF-κB activity in RPS-3 siRNA transfected cells This data corroborates well with results from p65 DNA binding assay

Figure 5.10

Figure 5.10 Effects of RPS-3 siRNA on NF-κB activity inNF-κB/SEAPorter™ HEK293 cell line NF-κB/SEAPorter™ HEK293 cell line were transfected with control siRNA or RPS-3 siRNA 48 h after transfection, TNF-α (50 ng/ml) was added to the cells for 24 h After 24 h, the cell culture media were harvested Amount of alkaline phosphatase secreted was measured using SEAP assay kit (n = 6) The cells were also harvest at the end of TNF-α stimulation RPS-3 mRNA levels of the cells harvested were evaluated (n = 6) Values shown are mean ± SEM * Significant difference from control siRNA, P <0.05

Abbreviations: Veh, vehicle control; Con, control siRNA

0100200300

UnstimulatedStimulated

00.40.81.2

UnstimulatedStimulated

SEAP Assay

TNF-α (-) TNF-α (+)

TNF-α (-) TNF-α (+)

5.2 Discussion

NF- κB exerts its fundamental role as a pleiotropic transcription factor by binding to the κB sites on the promoter sequences of its vast array of target genes It has been recently demonstrated in T cells that the binding of NF-κB to the κB site on the promoter a subset of genes — IL-2, IL-8 and IκBα —

is mediated by RPS-3, which has been identified as an integral non-Rel subunit of NF-κB In addition, this non-Rel subunit is also involved in NF-κB mediated expression of immunoglobulin-κ light chain gene and receptor editing in B cells (Cadera et al., 2009; Wan et al., 2007) RPS-3 is a 40S ribosomal protein that is encoded by RPS-3 gene However, RPS-3 is not a coactivator of NF- κB because it does not have transactivating property by itself The NF-κB is a dimer that consists normally consist

of p65 and p50 (Huxford and Ghosh, 2009; Smale, 2012) Based on results from proteomic screening, RPS-3 associates with p65 of NF-κB-IκB complex in the cytoplasm of the cell Upon activation of NF-κB signaling, IKK phosphorylates IκB Phosphorylation of IκB results in its degradation Subsequently, RPS-3- NF-κB complex is released and translocates into the nucleus (Wan and Lenardo, 2009b) In the nucleus, RPS-3- NF-κB complex binds to selected κB site to mediate gene transcription (Figure 1.13) Through its direct interaction with p65, RPS-3 stabilizes the binding between NF-κB and the κB site of a subset of genes— IL-2, IL-8, and IκBα This stabilization is essential for the subset of genes to be expressed (Wan et al., 2007)

Persistent NF-κB activation has been observed in allergic airway inflammation both in animal models

of asthma and in asthmatic patients (Gagliardo et al., 2003; Janssen-Heininger et al., 2009; Pantano et al., 2008b) Therapeutic strategies targeted at the NF-κB signaling pathway, including specific decoy oligonucleotide (Desmet et al., 2004), p65-specific antisense oligonucleotide (Choi et al., 2004) and inhibitory κB kinase β (IKKβ)-selective small molecule inhibitor (Birrell et al., 2005; Newton et al., 2007), have demonstrated beneficial effects in experimental asthma models, making it an attractive therapeutic target for asthma

Given the importance of RPS-3 in NF-κB signaling pathway and the association between NF-κB and asthma, we are interested to find out more about the role that RPS-3 might play in mediating inflammation in asthma Although the inhibitory effect of RPS-3 siRNA on NF-κB has been studied

in T cells, the potential NF-κB inhibitory effect of RPS-3 siRNA has yet to be studied in lung cells (Wan et al., 2007)

Using BEAS-2B (normal bronchial epithelial cells) and NCI-H292 (lung mucoepidermoid cells), we demonstrated that RPS-3 siRNA transfection markedly reduced TNF-α-induced NF-κB signaling TNF-α stimulation strongly increased p65 κB binding activity On the other hand, RPS-3 siRNA transfection prior to stimulation significantly inhibited this binding activity These observations corroborated well with NF-κB - dependent reporter gene assay, which also shows that RPS-3 siRNA suppressed TNF-α induced NF-κB activation In addition, we also show for the first time that this suppression of NF-κB activation by RPS-3 siRNA is accompanied by reduced MUC5AC and IL-8 production in lung cell lines Both MUC5AC and IL-8 are mediators of allergic airway inflammation Our results suggest that RPS-3 siRNA has potential therapeutic effect against airway inflammatory disease

NCI-H292 is a cell line commonly used to study mucus production (Bautista et al., 2009; Ikegami, 2009; Iwashita et al., 2010; Lora et al., 2005) TNF-α activation at 50 ng/ml for 24 h has been shown

to increase MUC5AC production in NCI-H292, in an NF-κB dependent manner (Lora et al., 2005) Our results show that down-regulation of RPS-3 mediated by RPS-3 siRNA inhibited TNF-α-induced MUC5AC up-regulation at mRNA and protein levels Down-regulation of MUC5AC production may offer therapeutic potential because pharmacological inhibition of mucin secretion using a MARCKS peptide attenuated metacholine-induced AHR in mouse asthma model (Agrawal et al., 2007) We also demonstrated that RPS-3 down-regulation significantly suppressed TNF-α-induced up-regulation of

inflammatory cytokine that can be secreted by epithelial cells Studies have shown that inhibition IKKα and IκBα phosphorylation suppressed IL-8 expression and secretion In line these findings, IL-8 contains κB site for NF-κB within its promoter (Simone et al., 2011) IL-8 has been reported to regulate mucin gene (MUC5AC) expression at post-transcriptional level in lung epithelial cells (Bautista et al., 2009) Therefore, inhibition of IL-8 secretion could contribute to the suppressed MUC5AC protein production observed in NCI-H292

Besides analyzing MUC5AC and IL-8 level, we also examined TNF-α-induced IL-6 and TSLP expression Both cytokines have shown to be inducible by TNF-α (Bao et al., 2009; Lee and Ziegler, 2007) It was reported that 6 h of TNF-α stimulation results in highest induction of TSLP at mRNA level TSLP mRNA level declines after 6 h of TNF-α stimulation (Lee and Ziegler, 2007) In order to monitor the effect of RPS-3 siRNA on TNF-α- induced TSLP expression, the cells were only stimulated with TNF-α for 6 h Although RPS-3 siRNA suppressed the expression of TNF-α-induced IL-8 and MUC5AC at mRNA and protein level, our results show that RPS-3 siRNA did not suppress the production of TNF-α induced IL-6 and TSLP Interestingly, IL-8, MUC5AC, IL-6 and TSLP are known to be NF-κB target genes (Bao et al., 2009; Edwards et al., 2009) Other studies showed that RPS-3 is only essential for selective p65-dependent genes (Wan et al., 2007) It is reported that KH domain of RPS-3 preferentially directs NF-κB molecules to κB site with particular sequence specificity Therefore, RPS-3 has been referred to as a “specifier” In line with this specificity, p65 binds to the 3’half of the κB site that is variable Furthermore, it has been proposed that there could be

an array of molecular complexes which contains RPS-3-like “specifier” subunits with different gene transcription activation specificities that all masquerade as a single NF-κB complex in the nucleus Consequently, not all NF-κB target genes are affected by RPS-3 inhibition It is not known what could

be the factor that determines the selectivity Therefore, further studies need to be performed (Wan and Lenardo, 2009b)