Alternatively Spliced Myeloid Differentiation Protein-2 Inhibits TLR4-Mediated Lung Inflammation

An essential component of the outer membrane of Gram-negative bacteria, LPS induces a powerful inflammatory response that can lead to septic shock and death (1). LPS signals through TLR4 by binding to its coreceptor myeloid differentiation protein-2 (MD-2; LY96) and other accessory molecules, including LPS-binding protein and CD14. Among these molecules, LPS-binding protein and CD14 are important for enhancing LPS binding to MD-2 but are not essential for triggering TLR4 signaling (2, 3). MD-2 is an essential component of the signaling receptor complex that recognizes and initiates the innate immune response to bacterial LPS (4). MD-2 is both a membrane bound and a secreted glycoprotein that binds both LPS and the ectodomain of TLR4, forming the TLR4/MD-2/LPS complex (5, 6). Loss-of-function approaches (knockdown and mutation) demonstrated that MD-2 is indispensable for LPS-induced immune cell activation (7–9) in both mouse and human models (10, 11). Upon LPS binding, a receptor multimer composed of two copies of the TLR4/MD-2/LPS complex is formed (10, 12), which triggers a downstream signaling cascade, culminating in the activation of proinflammatory transcription factors such as NF-κB and the IFN regulatory factors. We have shown that MD-2 interacts with Lyn kinase and is tyrosine phosphorylated following LPS-induced activation of the TLR4 signaling pathway. We have demonstrated that this posttranslational modification is required for TLR4/MD-2/LPS signaling (13).

Activation of the innate immune response is a critical step in the response to infection. LPS plays a major role in sepsis and septic shock pathogenesis (14, 15) with 57.2% of patients having either intermediate or high endotoxin levels on the first day of admission (16). As an acutely overactive innate immune responses can contribute to the pathogenesis of many inflammatory diseases (17), it is critical that innate immunity be tightly controlled, activated when necessary, and kept inactive when not. In addition to a complex regulatory mechanism that controls innate immune activation, there are also several negative regulatory pathways of the innate immune system that limit the initial response and the potential damage due to uncontrolled overactive chronic inflammation (18–23). Several negative regulatory pathways have been reported, including proteins that bind an inactivate TLR signaling (18–23) and microRNAs that regulate expression of TLR signaling genes (24). Another method of downregulating TLR4 signaling is to produce an inhibitory isoform by alternatively splicing specific genes encoding essential signaling components such as IL-1R–associated kinase 2, TLR3, MD-2, and MyD88 (25–29). For instance, an alternatively spliced short form of MyD88 acts as a negative regulator of IL-1R/TLR/MyD88-triggered signals, leading to transcriptionally controlled negative regulation of innate immune responses (30, 31).

We recently identified a novel alternatively spliced isoform of human MD-2, MD-2s, that lacks the region encoded by exon 2 of the MD2 (LY96) gene (REF) gene. This human isoform differs from one found in mice (25, 26). We determined that MD-2s is upregulated by IFN-γ, IL-6, and TLR4 signaling, is a negative regulator of LPS-mediated TLR4 signaling, and competitively inhibits binding of full-length MD-2 to TLR4 (32). We hypothesized that MD-2s would have beneficial effects in vivo in mitigating TLR4-mediated lung injury and inflammation models.

To investigate the in vivo effects of MD-2s on TLR4-mediated lung inflammation, we used two mouse models dependent on MD2/TLR4 signaling: LPS-induced acute lung injury (ALI) and house dust mite (HDM)–induced allergic airway inflammation. In both models, we found that overexpression of MD-2s led to marked reduction in markers of tissue injury and inflammation. These data suggest that MD-2s may serve as an effective inhibitor and a potential therapeutic candidate to treat human diseases induced or exacerbated by TLR4 signaling.

C57BL/6 mice were used for all of the experiments in this study. For all ALI experiments, C57BL/6 mice were bred in house (originally from The Jackson Laboratory, Bar Harbor, ME). For all HDM studies, C57BL/6 mice were purchased from The Jackson Laboratory. All animal experiments were performed according to the guidelines and approved protocols of the Institutional Animal Care and Use Committee, Cedars-Sinai Medical Center. Laboratory animals are maintained in accordance with the applicable portions of the Animal Welfare Act and the guidelines prescribed in the Guide for the Care and Use of Laboratory Animals. Cedars-Sinai Medical Center is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care and abides by all applicable laws governing the use of laboratory animals.

The murine macrophage/monocyte cell line RAW264.7 (BH-AC71; American Type Culture Collection) was maintained in DMEM (Sigma-Aldrich) containing 10% FBS (Hyclone), 100 U/ml penicillin, and 100 U/ml streptomycin. Primary normal human small airway epithelial cells (SAEC) as well as all the basal media and growth supplements were obtained from Lonza (Walkersville, MD). Cells were cultivated according to the instructions of the manufacturer on plastic dishes or flasks (BD Biosciences, Heidelberg, Germany). Passage number was kept to less than four passages from original stocks. SAEC were maintained in SAEC basal medium supplemented with 52 μg/ml bovine pituitary extract, 0.5 ng/ml human recombinant epidermal growth factor, 0.5 μg/ml hydrocortisone, 0.5 μg/ml epinephrine, 10 μg/ml transferrin, 5 μg/ml insulin, 0.1 ng/ml retinoic acid, 6.5 ng/ml triiodothyronine, 50 μg/ml Gentamicin/Amphotericin-B (GA-1000), and 50 μg/ml fatty acid–free BSA.

The human MD-2s gene was cloned into the pEntry CMV-IRES shuttle vector using AgeI-NotI sites. After sequence verification, the pEntry-CMV-MD-2s-IRES plasmid was recombined with pDest-Ad plasmid (Life Technologies). The recombinant p–adenovirus construct that expressed MD-2s (Ad-MD-2s) plasmid was selected in DH10B Escherichia coli using ampicillin. For generating recombinant Ad-MD-2s virus, the pAd-MD-2s plasmid DNA was transfected into HEK-293 cells. At 10–14 d posttransfection, viral plaques appeared, and the infected cell lysates were harvested. The Ad-MD-2s virus underwent two additional rounds of amplification. The amplified virus was subjected to two rounds of CsCl gradient ultracentrifugation and dialysis. The viral vector titer was measured in HEK-293 cells by serial dilution. A control Ad-EV virus was generated as described above.

Male, 6–8-wk-old C57BL/6 mice (n = 8 in each group) received a first-generation replication-deficient adenovirus serotype 5 containing human short MD-2s cDNA (Ad-MD-2s) (1 × 10⁹ viral particles in 50 μl PBS) or an empty vector lacking a transgene (Ad-EV) intratracheally (i.t.) 2 d before i.t. instillation of LPS (E. coli serotype O:111; Alpha Chemical and Plastics Co., Hollister, MO) at 0.25 mg/kg. The mice were sacrificed 6 h later to evaluate lung injury. For bronchoalveolar lavage fluid (BALF) transfer experiments, 12 mice in each group treated with 1 × 10⁹ PFU/mice of Ad-EV or Ad-MD-2s for 48 h and BALF collected, pooled, and concentrated by using centrifugal filter devices (Amicon Ultra-15; Merck Millipore) in sterile condition. A total of 100 μl BALF was administrated i.t. into six naive mice in each group, and 1 h later, mice were treated with 0.25 mg/kg LPS i.t. and sacrificed 6 h later.

Female, 6–8-wk-old C57BL/6 mice (n = 10 in each group) received Ad-MD-2s (5 × 10⁸ viral particles in 50 μl PBS), Ad-EV, or PBS. Forty-eight hours later, each of them treated with 50 μg HDM extract of Dermatophagoides pteronyssinus (Greer Laboratories, Lenoir, NC) once by i.t. for sensitization. Then mice were challenged with 50 μg/mice HDM by i.t. route on day 15 and 18 and sacrificed on day 20.

BALF was obtained by cannulating the trachea with a needle and infusing the lungs three times with 0.3 ml PBS. After centrifugation at 2000 × g for 10 min, cell-free BALF was collected and kept at −80°C. Total protein quantification in cell-free BALF was accomplished on aliquots of supernatants from all samples with the BCA Protein Assay Reagent Kit (Pierce Biotechnology, Rockford, IL). The cell pellet was resuspended in cold PBS, and the total cell counts were determined using a hemacytometer. The number of eosinophils was determined by FACS using PE-conjugated anti-Siglec F Ab. Lung tissues were homogenized in PBS, centrifuged, and supernatant of each sample kept at −80°C.

Inflammatory mediators IL-6, TNF-α, keratinocyte chemoattractant (KC), MIP-2, IL-5, and IL-8 were measured in BALF, supernatants of lung homogenate, and culture medium by using an ELISA kit (BD Biosciences, San Jose, CA and R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions.

Myeloperoxidase (MPO) activity was measured in the supernatants of lung tissue homogenate. Briefly, the lung tissues were homogenized in PBS and centrifuged. The MPO activity was assayed by measuring the absorbance spectrophotometrically at 450 nm using 0.167 mg/ml O-dianisidine hydrochloride and 0.0005% hydrogen peroxide. Data were expressed as fold increase comparing to PBS group.

For paraffin sections, tissues were fixed in 10% neutral-buffered formalin and embedded using standard techniques. Sections (5 μm) were cut, deparaffinized, and stained with standard H&E methods to evaluate the tissue histological alterations. For our allergic airway inflammation model, lung sections were also stained with Alcian blue reagent (American MasterTech, Lodi, CA) for detecting airway mucus production. For immunofluorescence, the tissues were fixed in 2.5% paraformaldehyde following dehydration with 30% sucrose overnight at 4°C. Transverse lung sections (30 μm) were obtained with a cryostat and processed for immunofluorescence staining. The tissues were then permeabilized and blocked with protein block (DakoCytomation). The sections were stained with combinations of Abs against E-cadherin (BD Biosciences) and anti-Myc conjugated with Alexa Fluor 488 in blocking solution overnight at 4°C. We used the secondary Alexa Fluor 589–conjugated polyclonal donkey anti-rabbit Ab (1:2500; Jackson ImmunoResearch Laboratories) for 1 h incubation. Images were obtained using a BZ-9000 microscope (Keyence, Itasca, IL).

Total RNA was extracted by the RNeasy extraction kit (Qiagen) and reverse transcribed with the iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA). Real-time PCR was performed using SYBR Green Master Mix (Takara Bio). GAPDH served as a loading control. PCR primers for human IL-8: forward 5′-TCTGGCAACCCTAGTCTGCT-3′ and reverse 5′-GCTTCCACATGTCCTCACAA-3′; and human CCL2: forward 5′-ATCCAGCTCCTTCCAGGATT-3′ and reverse 5′-ACACACCCACCCTCTCTTTG-3′.

Protein concentration was determined by a BCA Protein Assay Kit (Pierce) with BSA used as a standard. Equal amounts of protein were mixed with sample buffer, separated by 10% SDS-PAGE, and transferred onto a polyvinylidene difluoride membrane. Membranes were then incubated in blocking buffer (5% skim milk in 100 mM Tris-HCl [pH 7.5], 150 mM NaCl, and 0.1% Tween-20 [TBST]) for 1 h. Primary Abs were applied in blocking buffer overnight at 4°C. Membranes were washed three times, for 5 min each with TBST, and incubated with a secondary Ab conjugated with HRP (GE Healthcare) in blocking buffer for 1 h at room temperature. After washing the membranes with TBST four times (5 min each), signal was detected by chemiluminescence using the ECL Plus Western blotting Detection kit (GE Healthcare Life Sciences, Piscataway, NJ) and a BioSpectrum UVP Imaging system (Bio-Rad).

To check the overexpression of MD-2s in vitro, the cells were treated with 50:1 multiplicity of infection (MOI) of Ad-EV or Ad-MD-2s for 1 h, the culture medium changed, and then continuously incubated for 48 h. The cells were dislodged by scraping and washed with PBS. After fixation and permeabilization with Cytofix/Cytoperm kit (BD Biosciences), the cells were stained with anti-Myc Alexa Fluor 488 (Santa Cruz Biotechnologies, Santa Cruz, CA) on ice for 30 min. Cells were washed twice with PBS before fluorescence measurement. Eosinophil cell numbers were determined by staining with PE-conjugated anti-Siglec F. Fluorescence was assessed with a CyAn flow cytometer (Beckman Coulter), and data were analyzed using Summit (DakoCytomation, Carpinteria, CA) software.

Results are expressed as mean ± SD. Data were analyzed using Student t test. Significant differences were set at p < 0.05 for all studies. For multiple comparisons, statistical significance was evaluated by one-way ANOVA with the Tukey post hoc test.

We previously reported that MD-2s impedes LPS signaling in vitro by competitively inhibiting MD-2 binding to TLR4 (32). To investigate the potential therapeutic effects of MD-2s in vivo, we generated a Myc-tagged MD-2s adenoviral overexpression vector, Ad-MD-2s. Expression was confirmed by flow cytometry using an anti-Myc Ab compared with cells infected with Ad-EV (Supplemental Fig. 1A). We next infected RAW 264.7 cells to assess if Ad-MD-2s would inhibit LPS-induced signaling and inflammatory cytokine production. In response to 6-h LPS stimulation, the levels of TNF-α and IL-6 were significantly reduced in Ad-MD-2s–infected cells compared with Ad-EV–infected cells (Supplemental Fig. 1B, 1C). These data demonstrate that overexpression of MD-2s blunts LPS-mediated proinflammatory responses in cells.

To elucidate the immunomodulatory effect of MD-2s in LPS-induced lung damage, we examined if i.t. administration of Ad-MD-2s could elevate MD-2s levels in mouse lung. We administered C57BL/6 mice with 1 × 10⁹ PFU Ad-EV or Ad-MD-2s and 48 h later probed for the Myc tag in BALF and lung homogenates. Myc-tagged MD-2s was detected in both the BALF and lung homogenate as two major bands (32, 33) (Fig. 1A), indicating that MD-2s was differentially glycosylated as expected (17). We used immunofluorescence to determine which cells expressed MD-2s in the lungs. Frozen sections were stained with Abs specific for Myc conjugated with Alexa Fluor 488 (green) and the epithelial cell marker E-cadherin, followed by secondary Ab conjugated with Alexa Fluor 569 (red). MD-2s was seen in the bronchial and alveolar epithelial cells of the lung. There was no detectable Myc signal in Ad-EV–infected lungs (Fig. 1B).

To assess the therapeutic effects of MD-2s in vivo, we infected mice with Ad-EV or AD-MD-2s and then 48 h later challenged them with LPS and assessed pulmonary inflammation 6 h later. In the Ad-EV group, LPS treatment resulted, as expected, in thickened alveolar walls, alveolar congestion, and massive cell infiltration. These pathological changes were clearly attenuated in lungs of MD-2s–overexpressing mice (Fig. 2A). Mice that received Ad-MD-2s had significantly reduced levels of protein and cells in the BALF compared with Ad-EV controls (Fig. 2B, 2C). As a proxy for neutrophil infiltration into the lungs, we measured MPO levels in the lung homogenates and found a significant reduction in enzyme levels in mice that received Ad-MD-2s (Fig. 2D). We next measured the proinflammatory cytokine IL-6, as well as the neutrophil chemokines MIP-2 and KC, and found that all three were significantly reduced in both BALF and lung homogenates in mice that received Ad-MD-2s compared with mice that received the empty vector (Fig. 2E–G). Importantly, adenoviral vector alone did not alter the effects of LPS-induced ALI as measured by cellular recruitment and cytokine production (Supplemental Fig. 2). These observations indicate that Ad-MD-2s treatment attenuated lung injury in response to LPS by reducing vascular leak and decreasing neutrophil influx and production of proinflammatory cytokines and chemokines.

To determine if MD-2s functioned specifically by inhibiting TLR4 signaling and not other TLR pathways, we evaluated the effect of MD-2s on the inflammatory cell accumulation and cytokine production in the lungs following an i.t. challenge with Pam3CSK4, a potent activator of TLR2/TLR1 pathway. Pam3CSK4 induced marked infiltration of inflammatory cells into the alveolar space and protein in the BALF (Fig. 3A–C) and increased levels of MPO, IL-6, MIP-2, and KC (Fig. 3D–G). However, none of these endpoints differed between Ad-EV– or Ad-MD-2s–infected mice. These data demonstrate that overexpression of MD-2s did not inhibit TLR2-mediated inflammation and support our conclusion that MD-2s acts specifically to blunt TLR4 signaling.

A soluble form of MD-2 (sMD-2) is markedly elevated in plasma from patients with severe infections and in other fluids from inflamed tissues (34, 35). sMD-2–LPS complex plays a crucial role in the LPS response by activating epithelial and other TLR4-positive MD-2–negative cells in the inflammatory microenvironment. We reported that MD-2s exists also as a secreted form that retains its TLR4-inhibitory activity in cell models (32). To assess if soluble MD-2s produced in vivo from Ad-MD-2s could block TLR4 signaling, we transferred BALF from Ad-MD-2s– or Ad-EV–infected mice into naive mice then challenged with LPS (Fig. 4A). sMD-2s in BALF was detected by immunoblotting using anti-Myc Ab (Fig. 4B). Concentrated BALF was administrated i.t. 1 h before instillation of 0.25 mg/kg LPS, and the mice were sacrificed 6 h later. In the group that received Ad-EV BALF, LPS challenge led to increased total cell number and levels of MPO, IL-6, and KC (Fig. 4C–F). The pretreatment of BALF prepared from Ad-MD-2s–treated mice significantly blunted these proinflammatory responses (Fig. 4C–F). The pretreatment of BALF obtained from naive animals had no effect on LPS-induced inflammation (Supplemental Fig. 3A–D). Although it is possible that an unknown factor in the BALF is contributing the inhibition of LPS signaling, the data indicate that it is likely to be the secreted form of MD-2s.

Recent publications have described the MD-2–related lipid recognition family of proteins of which MD-2 is the prototypical member (36). More recently, the HDM major Ag Der p 2 was found to have many common structural and functional characteristics with MD-2 (37). In mouse models of allergic asthma, the effects of Der p 2 are markedly reduced in Tlr4 knockout mice and can be prevented in wild-type mice by administration of a TLR4 antagonist (38–40). The bronchiolar–alveolar epithelium is a primary target site for inhaled agents that cause lung injury and releases a broad range of mediators that influence other cell populations. New findings showed epithelial cells are in a pivotal position in the development of allergic inflammation though the activation of the TLR4 signaling pathway (38, 39, 41–43). However, in addition to its proinflammatory effects, due to its structural similarities to MD-2, Der p 2 can functionally replace MD-2 and facilitate TLR4 signaling in vitro (42). To determine if MD-2s inhibits HDM-induced activation of epithelial cells, we measured mRNA expression of CCL2 and IL-8, and the release of IL-8 protein from human primary SAEC. The cells were transduced with Ad-EV or Ad-MD-2s at 50:1 MOI for 1 h, washed, and incubated for 48 h. The cells were then stimulated with HDM or PBS. As expected, HDM stimulated the expression of CCL-2 and IL-8 mRNA as well as IL-8 secretion. However, transduction with Ad-MD-2s significantly inhibited the induction of these cytokines (Fig. 5). We also performed similar experiments using the human bronchial epithelial cell line (BEAS-2b) and obtained the same results (data not shown). These data demonstrate that MD-2s inhibits the expression of inflammatory mediators CCL-2 and IL-8 by airway epithelial cells exposed to HDM.

To explore further if MD-2s has a protective effect on allergic airway inflammation in vivo, we used the HDM-induced allergic inflammation model in mice. This model exhibits a number of features of human allergic airway inflammation including cellular infiltration into the lungs, increase in eosinophil numbers, and elevated levels of inflammatory cytokines in BAL fluid. As shown in Fig. 6A, mice were treated with Ad-EV or Ad-MD-2s (n = 10 for each group) at 5 × 10⁸ PFU/mice for 48 h, followed by sensitization with 50 μg HDM given by i.t. route. Mice were then challenged with 50 μg HDM on days 15 and 18 postsensitization and sacrificed on day 20 (Fig. 6A). Sections of the lungs stained with H&E showed marked histological alterations such as peribronchial and perivascular cell infiltration in the airway after HDM challenge in the PBS- or Ad-EV–pretreated groups. However, these changes were reduced in the mice that were treated with AD-MD-2s (Fig. 6B). MD-2s overexpression during sensitization also significantly reduced the influx of eosinophils into the airways, as well as the level of the Th2-associated cytokine IL-5 (p < 0.01) (Fig. 6C, 6D). Finally, MD-2s overexpression significantly reduced goblet cell hyperplasia and airway mucus production in murine lung when compared with mice treated with PBS or Ad-EV prior to HDM sensitization (Fig. 6E, 6F). It should be noted although there were no differences found between PBS groups and empty vector, there was a trend for reduced eosinophils with the vector control. It is possible that adenoviral transduction slightly inhibited Th2 responses by the induction of type I IFN. However, it is clear that MD-2s significantly reduced the overall allergic airway inflammation compared with both controls. These results suggest that in addition to inhibiting LPS-induced ALI, MD-2s can also inhibit HDM-induced allergic lung inflammation in mice.

In this study, we demonstrated that the alternatively spliced form of human MD-2 has clear beneficial effects on TLR4-mediated lung injury using the murine models of LPS-induced ALI and HDM-induced allergic airway inflammation. MD-2s markedly reduced the numbers of inflammatory cells, total proteins, MPO activity, and IL-6, KC, and MIP-2 levels in BALF and lung homogenate after LPS exposure to the lungs. Additionally, the results of the BALF transfer experiment showed that sMD-2s also ameliorated LPS-induced ALI in mice.

TLR4 alone is not sufficient for conferring LPS responsiveness. MD-2 is absolutely required for TLR4 signaling and considered the coreceptor for triggering of TLR4 signaling (44, 45). Indeed, addition of exogenous MD-2 protein could reverse LPS hyporesponsiveness in some experiments (33, 46). In addition, MD-2 knockout mice do not respond to LPS and survive against endotoxic shock (47). Hence, MD-2 is the logical target for pharmacological intervention against triggering of TLR4 signaling (48, 49). As LPS recognition is the first step of the signaling pathway, inhibiting this process could be an efficient way to suppress inflammatory responses before the signal has been transmitted into the downstream pathways. Some effective approaches targeting this step have already been described, but most of them are mainly focused on TLR4 and LPS, whereas few antagonists directed toward MD-2 have been reported (50–52).

Extensive work has suggested that a number of endogenous molecules such as nuclear protein high-mobility group box 1 or oxidized low-density lipoprotein may be potent activators of TLR4 and capable of inducing proinflammatory cytokine production by the monocyte–macrophage system and the activation and maturation of dendritic cells (53, 54). Thus, it is possible that MD-2s might also be able to downregulate this endogenous signaling. Further investigations into the scope of MD-2s inhibition are required.

As we described in our previous publication (32), MD-2s is generated by alternatively splicing out exon 2 of the MD-2 gene, which leads to an in-frame deletion of 30 aa spanning at positions 39–69. Based on the earlier studies (55–57) describing the essential residues in LPS- or TLR4-binding regions of the MD-2 protein, which are intact in our MD-2s. However, recent publications reported the importance of several amino acids in LPS-induced TLR-4/MD-2–mediated cell activation, which are missing in MD-2s (9, 58, 59). Kawasaki et al. (60) reported that the mouse MD-2 mutant G59A can form the cell-surface TLR4/MD-2 complex, but its ability to confer LPS responsiveness was reduced. Expression of the polymorphic variant G56R MD-2 reduces transfer of endotoxin from CD14 to MD-2, thereby reducing TLR4-dependent cell activation (61). These data suggest that amino acids encoded in exon 2 of MD-2 gene play key roles in the triggering and activating LPS signaling.

Particular innocuous environmental proteins act as allergens in susceptible hosts, but clear mechanistic explanations of allergenicity are still lacking. Concentrated in HDM fecal pellets, Der p 2 has the highest rates of skin test positivity compared with other mite allergens in HDM-allergic patients (62). Der p 2 belongs to the MD-2–related lipid-recognition domain family of proteins, binds LPS, and also interacts directly with the TLR4 complex, facilitating LPS signaling (40). Der p 2 drives Th2 inflammation in the airway in a TLR4-dependent fashion by mimicking MD-2 in the absence of MD-2. Thus, Der p 2 has autoadjuvant activity (37, 63). This is of special importance because airway epithelial cells express TLR4 but little or no MD-2 (64). Several studies have demonstrated a role for TLR4 signaling in type 2–mediated inflammation. In response to the major allergen HDM, specifically Der p 2, the mucosal epithelium induces proinflammatory mediators such as IL-8, TSLP, and IL-6 that causes the selective recruitment, retention, and accumulation of cells in the lung. In this study, we showed in vitro that MD-2s clearly suppresses HDM-induced IL-8 release in human primary SAEC and also human epithelial cell line. In our allergic lung inflammation model, MD-2s effectively inhibited the increase of Th2 cytokine, IL-5, and eosinophil count in BALF, and histological studies found that MD-2s substantially inhibited HDM-induced goblet cell hyperplasia in the airway. These results suggest that MD-2s may be useful for the treatment of HDM-related allergic lung inflammation via its inhibition of TLR4 signaling.

MD-2 also plays an important role in certain viral infections. During influenza infection, reactive oxygen species production leads to oxidized lipids that can potently signal though TLR4, requiring MD-2 in the process (65). Additionally, during respiratory syncytial virus infection, the F protein also signals through MD-2/TLR4 (66). In both of these cases, excess inflammation can lead to damaging pneumonia, which, when inhibited by TLR4 agonists, may help resolve the inflammation (65, 66). Finally, the HIV Tat protein has also been shown to bind the TLR4–MD-2 complex, leading to its activation and dysregulation (67). Thus, these viral infections may also present attractive targets for MD-2s–based immune regulation, and future experimental studies investigating the potential beneficial effects of MD-2s in these viral infections are warranted. The complexities involved in averting a prolonged and dysregulated immune response to LPS still need to be investigated further. Sepsis remains a leading cause of death worldwide. Despite years of extensive research, effective drugs that inhibit the proinflammatory effects of LPS and improve outcome when added to conventional sepsis treatments are lacking (68, 69). A naturally occurring alternatively spliced isoform, such as MD-2s, may behave like a decoy coreceptor to form a nonfunctional complex that negatively regulates downstream signaling (32). In this study, we showed that MD-2s attenuates TLR4-mediated inflammatory responses in vivo, suggesting that MD-2s may be useful in the treatment of inflammation associated with TLR4 and LPS. Thus, further investigations into the potential therapeutic aspects of MD-2d will be needed as well as a comprehensive examination of the regulation of the native MD-2s. In the meantime, our data suggest that in addition to several other alternatively spliced inhibitory check points, an MD-2s–mediated negative-feedback loop also ensures that innate immunity is self-limiting; this could be relevant to sepsis, asthma, cancer, and several other diseases with an inflammatory component.

We thank the Viral Vector Core Lab, Cedars Sinai Medical Center. We also thank P. Sun and G. Huang for technical assistance.

The authors have no financial conflicts of interest.