S79 αVβ5-MEDIATED TGFβ ACTIVATION BY AIRWAY SMOOTH MUSCLE CELLS IN ASTHMA

¹A. L. Tatler, ¹L. Jolly, ¹J. Porte, ²C. Brightling, ¹A. J. Knox, ¹L. Pang, ¹G. Jenkins. ¹Respiratory Biomedical Research Unit, University of Nottingham, Nottingham, UK, ²University of Leicester, Leicester, UK

Transforming growth factor β (TGFβ) is implicated in airway remodelling in severe asthma. It is synthesised and released from cells as a latent complex then sequestered in the extracellular matrix. Activation of TGFβ is a critical event in TGFβ bioavailability. Lysophosphatidic acid (LPA) is a bioactive phospholipid that induces contraction and/or the formation of stress fibres in many cell types. It is increased in bronchoalveolar lavage fluid following allergen challenge and may contribute to airway remodelling in asthma. We have previously shown that LPA causes stress fibre formation and TGFβ activation in airway epithelial cells. However, whether LPA can cause TGFβ activation in human airway smooth muscle (HASM) cells to promote airway remodelling is unknown.

We stimulated HASM cells with LPA and used a reporter cell co-culture assay to measure TGFβ activation. The mechanism of activation was investigated using an αVβ5 blocking antibody, the inhibitor of cytoskeletal reorganisation, cytochalasin D, and the β₂ agonist formoterol. The nature of the β5 cytoplasmic domain and cytoskeletal interactions was investigated using CS-1 cells, which express no endogenous β5, transfected with DNA constructs for both full-length β5 and polymorphic β5 that has a 9 bp deletion in its cytoplasmic domain. TGFβ activation in asthmatic HASM cells was compared with non-asthmatic HASM cells.

LPA induced a concentration-dependent increase in TGFβ activity, which was blocked by the αVβ5 antibody and cytochalasin D. Moreover, LPA increased the PAI1 (plasminogen activator inhibitor) gene, which was abrogated by TGFβ and αVβ5 neutralising antibodies, cytochalasin D and formoterol. Furthermore, cells transfected with polymorphic β5 could not activate TGFβ. Co-immunoprecipitation showed that polymorphic β5 does not interact with the cytoskeletal protein talin. Finally we showed that asthmatic HASM cells activate more TGFβ in response to LPA than control cells.

In conclusion, we show that LPA induces cytoskeletal reorganisation and αVβ5-mediated TGFβ activation in HASM cells. Interaction between talin and the β5 subunit is essential for αVβ5-mediated TGFβ activation. Furthermore, HASM cells from patients with asthma activate more TGFβ via αVβ5 than control cells. These data provide a potential mechanism through which uncontrolled airway contraction in asthma may promote airway remodelling through αVβ5 integrin-mediated TGFβ activation.

S80 SCREENING FOR NOVEL ANTI-INFLAMMATORY THERAPIES IN A WHOLE ANIMAL IN VIVO MODEL

A. L. Robertson, C. A. Loynes, M. K. Whyte, S. A. Renshaw. University of Sheffield, Sheffield, UK

Rationale Diseases of neutrophilic inflammation are common, affect many organ systems including the lung, and respond poorly to current therapies. There is a major unmet need to identify new ways to treat such diseases. Neutrophils are usually removed by macrophages having undergone apoptosis, but during inflammatory diseases survival signals delay this apoptosis, leading to enhanced inflammation.

Methods We have established a tractable model in transgenic zebrafish expressing green fluorescent protein (GFP) in the neutrophil lineage, in which inflammation resolution can be rapidly quantitated in vivo. Sterile physical injury to the tailfin of anaesthetised larvae leads to a reproducible and quantifiable neutrophilic inflammatory response which spontaneously resolves over time. This permits screening of compound libraries to identify compounds which accelerate the resolution of inflammation.

Results Preliminary experiments have demonstrated the practicality of such screens, and have identified several lead compounds that serve as “proof of principle”, demonstrating the utility of this approach in the identifcation of new immunotherapeutics. From the Spectrum collection (MSdiscovery), 960 compounds were tested, of which 12 were shown to have reproducible effects. These include several known anti-inflammatory agents. Of these, a number have been tested and shown to exhibit dose-dependent effects. Active compounds suppressed neutrophilic inflammation to levels below those seen with our positive control, a potent inducer of neutrophil apoptosis, pyocyanin. For example, in control fish, the number of neutrophils present at the site of injury at 24 h postinjury was 30.24±1.47. In pyocyanin-treated fish it was 19.04±1.93, and for one compound it was 12.33±1.85 (mean±SEM, p<0.05 for both treatments vs control, one-way analysis of variance (ANOVA) with Bonferroni post-test correction, n = 49, 24 and 9, respectively). These compounds are under further investigation for their ability to modulate human neutrophil function and will be assessed for enhancement of resolution of inflammation in mammalian models of neutrophilic pulmonary inflammation.

Conclusions These data show the ability of this model to identify novel therapeutics with dramatic immunomodulatory properties. Some of these compounds may be useful lead compounds for the identification of novel therapeutic entities.

S81 A COMPARISON OF TWO METHODS TO QUANTIFY TENASCIN C EXPRESSION WITHIN THE RETICULAR BASEMENT MEMBRANE OF PAEDIATRIC ENDOBRONCHIAL BIOPSIES

¹R. O’Reilly, ¹N. Ullmann, ²J. Zhu, ¹A. Bush, ²P. K. Jeffery, ¹S. Saglani. ¹Respiratory Paediatrics, NHLI, Imperial College London, UK, ²Lung Pathology, NHLI, Imperial College London, UK

Introduction Tenascin C (TN) is an extracellular matrix glycoprotein with increased expression in the epithelial reticular basement membrane (RBM) in adult asthma. The RBM is thicker in preschool children who wheeze, when compared with age-matched non-wheezing controls.1 We believe that increased TN expression may be a more sensitive marker for severe preschool wheeze (and future asthma) than RBM thickness alone.

Aim To use two methods of quantifying immunohistochemically detected TN expression within the RBM in endobronchial biopsies (EBs) stained with immunoperoxidase.

Methods EBs from preschool children undergoing clinically indicated bronchoscopy were obtained from both wheezers (n = 29, median age 22.5 months) and non-wheezers (n = 10, median age 25 months). Wheezers were divided into confirmed wheeze (CW, n = 15) or reported wheeze (RW, n = 14) based on parent identification using a video questionnaire. Paraffin sections of 5 μm were cut and stained for TN as previously described.2 For TN quantification, method 1 calculated the mean of all TN thickness measurements at 20 μm intervals along the whole RBM, using computer-aided image analysis.3 Method 2 used a semi-quantitative scale to grade the TN-positive proportion of RBM: grade 1 (<1/3), grade 2 (1/3–2/3) and grade 3 (⩾2/3).4 All biopsies were assessed by a single, blinded observer. Non-parametric tests were used to analyse the data.

Results Method 1 showed a significant difference in TN thickness between CW and controls (p = 0.01), but method 2 showed no group differences (p = 0.45). Intraobserver repeatability (coefficient of variation) for method 1 was 0–34%; the intraclass correlation coefficient for repeated measurements for method 2 was 0.93 (p<0.0001). There was no correlation between age and TN expression in the controls.

Conclusion There was increased TN in the RBM of preschool CW compared with controls using method 1. However, in contrast to direct measurements of RBM thickness, there is large variability in TN expression within the RBM which impacted on reliability of both the above methods; method 1 had poor repeatability and method 2 had poor discriminative power. Stereological techniques are being evaluated to overcome these limitations.