Fig 1.
Identification of smaller periostin fragments.
(A) Immunoprecipitated proteins by SS19C blotted by SS19C. The arrows indicate intact periostin and smaller-sized periostin. The smaller product was cut from the gel and subjected to MALDI/TOF-MS. (B) Reactivities of cleaved periostin against anti-periostin mAbs. Periostin purified from human serum was immuno-depleted by SS16A, SS17B, SS18A, SS19A SS19C, SS20A, or SS21A. Western blotting of the immunoprecipitate by the indicated mAbs, followed by blotting with SS19C under reducing conditions is depicted. (C) Antibody binding sites in the original (Assay A) and novel (Assay B) periostin ELISA systems. The arrow indicates the possible cleavage site of the smaller-sized product of periostin. Dark gray parts indicate the composition of the smaller-sized periostin. The capturing mAb (SS18A) and secondary mAbs (SS17B and SS19C) used in Assay A and Assay B are shown. (D) Reactivities of periostin for Assay A and B showing periostin oligomers, monomers and smaller-sized periostin fragments.
Table 1.
Clinical characteristics of BIOAIR study participants.
Table 2.
Correlations between serum and sputum periostin and clinical characteristics in patients with asthma from BIOAIR study.
Fig 2.
Association between sputum IL-13 and sputum periostin.
The relationship between sputum IL-13 and sputum periostin (assay B) was examined in an additional, independent cohort of asthma patients. Panel A shows a scatterplot of sputum IL-13 and periostin levels analysed by linear regression (R2 = 0.4, p = 0.0013). The Spearman r for this analysis was 0.59, p = 0.0031. Panel B shows sputum IL-13 levels in patients with sputum periostin < or ≥ the group median, termed low sputum periostin and high sputum periostin. Individual data points are shown with line at group median.
Fig 3.
Periostin levels in serum and sputum samples subdivided according to sputum inflammatory profile.
Periostin levels were measured in serum and sputum (Assay A and Assay B) in BIOAIR patients and subdivided according to A) sputum eosinophils (eosinophilic = 3% or greater eosinophils, non-eosinophilic below 3% eosinophils) and B) sputum inflammatory profile. Briefly, the four inflammatory profiles were defined as follows: neutrophilic = 61% or greater sputum neutrophils (eosinophils under 3%), eosinophilic = 3% or greater eosinophils (neutrophils under 61%), paucigranulocytic = less than 61% neutrophils and less than 3% eosinophils, and mixed granulocytic = 61% or greater neutrophils and 3% or greater eosinophils. 33% of patients were paucigranulocytic, 21% were neutrophilic, 35.5% were eosinophilic and 10.5% were mixed granulocytic.
Fig 4.
Effect of oral corticosteroid treatment on serum and sputum periostin.
Periostin levels in matched serum and sputum samples were measured before and after 2 weeks of treatment with oral prednisolone. Paired comparisons by Wilcoxon matched-pairs signed rank test were possible for n = 20 patients with mild-to-moderate and severe asthma from the BIOAIR study.
Fig 5.
Evidence of intact and cleaved periostin products in sputum from asthma patients.
Western blot showing intact and/or cleaved periostin products as well as estimated periostin values obtained by ELISA. The results of Assay A or Assay B in sputum samples from 10 asthma patients are depicted. In the Western blotting image, the monomeric periostin represents the reaction with the SS17B antibody and the cleaved product with the SS19C antibody.