Investigating the role of pentraxin 3 as a biomarker for bacterial infection in subjects with COPD

Background Pentraxin 3 (PTX3) is an acute phase protein, involved in antibacterial resistance. Recent studies have shown PTX3 levels to be elevated in the presence of a bacterial infection and in a murine sepsis model. Objective We aim to investigate if sputum PTX3 can be used as a biomarker for bacterial infection in subjects with COPD. Materials and methods Sputum samples from 142 COPD patients (102 men) with a mean (range) age of 69 years (45–85) and mean (SD) post-bronchodilator percentage predicted forced expiratory volume in 1 second (FEV1) of 50% (19) were analyzed for PTX3, using a commercial assay at stable state and during an exacerbation. Association with bacteria, from culture, quantitative real-time polymerase chain reaction (qPCR) and colony-forming units (CFU) was investigated. Results The geometric mean (95% CI) PTX3 level at stable state was 50.5 ng/mL (41.4–61.7). PTX3 levels correlated with absolute neutrophil count in sputum (r=0.37; P<0.01), but not FEV1 or health status. There was a weak correlation between PTX3 and bacterial load (CFU: r=0.29, P<0.01; 16S qPCR: r=0.18, P=0.05). PTX3 was a poor predictor of bacterial colonization (defined as >105 CFU/mL at stable state) with a receiver-operating characteristic (ROC) area under the curve (AUC) of 0.59 and 95% confidence interval (CI) 0.43–0.76 (P=0.21). During an exacerbation, there was a modest increase in PTX3 (fold difference 0.15, 95% of difference 0.02–0.29; P=0.02), and PTX3 fared better at identifying a bacteria-associated exacerbation (ROC AUC 0.65, 95% CI 0.52–0.78, P=0.03). Conclusion PTX3 is associated with bacterial infection in patients with COPD, but its utility as a biomarker for identifying a bacteria-associated exacerbation warrants further studies.


Introduction
COPD is characterized by persistent airflow limitation, which is progressive and associated with an enhanced inflammatory response to noxious agents. 1 Exacerbations of COPD have a significant effect on the health of patients 2 and are associated with microbial 3-7 and airway inflammation. 8 Chronic bacterial infection, defined as .10 5 colony-forming units (CFU), 9 occurs in 30% of patients at stable state 10 and increases up to 50% during an exacerbation. 3 Guidelines for treatment of an exacerbation advocate the use of systemic corticosteroids 11 and antibiotics; 12 however, this is usually without evidence of a bacterial infection as standard culture techniques are not rapid. 13 The occurrence of resistant bacteria is a growing concern 14 and deemed an emerging medical catastrophe. 15 The rapid identification of a bacterial exacerbation is urgently needed to aid clinical treatment decisions. One such possible marker is pentraxin 3 (PTX3). PTX3 is a soluble pattern recognition receptor, 16 recognizing pathogen-associated molecular patterns expressed by microorganisms. 17 PTX3 is induced in response to proinflammatory stimuli 16 and toll-like receptor (TLR) interactions 17 and contributes to innate resistance to pathogens. 16 PTX3 levels have been shown to be elevated in a murine bacterial infection model, 18 in sepsis, [19][20][21] and in inflammatory rheumatic disease. 22 In this study, we aim to determine whether sputum PTX3 is a sensitive biomarker for bacterial colonization at stable state and for a bacterial exacerbation in patients with COPD.

Materials and methods subjects and sampling
COPD subjects entering a longitudinal study looking at biomarkers in COPD were analyzed, where subject inclusion and exclusion criteria, study design, and measurements are as previously described. 3 In brief, subjects attended a stable state visit every 3 months over a 12-month period and also during exacerbations and 2 weeks post-exacerbation. An exacerbation event was defined according to Anthonisen criteria 23 and NICE guidance, 24 and a bacteria-associated exacerbation was defined as .10 7 CFU/mL as previously described. 3 At each visit, participants underwent pre-and postbronchodilator spirometry; blood was collected by standard venipuncture and sputum induction for sputum collection. If participants were unable to perform sputum induction, spontaneous sputum was collected. Health status and symptoms were measured using the St George's respiratory questionnaire (SGRQ), 25 MRC dyspnea scale, 26 the chronic respiratory disease questionnaire (CRQ), 27 and the visual analog score (VAS). 28 Chronic bacterial infection (colonization) was defined as .10 5 CFU/mL at stable state 9 and in this cohort was associated with a high positive predictive value of a respiratory pathogenic microorganism (Table 1). All subjects gave written informed consent, and the study was approved by the Leicestershire, Northamptonshire and Rutland ethics committee (reference number: 07/H0406/157).

sputum processing
Sputum processing involved plug selection, followed by a dispersion step with Dulbecco phosphate-buffered saline and a mucolytic step with dithiothreitol as previously described. 3 A filtration step to remove debris for cytospin preparation and quantification of cell differential count was then performed. 29 Samples with a cell viability of ,40% and a squamous contamination .20% were excluded. CFUs were prepared for semi-quantitative analysis using standard techniques. 30 A further 500 μL of the filtrate was processed by SYBR green (Applied Biosystems ® ; Life Technologies Corp., Carlsbad, CA, USA) quantitative polymerase chain reaction (qPCR) for Haemophilus influenzae and Staphylococcus aureus bacterial DNA, and a Taqman qPCR assay for the quantification of Streptococcus pneumoniae and Moraxella catarrhalis as described previously. 3 Viral RNA was extracted, using in-house assays, from an additional sputum plug for PCR analysis, as previously described. 3 Only samples with available sputum microbiology were analyzed. A sputum eosinophilia was defined as a sputum eosinophil count of .3%. 31

Pentraxin 3 Quantikine elIsa
A Human Pentraxin 3/TSG-14 Quantikine ELISA Kit (R&D Systems, Oxfordshire, UK) was used to measure PTX3 levels; the assay was carried out as per manufacturer's protocol and read using an EnVision plate reader (Perkin-Elmer, MA, USA). The lower limit of detection for PTX3 is 0.31 ng/mL. Readings taken at 560 nm wavelengths were subtracted from the readings taken at 450 nm to correct for any optical imperfections within the plate. The average optical density from the negative control was subtracted from each well to remove any background noise present. All standards, negatives, and samples were run in duplicate on each plate. The standards were plotted against the optical density value. The R 2 value was .0.97.

statistical analysis
GraphPad Prism version 6 (GraphPad Software, Inc., La Jolla, CA, USA) and Statistical Package for the Social Sciences (SPSS) Statistics version 22 (SPSS, Inc., Chicago, IL, USA) were used for statistical analysis. The Kolmogorov-Smirnov test was applied for normality. PTX3 levels were log-normal transformed. All parametric data were displayed as mean and standard deviation (SD), all non-parametric data were displayed as median (interquartile range [IQR]), and log-transformed data were presented as geometric mean and 95% confidence interval. Paired and unpaired t-test and one-way analysis of variance (ANOVA) tests were used to compare two, three, or more groups, respectively. Receiveroperating characteristic (ROC) curves were utilized to measure the sensitivity and specificity of PTX3 as a biomarker. Repeatability analysis was conducted in 10 subjects over  Figure 1A) and neutrophilic airway inflammation (r=0.37, P,0.01; Figure 1B). PTX3 was associated with bacterial load, measured by CFU/mL (r=0.29, P,0.01; Figure 1C) and total 16S qPCR (r=0.18, P=0.05; Figure 1D).  21) and was stable over time with a coefficient of variation 10% (Figure 2).

at exacerbation
During an exacerbation, there was a significant increase in PTX3 levels from a stable to exacerbation state (fold difference 0.15, 95% of difference 0.02-0.29; P=0.02). Subjects with an increase in PTX3 at the exacerbation were also more likely to culture a pathogen (  (Figure 3). In a small subset of subjects (n=11) with sputum PTX3 measured at a stable, exacerbation and exacerbation follow-up visit, a trend to increase in PTX3 from stable to exacerbation state

1202
Thulborn et al was seen in the 11 paired subjects and a trend to reduction at follow-up ( Figure 4).

Discussion
In this study, we aimed to determine if PTX3 was a sensitive biomarker for bacterial colonization at stable state and for a bacterial exacerbation in subjects with COPD, by measuring PTX3 concentrations in sputum. We have demonstrated for the first time that PTX3 in subjects with COPD correlates with bacterial burden, both at stable state and exacerbations and increases during an exacerbation event. PTX3 did not correlate with symptoms or lung function and was a reasonable predictor of a bacteria-associated exacerbation. Previous studies in non-COPD cohorts have shown PTX3 to rapidly increase due to an infection. 19,33 PTX3 has been shown to be elevated in COPD subjects compared with healthy controls in plasma 34 and sputum. 35 PTX3 levels have been correlated with the number of neutrophils present   in sputum and have been shown to be elevated in current and ex-smokers compared with non-smokers. 36 Our results support these statements. To our knowledge, no study has looked into PTX3 as a viable biomarker for a bacterial colonization or bacterial exacerbation in COPD subjects, though this has been suggested. 37 PTX3 levels have previously been linked to inflammatory diseases; 22 so it was not surprising to see a correlation between PTX3 levels and neutrophils in sputum, a pro-inflammatory cell, 38 and also a cell involved in PTX3 storage. 17 PTX3 has also been implemented in bacterial infection, [18][19][20][21] and our study supports this as we found a correlation between bacterial load (CFU/mL) in sputum at stable and exacerbation states implying that PTX3 expression in the lungs is driven by bacterial infection. This was further highlighted when PTX3 expression was relatively unchanged with the presence of a virus. Although non-significant, we did identify a moderate correlation of H. influenzae individual pathogen-specific qPCR with PTX3 during an exacerbation but not at stable state. This is unsurprising as H. influenzae is a commonly isolated pathogen in COPD patients 39 and is the main driver of inflammation. 40 Further work is needed to fully investigate the role of individual pathogens in PTX3 expression.

Limitations
This study has a few limitations. First, we did not have data on PTX3 in healthy controls, which would support whether levels detected in the COPD population are elevated; an additional infection control, such as bronchiectasis, would provide added value. However, this is the first study to look at infection (chronic and thus colonization) versus acute (bacteria-associated) exacerbations demonstrating modest utility of PTX3. Finally, advances in qPCR mean it is now possible to measure bacteria more accurately using plasmids 41 and to exclude dead bacteria. 42 As this was not available at the time of this work, we cannot confirm if the modest associations seen are a consequence of the inclusion of live and dead bacteria, and both these advancements in the field would greatly enhance this study.

Conclusion
To conclude, this study found that PTX3 correlates with bacterial load and that PTX3 is not sensitive or specific enough to be used as a biomarker for bacterial colonization but could be a potential biomarker for a bacteria-associated exacerbation in patients with COPD. However, further work is warranted to fully evaluate this.

Publish your work in this journal
Submit your manuscript here: http://www.dovepress.com/international-journal-of-chronic-obstructive-pulmonary-disease-journal The International Journal of COPD is an international, peer-reviewed journal of therapeutics and pharmacology focusing on concise rapid reporting of clinical studies and reviews in COPD. Special focus is given to the pathophysiological processes underlying the disease, intervention programs, patient focused education, and self management protocols.
This journal is indexed on PubMed Central, MedLine and CAS. The manuscript management system is completely online and includes a very quick and fair peer-review system, which is all easy to use. Visit http://www.dovepress.com/testimonials.php to read real quotes from published authors.