Exhaled Breath Condensate for Proteomic Biomarker Discovery

Exhaled breath condensate (EBC) has been established as a potential source of respiratory biomarkers. Compared to the numerous small molecules identified, the protein content of EBC has remained relatively unstudied due to the methodological and technical difficulties surrounding EBC analysis. In this review, we discuss the proteins identified in EBC, by mass spectrometry, focusing on the significance of those proteins identified. We will also review the limitations surrounding mass spectral EBC protein analysis emphasizing recommendations to enhance EBC protein identifications by mass spectrometry. Finally, we will provide insight into the future directions of the EBC proteomics field.


Introduction
Exhaled breath condensate (EBC) has quickly become a subject of research interest due to the high potential for respiratory disease biomarker discovery.Considered an aqueous matrix featuring condensed volatile organic compounds (VOC) from breath and non-volatile droplets of liquid from the respiratory tract, EBC represents an expanded potential for respiratory biomarker discovery beyond VOC breath analysis [1][2][3][4][5].These properties have allowed EBC to come to the forefront of collection methods for biomarker discovery in respiratory diseases.
Historically, the biomarker discovery field has relied heavily on mass spectrometric approaches for characterizing the contents of complex biological samples and fluids.Exhaled breath condensate is no different.Both gas chromatography and liquid chromatography mass spectrometry (GC-MS and LC-MS/MS respectively) approaches have been applied to EBC to identify molecules of prognostic and predictive significance.This review will discuss the proteins identified using mass spectral methodologies for EBC biomarker discovery with a focus on the significance and limitations associated with EBC analysis.

Collection and Sampling
In contrast to invasive and minimally invasive collection methods, such as tissue biopsy, blood and bronchoalveolar lavage (BAL), exhaled breath condensate is obtained by a completely non-invasive method making it ideal for ailing individuals [6,7].Furthermore, EBC collection is simplistic, requiring minimal technical skills, allowing for sampling from both children and adults alike [6,7].These properties make EBC collection and analysis applicable to many respiratory diseases, such as asthma, cystic fibrosis and chronic obstructive pulmonary disease (COPD) [6,7].
Sampling of EBC can be conducted on commercially available equipment, most notably the Ecoscreen or RTube, or homemade devices such as that put forth by Schleiss et al. [4,8,9].See Huttmann et al., for details regarding the Ecoscreen and RTube [10].All these sampling devices strive to condense humid breath, without salivary contamination, into liquid or ice, based on the condensation temperature, for collection and analysis [11].For review of collection, device variability, sampling pitfalls and recommendations for collection method standardization, in EBC sampling see Horvath et al., Montuschi and Grob et al. [1,8,12].Since EBC is >99% water, analytes are diluted to detection limits or lower [13].Steps to concentrate analytes, such as centrifugal evaporation or lyophilization, have been employed to overcome this limitation [4,8].Despite this analytical obstacle, it is these concentrated samples that hold the most promise for protein biomarker discovery.

Exhaled Breath Condensate Proteomics
In contrast to global bottom-up proteomics, EBC proteomic analysis has yielded only a modest number of protein identifications (Table 1).For example, shotgun proteomics of global lysates regularly identify several hundred proteins in a single study with significant protein scores [14].Conversely, the largest number of significantly identified proteins from EBC using bottom-up approaches, in a single study, was accomplished by Fumagalli et al., with 44 total proteins [15].This has been attributed to the low concentration, <1 µg/mL, of proteins contained in EBC [16].While few in number, those already discovered have the potential for high physiologic relevance.The dominant proteins found in EBC are cytokeratin proteins.Most often Cytokeratins 1, 2, 9 and 10 are more prevalent while 4, 5, 6A, 6B, 8, 14, 16, 17, 25, and 26 are also present at a lower frequency [15][16][17][18][19][20][21][22][23][24][25].Several of these cytokeratins, specifically 5, 6, 8, 14 and 17, are expressed in alveolae and bronchii of the lung suggesting EBC is a representative sample of the lower respiratory tract environment [18].Similarly, Gianazza et al., have reported a 3× increase in the keratin content of EBC from smokers when compared to non-smokers [23,26].However, many of these cytokeratins and other proteins have been reported as contaminants, associated with either the sampling environment or patient derived, dust and skin debris in ambient air [18,24].Cytokeratins 1, 2 and 14, dermcidin and hemoglobin fall into this category (Table 1) [16,18,24].Potentially more important than cytokeratin identification, a recent study has found EBC derived inflammatory mediators, e.g., IL-1α, IL-1β, IL-2, IL-12α, IL-12β, IL-15, IFN-α, IFN-ϒ, and TNF-α, suggesting the proteomic community is progressing toward the lower end of the dynamic range of EBC [15].
Cytokines in exhaled breath condensate have been reported in the low pg/mL range often bordering the lower limits of detection and/or quantitation by common immunological methodology [27].For example, tumor necrosis factor alpha (TNF-α) was shown by Garey et al., by ELISA to range from 7.4 ± 17.5 to 3.9 ± 8.5 pg/mL between smokers and non-smokers which is near their reported 2 pg/mL detection limit for the assay [28].Mass spectral detection of these potentially low abundant and highly localized proteins by Fumagalli et al., in EBC suggests the methodological and instrumental limitations of EBC proteomics are slowly being overcome [15].Such results hold promise for using MS to characterize the EBC protein content, across many diseases, expanding our understanding of EBC as a source of protein biomarkers.
The identification of cytokines in exhaled breath condensate also holds diagnostic significance in determining lung inflammation.For example, IL-2 was detected in the EBC of children with asthma and CF while remaining undetectable in control children [29].Furthermore, protein array data provided by Matsunaga et al., showed increases in IL-4, IL-8, IL-17, TNF-α, RANTES, IP-10, TGF-β, MIP-1α and MIP-1β in asthmatics when compared to healthy individuals [30].Matsunaga et al., also correlated values of EBC RANTES, TNF-α, and TGF-β with physiological parameters of airway disease [30].Additionally, Colombo et al., demonstrated IL-8 correlated with clinical biomarkers of cystic fibrosis by biochip array [31].However, more sensitive assays, such as multiple reaction monitoring (MRM), are needed as these cytokines are often near immunological reagent dependent assay's detection limits [29].These examples emphasize the potential for using EBC as a medium to monitor lung inflammatory mediators.Additionally, these data stress the need for novel highly sensitive methods for EBC analysis.

Exhaled Breath Condensate Proteomic Limitations
In contrast to the numerous small molecule biomarkers identified, such as leukotrienes (LTB 4 , CysLT) and nitric oxide (NO), proteomic analysis of EBC has had a limited number of potential protein biomarkers identified due to methodological and instrumental hurdles [4,8,32,33].The analytical difficulties of EBC proteomics can be attributed to several factors including, low protein concentration, inconsistent sample preparation and sample loss and instrumental sensitivities [34][35][36].Although present, the limitations surrounding EBC proteomics can be potentially overcome.
Low protein content, <1 µg/mL, in EBC requires measures to concentrate the proteins prior to analysis [16].Traditionally accomplished through centrifugal evaporation or lyophilization, these steps are reported to concentrate samples up to 20% [4].Although beneficial, lyophilization has been shown to reduce peptide identifications 50%-90% suggesting other methods or combination of methods such as gel electrophoresis, ultrafiltration, protein precipitation and solid-phase extraction, for sample concentration could improve sample quality and protein ID quantity [15][16][17][18][19][20][21][22][23][24][25].Additionally, lyophilizing instruments are not readily available at most institutions limiting their potential use in EBC methodology [1].These results suggest lyophilization for sample concentration alone is insufficient to probe low abundant proteins in EBC.
The most common sampling procedure for EBC collection involves normal tidal breathing through a collection device for 10 minutes, ultimately yielding approximately 1 mL of condensate [1,2].Although variations exist, i.e., condensing temperature, collection device, etc., this procedure has been used routinely for many applications including proteomics [15][16][17][18][19][20][21][22][23][24][25].A 1 mL sample of EBC for proteomics is insufficient due to the low overall protein abundance.While increasing the sampling time is ideal, it may provide added stress to an already sick individual [16].To circumvent this limitation, Fumagalli et al., pooled samples from healthy smokers and non-smokers to increase sample protein concentration and obtain MS spectra from the low end of the dynamic range [15].While they could not characterize distinct individuals in this manner, sample pooling did allow for the MS detection of the highest number of proteins from EBC to date [15].
The proteomic analysis of EBC has been conducted on many different instruments, such as MALDI-TOF, SELDI-TOF, Ion-trap LC-MS/MS and LC-FT-ICR, each with variable sensitivities (Table 1) [15][16][17][18][19][20][21][22][23][24][25].As newer more sensitive instruments with high MS N scan rates become available, the field of EBC proteomics will quickly progress.Additionally as the interest in exhaled breath as a source of protein biomarkers grows, more researchers will become involved in EBC protein analysis further advancing the field.

Future Directions
Although significant strides have been made in qualitatively analyzing EBC, i.e., small molecule, proteomics, pH, etc., the clinical significance and application of these discoveries currently remains unclear.Further development of MS techniques and instrumentation, sample collection and preparation are required to advance the proteomic characterization to the low end of the dynamic range.Such research will allow for the identification of low abundant disease biomarkers facilitating the development of highly sensitive quantitative MS assays, i.e., multiple reaction monitoring, supporting large-scale clinical biomarker studies.It is these low abundant proteins in conjunction with high sensitivity quantitative assays that could hold significant clinically relevant diagnostic and therapeutic information.

Table 1 .
List of proteins identified in exhaled breath condensate by mass spectrometry in various diseases.Disease Abbreviations: NS = Non-Smoker, S = Smoker, ES = Ex-Smoker, LT = Lung Transplant, AATD = Pulmonary Emphysema Associated with α1