Anal. Chem. 2009, 81, 662–668
Airway Monitoring by Collection and Mass Spectrometric Analysis of Exhaled Particles Ann-Charlotte Almstrand,† Evert Ljungstro ¨ m,‡ Jukka Lausmaa,§ Björn Bake,| Peter Sjo¨vall,§ and ,† Anna-Carin Olin* Occupational and Environmental Medicine, Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden, Atmospheric Science, Department of Chemistry at University of Gothenburg, Gothenburg, Sweden, Chemistry and Materials Technology, SP Technical Research Institute of Sweden, Borås, Sweden, Respiratory Medicine and Allergology, Sahlgrenska University Hospital, Gothenburg, Sweden We describe a new method for simultaneously collecting particles in exhaled air for subsequent chemical analysis and measuring their size distribution. After forced exhalation, particles were counted and collected in spots on silicon wafers with a cascade impactor. Several phospholipids were identified by time-of-flight secondary ion mass spectrometric analysis of the collected spots, suggesting that the particles originated from the lower airways. The amount of particles collected in ten exhalations was sufficient for characterizing the phospholipid composition. The feasibility of the technique in respiratory research is demonstrated by analysis of the phospholipid composition of exhaled particles from healthy controls, patients with asthma, and patients with cystic fibrosis. We believe this technology will be useful for monitoring patients with respiratory disease and has a high potential to detect new biomarkers in exhaled air. Noninvasive monitoring of respiratory diseases has traditionally involved measuring respiratory volume. Lung volume measurements were initiated in the 17th century and established as a diagnostic tool (mainly for tuberculosis) in the 19th century. In the 1950s, exhalation flow was related to expiratory volume, and forced expiratory volume in 1 s (FEV1) was introduced, providing more sensitive information on lung disorders.1 Breath analysis started in the late 18th century with the work of Lavoisier, who was the first to analyze oxygen and carbon dioxide in exhaled air.2 More than 500 volatile compounds have been detected in exhaled air, and considerable effort has been expended to develop methods for sampling and analyzing those compounds. Exhaled nitric oxide, generated primarily in the bronchial system, is now used to diagnose allergic asthma, and volatile organic compounds in exhaled air are being investigated as * To whom correspondence should be addressed. Address: Occupational and Environmental Medicine, Sahlgrenska Academy, Box 414, SE-405 30 Gothenburg, Sweden. Phone: +46 (0)31 786 62 91. Fax: +46 (0)31 40 97 28. E-mail:
[email protected]. † Sahlgrenska Academy at University of Gothenburg. ‡ Department of Chemistry at University of Gothenburg. § SP Technical Research Institute of Sweden. | Sahlgrenska University Hospital. (1) Gibson, G. J. Breathe 2005, 1, 206–216. (2) Phillips, M. Sci. Am. 1992, 267, 74–9.
662
Analytical Chemistry, Vol. 81, No. 2, January 15, 2009
potential biomarkers for systemic diseases such as lung and breast cancer.3,4 Exhaled air also contains nonvolatile compounds, which are transported by aerosol particles, presumably originating from the respiratory tract lining fluid (RTLF).5 The composition of RTLF varies in different respiratory compartments. A major component is surfactant, which originates from the alveoli and contains phospholipids, nucleic acids, proteins, and peptides.6 Other components of RTLF are glycoproteins and mucins from the conducting airways. Surfactant is a bronchial muscle relaxant and may be pivotal in modulating innate immunity.7,8 Its phospholipid and protein composition is altered in airway disease.9 Identification and quantification of the compounds present in RTLF may therefore be related to early detection of lung disease, disease progression, severity of disease, and to therapy response. Particles exhaled during tidal breathing are normally smaller than 1 µm.10 The mechanism and exact location of particle formation in the airways are unclear, and the chemical composition of particles has never been analyzed specifically. In this work, we describe a technique for simultaneously collecting exhaled particles and measuring their size distribution. The method is based on maximizing the number of collected exhaled particles and concentrating them onto a solid substrate with a cascade impactor. The collected material is directly available for subsequent mass spectrometric chemical analysis by time-offlight secondary ion mass spectrometry (TOF-SIMS). After dissolution or additional sample preparation steps, the collected material can also be analyzed with other techniques, such as LC-MS, MALDI-MS, or biochemical assays, although such analyses are not reported here. (3) Phillips, M.; Cataneo, R. N.; Cummin, A. R.; Gagliardi, A. J.; Gleeson, K.; Greenberg, J.; Maxfield, R. A.; Rom, W. N. Chest 2003, 123, 2115–23. (4) Phillips, M.; Cataneo, R. N.; Ditkoff, B. A.; Fisher, P.; Greenberg, J.; Gunawardena, R.; Kwon, C. S.; Rahbari-Oskoui, F.; Wong, C. Breast J. 2003, 9, 184–91. (5) Scheideler, L.; Manke, H. G.; Schwulera, U.; Inacker, O.; Hammerle, H. Am. Rev. Respir. Dis. 1993, 148, 778–84. (6) Harwood, J.; Morgan, L.; Greatrex, T. In Pulmonary Biology in Health and Disease; Springer: New York, 2004; pp 44-63. (7) Koetzler, R.; Saifeddine, M.; Yu, Z.; Schurch, F. S.; Hollenberg, M. D.; Green, F. H. Am. J. Respir. Cell Mol. Biol. 2006, 34, 609–15. (8) Chiba, H.; Piboonpocanun, S.; Mitsuzawa, H.; Kuronuma, K.; Murphy, R. C.; Voelker, D. R. Respirology 2006, 11, S2-6. (9) Griese, M. Eur. Respir. J. 1999, 13, 1455–76. (10) Papineni, R. S.; Rosenthal, F. S. J. Aerosol Med. 1997, 10, 105–16. 10.1021/ac802055k CCC: $40.75 2009 American Chemical Society Published on Web 12/05/2008
Table 1. Basic Data on Healthy Subjects and Subjects with Asthma or CF
healthy asthma CF
n
age, years; mean (range)
sex (F/M)
FEV1, % pred; meana (range)
4 4 4
37 (26-46) 35 (25-50) 25 (17-37)
3/1 3/1 1/3
105.0 (97-112) 100.5 (82-115) 94.7 (69-123)
a FEV1 is the forced expiratory volume in 1 s and is here expressed as percentage of the predicted value.
Figure 1. Experimental setup: (1) Thermally insulated box of dimensions 52 × 46 × 121 cm. (2) Mouthpiece through which the patient inhales and exhales. Inhalation of room air takes place through the particle filter (3). The mouthpiece temperature is maintained by a temperature controller (4) using a Pt-100 thermometer and a heating tape surrounding the mouthpiece. (5) Particle counter/sizer. (6) Inertial impactor. (7) Vacuum pump serving the impactor. (8) Tubular reservoir for exhaled air. (9) Flow meter for measurement of excess air venting into the room. (10) Box temperature control consisting of a Pt-100 thermometer attached to the tubular reservoir, an electrical heater, circulation fan, and a controller. (11) Moist air inlet, with a flow slightly larger than the combined flow of the particle counter (5) and the impactor (6). (12) Air humidifier consisting of a flask containing distilled water, heated by a heating mantle. Clean, filtered air enters from 13 and becomes saturated with water vapor at ∼50 °C. This air then passes through the gas cooler, kept at 36 °C by circulating water from a thermostat bath. The excess water is lost by condensation until the dew point is reached.
EXPERIMENTAL SECTION Collection of Exhaled Particles. The apparatus for collecting particles in exhaled air (schematically illustrated in Figure 1) was designed to provide an efficient collection of exhaled aerosol without altering particle size through evaporation or condensation. To facilitate chemical analysis, particles were recovered in a concentrated form. Through a mouthpiece with a two-way valve, the subject inhaled filtered, particle-free room air and exhaled sample air into a reservoir. Both the reservoir and the mouthpiece are thermostat-controlled at 36 °C. A Grimm 1.108 optical particle counter continuously draws 1.2 L min-1 sample air from the reservoir and counts and sizes particles from 0.3 to 20 µm in 15 size intervals with 6 s time resolution. In addition, a threestage inertial impactor/pump (three-stage PM10 Impaktor Dekati LTd, Tampere, Finland) draws a continuous sample at 16 L min-1. Clean, particle-free air, saturated with water vapor at 34 °C, is added at the end of the reservoir at 18.5 L min-1 to serve as a buffer and to prevent room air from entering the system through the vent. The impactor collects particles according to their inertia, a quantity related to their size.11 The exhaled aerosol is passed through a nozzle and directed toward an impaction plate, placed at 90° to the nozzle axis to deflect the gas stream. Particles with sufficient inertia will strike the plate, while particles with less (11) Finlayson-Pitts, B. J., Pitts, J. N., Jr. In Chemistry of the upper and lower atmosphere; Academic Press: San Diego, CA, 2000; p 547-656.
inertia will remain in the gas stream, passing the plate and remaining airborne. Particles can be collected by size by arranging several stages with nozzles and plates so that the gas velocity in the nozzles increases for each stage, and smaller particles with less inertia impact on each ensuing stage. The impactor has cutoff sizes of 7, 1.5, and 0.5 µm for the three stages, respectively. For design reasons, each stage may have several nozzles, and the deposited material on the plate underneath the nozzles reflects the nozzle pattern (e.g., as 10 small “spots” of collected material for the last stage). Silicon wafers were used as impaction plates and placed on the last stage (cutoff size 0.5 µm) of the impactor. On this stage, particles in an estimated size interval of 0.5-2.0 are collected. Before use, the wafers were ultrasonically cleaned sequentially with heptane, acetone, ethanol, and Milli-Q water, dried with a nitrogen stream, treated with UV-ozone for at least 10 min, washed with Milli-Q water, and dried with a nitrogen stream. After this protocol, the wafer surfaces were highly hydrophilic. Expiratory flow rate was recorded with an ultrasonic flow meter (OEM Flow Sensor Spiroson-AS, ndd Medical Technologies, Zu¨rich, Switzerland). The flow rate is displayed on a computer screen, so subjects can observe and control their breathing. In preliminary studies, forced breathing produced significantly higher particle concentrations than tidal breathing (unpublished data) and was therefore used during sampling. The subjects were trained to perform repeated consecutive exhalations (with nose clips) with a maximal exhalation flow corresponding to 80% of their individual maximal FEV1 value. A 10% deviation from the target flow was considered acceptable. For each measurement period, the concentrations of particles between 0.5 and 2.0 were summed and averaged. Particle Sampling. Introductory studies for method evaluation were performed on four nonsmoking healthy subjects with normal lung function (ages 26-57, 3 females). Sampling was performed for 15 min on the morning of day 1 and repeated on day 2. Particle numbers were recorded, and particles were collected on silicon wafers. In addition, a saliva droplet from all subjects was collected on silicon wafers on day 2. The wafers were stored at room temperature until TOF-SIMS analysis, which was performed on day 2. Exhaled Particles in Healthy Subjects and Patients with Asthma or Cystic Fibrosis. On the basis of the design developed in the introductory study, the method was tested in four healthy subjects, four patients with physician-diagnosed asthma, and four patients with cystic fibrosis (CF). All subjects were nonsmokers. For basic data on the subjects, see Table 1. The subjects with asthma reported asthma symptoms during the last month, but were considered as having a well-controlled and stable disease. Analytical Chemistry, Vol. 81, No. 2, January 15, 2009
663
Two of them used inhaled steroids, and two used only short-acting β2-agonists. Two subjects with CF had exacerbation of their disease at the time of sampling (indicated by an increase in symptoms, increase in C-reactive protein, and a need for i.v. antibiotic therapy). The other two subjects with CF had stable disease. Subjects breathed particle-free air during a 3-min washout period before sampling. Subjects then performed 10 consecutive forced exhalations. Particle concentration was recorded, and exhaled particles were collected on silicon wafers for subsequent analysis by TOF-SIMS. The samples were stored at -20 °C until analysis, which was performed within 4 weeks. The study was approved by the local ethics committee of the Sahlgrenska Academy of Gothenburg. TOF-SIMS Analysis. Sample spots were analyzed with a TOFSIMS IV instrument (IONTOF, Mu¨nster, Germany). Spots containing the collected particles were localized with a video camera in the instrument. The analysis was done by rastering a 25-keV Bi3+ primary ion beam over an area of 500 × 500 µm2 centered on the selected spot unless stated otherwise. Mass spectra of positive and negative secondary ions were recorded with the instrument optimized for maximal mass resolution (bunched mode: m/µm ∼3000-6000, beam diameter of ∼5 µm), at a repetition rate of 6.7 kHz and a pulsed primary ion current of 0.1 pA. The accumulated primary ion dose was kept well below the so-called static limit of
