| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Respiratory Medicine and Allergy (Drs. Helleday and Blomberg) and Public Health and Clinical Medicine, Environmental and Occupational Medicine (Drs. Segerstedt, Forsberg, and Nordberg), Umeå University, Umeå, Sweden; Lung Biology, Pharmacology, and Therapeutics (Dr. Mudway), Pharmaceutical Sciences Research Division, King’s College London, London, UK; and Industrial Toxicology Unit (Dr. Bernard), Faculty of Medicine, Catholic University of Louvain, Brussels, Belgium.
Correspondence to: Anders Blomberg, MD, PhD, Department of Respiratory Medicine and Allergy, University Hospital, SE-901 85 Umeå, Sweden; e-mail: anders.blomberg{at}lung.umu.se
Abstract
We have previously demonstrated Clara cell protein (CC16) [secretoglobin 1A1] in serum to be a highly sensitive biomarker of altered lung epithelial permeability after ozone challenge. As a previous experimental study has indicated a diurnal variation in serum CC16 in humans, the aims of the present investigation were to confirm this observation and to attempt to model the diurnal variation in CC16 concentrations. In 18 healthy nonsmoking subjects, peripheral blood samples were drawn at six sampling points over a 15-h period and repeated twice within 3 to 4 weeks. A clear within-day variation was revealed in serum CC16 concentrations, falling significantly from baseline levels between the 11:30 AM and 10:00 PM time points (p = 0.000). Furthermore, it was shown that this within-day variation was reproducible regardless of subject or day, enabling the diurnal variation in serum CC16 to be modeled and fitted a second-degree polynomial for the observed time span. In conclusion, the present data demonstrate a pronounced time-dependent diurnal variation in serum levels of CC16, which can be mathematically compensated for, when addressing the issue of an air pollution-induced effect on CC16 in field studies.
Key Words: biomarker • Clara cell protein • diurnal variation • lung permeability • secretoglobin 1A1
Clara -cell protein (CC16) [secretoglobin 1A1], a microprotein of 16 kd, is an antiinflammatory protein and one of the major secretory products of the nonciliated bronchiolar Clara cells.12 A number of human and animal studies34 have reported that CC16 is present in high concentrations in airway lavage fluids, reflecting an intense secretion of CC16 into the respiratory tract lining fluids. Furthermore, CC16 also appears in serum, where its presence has been proposed to demonstrate a noninvasive and sensitive marker of lung epithelial injury, based on a passive leakage of CC16 across the lung epithelium into the blood circulation.5
CC16 in serum has been used to detect enhanced lung permeability in a variety of clinical and experimental situations, including sarcoidosis and exposure to tobacco smoke and ozone. Data from a field study67 have suggested that CC16 is a sensitive marker of ozone-induced lung epithelial damage. An experimental exposure chamber study8 has confirmed that serum CC16 is altered as a consequence of ozone exposure, and implied that CC16 may be used as a biomarker for ozone-induced lung epithelial injury. However, one of the surprising findings of that study was the appearance of a time-dependent diurnal variation in baseline serum CC16 concentrations in humans, complicating the interpretation of an ozone-induced response in CC16 and its simplistic use as a biomarker for ozone exposure. The aims of the present investigation were therefore to confirm the presence of a time-dependent diurnal variation in the serum concentrations of CC16 in healthy human subjects and, if so, to develop a mathematic model in order to correct for these time-dependent effects on serum CC16 when used as a biomarker for ozone exposure in field studies.
Materials and Methods
Subjects
Eighteen healthy volunteers participated (5 men and 13 women; mean age, 32 years; range, 22 to 51 years). All were never-smokers without a history of asthma or allergy and with normal lung function. Volunteers provided written informed consent, and the study was approved by the Umeå University Ethics Committee.
Study Design and Blood Sampling
In order to avoid confounding effects by ambient air pollution, the study was performed during the autumn, when background concentrations of ozone in the northern Sweden are low (< 0.02 ppm). Peripheral blood samples (10 mL) for analysis of CC16 in serum were obtained by venipuncture with collections into dry tubes (BD Vaccutainer, type ST; Belliver Industrial Estate; Plymouth, UK). Blood samples were drawn at six occasions over a 24-h period (7:00 AM, 9:00 AM, 11:30 AM, 2:30 PM, 6:00 PM, and 10:00 PM) and repeated on two separate dates within 3 weeks. Each blood sample was allowed to clot for a minimum of 1 h at room temperature. Samples were then centrifuged at 3,000g for 12 min. Serum was decanted and stored at – 80°C until protein analysis. CC16 was determined by latex immunoassay as previously described.8
Statistical Analysis
Statistical analysis was performed using a univariate general linear model. The factors used in the model were day, subject, and time of day. Day and subject were entered as random factors, and time of day was entered as a fixed factor. In the initial model, all two-factor interactions were included. Time of day was parameterized using simple contrast with the 7:00 AM time point as reference. In the second model, trying to develop a mathematical model for the time dependent changes, the time of day contrast was replaced with a second-degree polynomial measuring hours from 7:00 AM. Insignificant (p > 0.10) interactions were removed from the model. All statistical analyses were performed using statistical software (SPSS 11 for Windows; SPSS; Chicago, IL).
Results
An analysis of variance table for the initial model is presented in Table 1 , and the time of day contrasts are shown in Table 2 . The time effect showed a significant difference in CC16 concentrations from the 11:30 AM time point to the 10:00 PM time point, compared to the 7:00 AM baseline concentration (p = 0.000) [Table 2; Fig 1 ]. This effect, together with no observed significant effect for either the day x time or the subject x time interactions (Table 1) indicates a pronounced within-day variation and that the within-day variation is reproducible regardless of subject or day.
|
|
|
|
CC16 = – 0.582t + 0.032t2, where CC16 = change in CC16 serum concentration from the reference time point 7:00 AM, and t = number of hours after the reference time point. Both coefficients for t and t2 were significant, implying a nonlinear association between CC16 and the sampling time point. A third-degree polynomial was also estimated. However, the contribution of the third degree was marginal and not significant (data not shown). Discussion
The present data demonstrate a pronounced diurnal variation in the serum concentrations of CC16 with significant drops in the CC16 levels between 11.30 AM and 10:00 PM. This time-dependent diurnal variation in serum CC16 concentrations must be considered when CC16 is used as a biomarker in studies of exposure-induced effects.
In both a field study6 and an experimental exposure study,8 CC16 in serum has been suggested to represent a marker of ozone-induced epithelial damage68 that is much more sensitive than a traditional permeability marker such as albumin in airway lavage. An ideal biomarker for lung epithelial injury has to meet some important criteria. It has to be produced specifically by lung tissue and has to be easily measurable in extrapulmonary fluids. CC16 has both these characteristics. It is produced and secreted almost uniquely by the nonciliated bronchiolar Clara cells and is easily determined in peripheral blood using well-validated methods.15 However, as a suitable biomarker, the serum concentration of CC16 also has to be affected by a specific lung injury, such as ozone exposure in this case. As previously indicated, it has been shown that the serum concentration of CC16 increases significantly 2 to 4 h after exposure to ozone.8 However, in that study we found a consistent fall in serum CC16 concentrations after exposure to filtered air, implying the presence of a background diurnal variation in CC16 concentrations and hence of epithelial permeability. Without knowledge of the characteristics of this background diurnal variation in serum CC16 concentrations, that finding clearly complicates the use of CC16 as a biomarker for ozone exposure. In previous studies67 in which no compensation for a diurnal variation of serum CC16 was performed, the ozone-induced increases in CC16 have probably been underestimated. Further, it has to be acknowledged that exposures to tobacco smoke9 or nitrogen trichloride10 have been shown to affect the serum levels of CC16, indicating that other occupational exposures apart from ozone also may influence airway epithelial permeability in terms of serum CC16 levels. This fact has to be accounted for when using serum CC16 as a marker of ozone-induced epithelial damage.
In the present investigation, the most prominent effect on the CC16 concentration was a drop > 2.5 µg/L between baseline concentrations at 7:00 AM and those measured at 2:30 PM. As CC16 serum concentrations in normal healthy subjects, on average, range from 10 to 15 µg/L, a fall of this magnitude reflects a substantial reduction in the baseline CC16 concentrations, an effect that may be in the same range as, or even greater than, any air pollution-induced effect on circulating CC16 levels. Therefore, the time-dependent diurnal variation must be considered when CC16 is used as a biomarker for air pollution exposure. The present study also confirmed the presence of a major interindividual difference. Of importance, however, was that the time-dependent diurnal change in CC16 was similar regardless of the individual and the individual baseline concentrations. This made it possible to develop a mathematic model consisting of a second-degree polynomial to determine the fall in serum CC16 across a day, given a known baseline morning concentration. Furthermore, it does necessitate that serum collected for field study use has to be controlled for sampling time as well as compared to an individual baseline serum sample drawn on the same day. Given these precautions, CC16 may be employed as a biomarker for ozone exposure in field studies. An alteration from the calculated CC16 value at a certain time point, using the mathematic model, can thus be interpreted as an air pollution-induced effect on CC16.
It is well known that many proteins and hormones in the human body follow circadian rhythms. In the present study, we did not aim to study the reasons for the diurnal variation in the serum concentrations of CC16, and nothing is known about the underlying mechanisms here. Thus, at this stage, a few possible explanations can be speculated, which may be relevant either alone or, more probably, related to each other. A diurnal difference in production of CC16 in the Clara cells or a variation in CC16 secretion from Clara cells may be suggested. Differences in transepithelial leakage due to cyclic changes in the tightness of the epithelial tight junctions may also be implied. Other possible explanations may lie in pulmonary hemodynamics. During the night, CC16 may accumulate in the lung due to body position during sleep and maybe also in the serum as a consequence of decreased blood flow and reduced renal circulation, as CC16 is rapidly cleared by the kidneys. As noted in Figure 2, for 12 of 15 subjects, CC16 levels at 7:00 AM were higher on the subjects’ second sampling day compared to the first day. A plausible confounder not recorded in the present analysis may be the time frame between awakening and 7:00 AM. Studies are pending to address in more detail the mechanisms behind the diurnal variation in serum CC16 concentrations.
In conclusion, the present data demonstrate a pronounced time-dependent diurnal variation in the serum levels of CC16, which has to be considered when CC16 is used as a biomarker in studies of exposure-induced effects. However, as this variation follows a second-degree polynomial, it can be mathematically compensated for when addressing the issue of an air pollution-induced effect on CC16 in field studies.
Acknowledgements
The authors thank Mona Svensson, Annika Johansson, Helena Tjällgren, and Xavier Dumont for technical assistance, and Marie Eriksson and Leif Nilsson for statistical advice.
Footnotes
Abbreviations: CC16 = Clara cell protein; CI = confidence interval
No author has any pertinent involvement in any organization with a direct financial interest in the subject of this article.
This study was supported by the European Commission HELIOS project (QLK4CT-199-01308), the Swedish Heart-Lung Foundation, and Umeå University.
Received for publication November 24, 2005. Accepted for publication March 3, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
