|Home | About | Journals | Submit | Contact Us | Français|
Adenotonsillar hypertrophy is the major pathophysiological mechanism underlying obstructive sleep apnea (OSA) and recurrent tonsillitis (RI) in children. The increased expression of various mediators of the inflammatory response in tonsils of OSA patients prompted our hypothesis that the enhanced local and systemic inflammation in OSA children would promote tonsillar proliferation. Mixed cell cultures from tonsils recovered during adenotonsillectomy in children with OSA and RI were established, and proliferative rates were assessed. Cells were also cultured to determine levels of pro-inflammatory cytokines and anti-oxidant protein levels and mRNA expression. Global cell proliferative rates from OSA tonsils were significantly higher than RI (P<0.01), with CD3+, CD4+, and CD8+ cell proliferation being higher in OSA (P<0.05). Moreover, pro-inflammatory cytokines such as TNF-α, IL-6, and IL-1α were highly expressed in OSA-derived tonsils. Furthermore, thioredoxin (TRX), an anti-oxidant protein, was also highly expressed in OSA tonsils at the mRNA and protein levels (p<0.01). Thus, T-cells are in a highly proliferative state in the tonsils of children with OSA, and are associated with increased production of proinflammatory cytokines and TRX, when compared to children with RI.
Obstructive sleep apnea (OSA), a condition characterized by repetitive increases in upper airway resistance and collapse, is a common health problem in children affecting 1–3% of children during their first decade of life (1). In a series of previous studies from our and other laboratories, it has become apparent that children with OSA are at increased risk for a vast array of morbidities, principally affecting the CNS and cardiovascular systems (2–9). Adenotonsillar hypertrophy has been recognized as the major pathophysiological contributor of OSA in children (10), and plays an important role as well in recurrent tonsillitis (RI) (11). Consequently, adenotonsillectomy (T&A) is currently the first line of treatment for children these conditions (12,13). However, the exact mechanisms underlying follicular lymphoid proliferation and hyperplasia remain extremely poorly understood. In adults, there are several lines of evidence suggesting that both local upper airway and systemic inflammation are implicated in the pathophysiology of this a priori mechanical dysfunction of the upper airway. For example, the number of immune cells is substantially increased in the upper airway mucosa and muscle of adults with OSA (14). Similar increases in regional and systemic inflammatory markers have also been reported in children with OSA (15–17). In addition, increased expression of mediators of the inflammatory response such as cysteinyl leukotrienes and glucocorticoid receptors is consistently found in tonsillar tissues of children with OSA (18–20). Therefore, we hypothesized that cellular proliferation of tonsillar tissues in children with OSA would differ from that of children RI, possibly reflecting different pathologic mechanisms and cell type involvement leading to the enlargement of upper airway lymphoid tissue in these 2 conditions.
The study was approved by the University of Louisville Human Research Committee, and informed consent was obtained for each participant. Consecutive children with adenotonsillar hypertrophy who underwent tonsillectomy for either OSA or RI were identified. Overnight polysomnography was performed using standard methods (21–23). The diagnosis of OSA was defined as an obstructive apnea-hypopnea index (AHI) ≥ 5 / hour of total sleep time. Diagnosis of recurrent tonsillitis (RI) was based on a history of >5 tonsillar infections over a period of < 6 months, requiring administration of antibiotics, in the absence of any symptoms suggestive of OSA using a validated questionnaire (24). Overnight sleep studies were performed in the majority of RI children (13 of 20), and showed AHI <1/hour of total sleep time.
Tonsils were placed in ice-cold PBS plus antibiotics and processing began <30 minutes after surgical excision. Tonsils were manually dissected, and gently grounded with a syringe plunger through a 70 µ mesh screen. Red blood cells were removed by lysis buffer. Cellular viability was determined by trypan blue exclusion. Specimens with a viability < 70% were discarded. Cell cultures were established in standard medium RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum (FBS) plus antibiotics, which included streptomycin, fungisone, gentamycin, and penicillin. Mixed cell suspensions were transferred onto 96-round bottom-well plates at a concentration of 1×106 cells/well, and cultured in a 5% CO2 incubator at 37°C for 48 hours. Cells were also cultured using 24-well plates to determine pro-inflammatory cytokine levels, to conduct flow cytometric analysis, and to extract RNA for real time quantitative PCR assays or protein for western blot analyses.
Cell proliferation was measured using a CellTiter 96 AQ nonradioactive cell proliferation assay (Promega, Madison, WI). Briefly, cells were plated in 96-well plates at a density of 1×106 cells/well in 200 µl of medium for 48 hours; then MTS [3-(4,5-dimethylthiazol-2yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium]-phenozine methosulfate solution (30 µl/well) was added to the wells. After incubation for 4 hours, the absorbance was measured at 490 nm using a plate reader (Multiscan EX, Thermo Scientific, Waltham, MA). Data representing the average of four wells were considered as one experiment.
To detect T-cell and B-cell specific proliferation, we employed BrdU pulsed proliferation analysis using flow cytometry. All procedures were measured using the APC BrdU flow kit (BD Biosciences, San Diego, CA) as recommended by the manufacturer. In brief, at the end of 48 hours of cell culture, cells were pulse-labeled with 1 mM BrdU for 4 hours. BrdU labeled cells were stained with a 3-color anti-body combination consisting of mouse anti-human CD45/PerCP Cy7, CD3/PE, and CD19/APC-Cy7 antibodies (BD Biosciences, San Diego, CA) in 50 µl staining buffer for 15 min on ice. Cells were fixed and permeabilized with cytofix/cytoperm buffer, suspended with DNase (300 µg/ml) for 1 hour at 37°C, and anti-BrdU APC antibody was added and incubated for 20 min at room temperature. Isotype controls relevant for each antibody were used to establish background fluorescence. Negative control was used as a sample, which was untreated with BrdU and was not stained with specific fluorescence antibodies. Data were acquired on a FACS Aria flow cytometer using the FACS Diva 5.5 software (BD Biosciences, San Diego, CA). After gating of lymphocytes based on CD45+ cells, T-cell and B-cell numbers were calculated as CD3+/CD19− and CD3−/CD19+ cell populations, respectively. Moreover, counting CD3+/BrdU+ and CD19+/BrdU+ cell populations identified proliferation of T-cells and B-cells. The results were displayed as two color dot-plots and analyzed by FlowJo software (Tree Star, San Carlos, CA). All data are expressed as the percentage of positive cell from the total cell population.
Total RNA was isolated with Qiagen RNAeasy mini kit (Qiagen Inc. CA). Total RNA was quantified by spectrophotometer based on the absorbance A260/A280 ratio. Quantitative real-time PCR were performed using TaqMan one-step RT-PCR reagents kits (Applied Biosystems, Foster City, CA) according to manufacturer’s instructions. The genes examined included TNF-alpha (NM_000594), IL-6 (NM_000600), IL-1 alpha (NM_000575), TRX (X77584), CD4 (NM_000616.), and CD8a (NM_171827). In brief, all reactions were carried out in a final volume of 20 µl containing: 11.5 µl 2× Master Mix, 0.5 µl RNase Inhibitor Mix, 1 µl primer (20 µM), 200 ng total RNA, and the volume of RNAse free water was adjusted with template. Thermo-cycling was run on an ABI 7500 real-time PCR system (Applied Biosystems, Foster City, CA) as follows: 30 min at 50°C, 10 min at 95°C, 40 cycles of 15 s at 95°C, 1 min at 60°C. Samples were normalized using the housekeeping gene, ribosomal 18S rRNA. Individual gene expression was calculated using the comparative Ct method (25). Specific mRNA expression in OSA was expressed as the relative fold increase compared to RI samples.
Pro-inflammatory cytokines levels such as TNF-α, IL-6, and IL-1 α were measured in cell-free supernatants with commercially available kits (R&D system, Minneapolis, MN) according to protocols provided by the manufacturer. Each of samples was assayed in duplicate. Inter-assay and intra-assay coefficients of variability ranged from 3.4 to 5.3% and 3.4 to 4.5%, respectively.
Cells were lysed using a nuclear extract kit (Active Motiff, Carlsbad, CA). Protein concentrations were measured using the DC protein assay (BioRad, Hercules, CA). Thioredoxin (TRX) expression was assessed by immunoblotting with anti-human TRX monoclonal antibody (1:1000, BD Biosciences, San Diego, CA) and mouse anti-rabbit IgGs that were conjugated to horseradish peroxidase in a Tris-buffered saline supplemented with 5% nonfat dry milk. Bands corresponding to TRX were visualized with ECL and quantified in a scanner with ImageQuant software (Molecular Dynamics, GE Health Science). Normalization of integrated densities was performed by reprobing membranes with β- actin antibody.
Coronal sections (40 µm) of tonsils from OSA and RI were initially incubated in 1xcitrate buffer (Lab Vision Corporation, Fremont, CA) at 95°C for 45min, washed several times in PBS, and blocked with a PBS/0.4% Triton X-100/0.5% TSA (Tyramide Signal Amplification, Perkin Elmer Life Sciences, Boston, MA) blocking reagent/10% normal horse serum for 1 hour. Sections were then serially incubated with anti-CD4 antibody (1:300, Santa Cruz Biotechnology, Santa Cruz, CA) or anti-CD8 antibody (1:1000, BD Pharmagen, San Jose, CA) at 4°C for 24 h, and then washed in PBS six times for 5 min each wash. Sections were incubated for 1 hour in horse anti-mouse biotinylated antibody (1:400, Vector Labs, Burlingame, CA) in a PBS/0.4% TSA blocking reagent/10% horse serum solution, and then with streptavidin-horseradish peroxidase diluted 1:100 in PBS/0.5% TSA blocking reagent. Subsequently, the sections were incubated with TSA fluorescein reagents diluted 1:50 in amplification diluent (Perkin Elmer Life Sciences) for 2 min. Sections were then washed and mounted onto glass slides. Negative controls were prepared by either omitting the primary or the secondary antibody. Sections were prepared from five sets of tonsils from either OSA or RI groups, and were visualized using a fluorescent microscope by an investigator who was blinded to the sample source.
All data were expressed by mean±SD. Statistical analyses were performed using SPSS software (version 16.0; SPPS Inc., Chicago, IL). All p-values reported are 2-tailed with statistical significance set at <0.05.
The demographic, polysomnographic, and tonsillar cell distribution characteristics of the OSA and RI cohorts are shown in Table 1. The mean age of children with OSA and RI was 6.2±2.8 years and 5.8±2.2 years, respectively, and genders and ethnicities were similarly represented. All OSA subjects had AHI >5 /hrTST, with 12 subjects having AHI < 10/hr TST, 10 subjects with AHI ≥ 10 but < 20 /hrTST, and 3 subjects with AHI > 20/hrTST. The total percentage of lymphocytes was higher in OSA children, and the percentage of CD3+ T lymphocytes was also higher in OSA children compared to RI (18.1±4.1% vs. 9.3±3.9%, p<0.05). In contrast, CD19+ B lymphocytes tended to be less abundant in OSA (42.8±3.8% vs. 48.1±3.5%; p=0.055).
Cellular proliferative assays showed that the tonsils of OSA children had significantly higher proliferative rates than those of RI (Figure 1; optical density units: 1.45±0.06 vs. 0.82±0.02, p<0.01).
To assess T cell and B cell proliferation, we employed the BrdU proliferation assay in combination with flow cytometry. Figure 2-A illustrates the strategy for identification of and sorting of T-cells and B-cells. As shown in figure 2-B, the proportion of proliferating of CD3+ T cells was significantly higher in OSA children compared to RI children (OSA vs. RI: 5.49±2.10% vs. 2.74±0.97%, P<0.05). However, proliferation of CD19+ B cells in children with OSA was reduced (OSA vs. RI: 2.80±1.04% vs. 4.3±1.41%, P<0.05). In addition, mRNA expression studies in tonsil cultures from children with OSA and RI confirmed that both CD4+ and CD8+ mRNA were highly expressed in children with OSA compared to RI (Figure 3), with CD4+ being the predominant cell type (Figure 3). Both these T-cell types were primarily located in the tonsillar extrafollicular area.
Figure 4 shows mRNA and protein expression of TRX in cells derived from tonsils in OSA and RI children. As seen in figure 4A, there was a 2.6 fold increase in TRX mRNA expression in OSA children (P<0.01). In addition, western blots confirmed these findings, such that TRX protein expression was markedly increased in tonsils from OSA children (OSA vs. RI; relative intensity: 0.86±0.19 vs. 0.28±0.16, P<0.01).
Figure 5A shows the concentrations in supernatants of proinflammatory cytokines. TNF-α and IL-6 concentrations were higher in children with OSA than in cultures from RI children (OSA vs. RI: TNF-α: 29.2±6.2 pg/ml vs. 10.5±5.6 pg/ml, P<0.01; IL-6: 49.6±7.3 pg/ml vs. 22.6±9.7 pg/ml, P<0.01). Similarly, IL-1α levels were also increased in OSA samples (OSA vs. RI: 42.2±9.81 pg/ml vs. 25.5±8.7 pg/ml, P<0.05). As seen in Figure 5B, mRNA expression of TNF- α, IL-6, and IL-1α in tonsillar cells from OSA children was significant higher than in children with RI (P<0.05).
In this study, we show that proliferative rates are increased in tonsil mixed cell cultures harvested from children with OSA during routine T&A compared to children with RI, and this process appears to be mediated by T-cells, while the reverse seems to occur in RI, with B cells assuming a more predominant role. Furthermore, both expression and release of pro-inflammatory cytokines to the supernatants, such as TNF- α, IL-6, and IL-1α, were increased in OSA, and in addition, evidence for up-regulation of the gene thioredoxin and its transcript protein in OSA-derived cells. Taken together, the data presented herein suggest that the mechanisms underlying upper airway lymphadenoid tissue proliferation in OSA and RI are distinct, and may allow for future non-surgical disease-specific therapeutic interventions that may ultimately obviate the need for T&A, while leading to involution of the hypertrophic adenoids and tonsils.
Several methodological issues deserve to be addressed. Firstly, the remarkable similarity between the qualitative findings in the tonsillar immunohistochemistry regarding cell type distribution and the percentage of lymphocytes sub-types identified in cell culture using flow cytometry would suggest that processing of the tonsillar tissues harvested during surgery did not alter the constitutive cell populations of these tissues. Moreover, the standard procedures used herein would be expected to have a similar influence on tonsils collected from children with OSA and with RI. Thus, the differences in proliferation, cytokine release, and TRX expression between the two cohorts appear to reflect the divergent intrinsic properties of these tissues, rather than reflect consequences of procedural methodologies. However, we should also remark that some differences in lymphocyte properties have been noted when different tissue processing procedures were used (26). Secondly, only palatine tonsils were included in this study, and therefore, differences in cytokine networks could emerge between nasopharyngeal lymphadenoid tissues and palatine tonsils, and future studies will have to examine this issue (27). Thirdly, the 2 groups had similar age, gender, and ethnicity, and overnight sleep studies were conducted in the majority of children with RI, such as to more objectively confirm the 2 major diagnoses leading to T&A in children. Obviously, tonsil tissues from healthy children are unavailable for ethical reasons. Finally, we did not specifically assess the contributions of allergic sensitization and viral exposures on the etiology of tonsillar proliferation in either OSA or RI. Previous studies have reported a high prevalence of allergic sensitization in children with OSA (28,29), and early exposure of respiratory viruses could modify the proliferative properties of tonsils through up-regulation of nerve growth factor and neurokinin 1 receptor dependent pathways (22,30,31). Thus, further studies regarding the potential contributions of these factors are needed.
There is no doubt that adenotonsillar hypertrophy constitutes the primary contributor to OSA in children, and even if T&A does not always result in cure, significant improvements in the severity of sleep-disordered breathing are usually the rule (32–35). However, very little is known on the mechanisms that mediate proliferation of tonsils in children with either OSA or RI. We have previously shown that pathways of inflammation play a role, since both topical corticosteroids and leukotriene receptor modifiers were found to be highly effective in reversing tonsillar hypertrophy and their receptors display increased expression patterns in the upper airways of children with OSA (16,18–21,36,37). Moreover, nerve growth factor and substance P, a major controller of sensorineural development and neuroimmuno-inflammatory responses, are also overexpressed in tonsils tissues from children with OSA (22). Notwithstanding such encouraging results, these studies did not quantitatively assess the proliferative characteristics of tonsillar tissues in OSA or RI, and did not provide information as to the major cellular populations that are regulated in this process. Therefore, we used a mixed in vitro cell culture model to more quantitatively investigate potential relationships between inflammatory responses and cell proliferation.
Considering the increased inflammatory markers in the upper airway of pediatric OSA patients (16–19,21,22), the increased proliferative activity in tonsils from children with OSA was not unanticipated. However, and of particular interest, were the findings on the differential patterns of proliferation among T-cells and B-cells in the 2 patient cohorts, with a T-cell predominant response in OSA being associated with higher expression and release of the proinflammatory cytokines TNF-α, IL-6, and IL-1 α. It is now accepted that recurrent vibration in the upper airway may promote the development of local inflammatory responses, leading to mucosal swelling (38,39), lymphadenoid tissue proliferation, and upper airway obstruction. Indeed, CD4+, CD8+, and activated CD25+ T cells were preferentially present in the mucosa and underlying muscle of the upper airway of adult patients with OSA (14). These findings have been recently confirmed (40). In children, Li et al (17) showed marked increases in sputum neutrophil counts in OSA that correlated with the severity of the disease. The increases in serum high sensitivity C-reactive protein levels in children OSA would further attest to the presence of a systemic, inflammatory process that could also contribute to the increased proliferation of the upper airway lymphoid tissues (41–44). Based on aforementioned considerations, the favorable response of children with OSA to topical corticosteroids (37,45), and current findings, we propose that local and systemic activation of inflammatory pathways will promote preferential T-cell proliferation and upper airway lymphoid hyperplasia.
TRX has been characterized as a new oxidative stress inducible protein, and plays an important role in intracellular signaling and resistance against oxidative stress (46). A multitude of stimuli may lead to increased TRX cellular expression, including hypoxia, lipopolysaccharide, hydrogen peroxide, and viral infections (47,48). TRX is actively secreted by a variety of normal and transformed cells, including fibroblasts, airway epithelial cells, and activated T cells (49). In a previous study, Park and colleagues (50) proposed that TRX is mechanistically involved in intermittent hypoxia-mediated alterations in the susceptibility of the heart to oxidative stress. Recently, and in support for a potential role of TRX in OSA, adult patients with this condition had higher plasma TRX levels that were reduced after treatment with continuous positive airway pressure (51). In the present study, we found that expression of TRX was markedly higher in tonsillar cells derived from children with OSA. While we are still unclear as to the mechanisms responsible for the differential expression of TRX, we postulate that TRX may be involved in protection from oxidative stress, and may also be modulating T-cell proliferative activity, although we cannot exclude a role in other inflammatory cell types, such as neutrophils. If the latter proves to be accurate, it may provide a viable target for development of interventional approaches for treatment or prevention of tonsillar hypertrophy in children with OSA.
In summary, we have established that T-cells are in a highly proliferative state in the tonsils of children with OSA, and are associated with increased production of pro-inflammatory cytokines and TRX, when compared to children with RI. These findings not only shed additional light on the differential regulatory mechanisms underlying tonsillar hypertrophy in 2 common pediatric conditions, namely, RI and OSA, but also provide an in vitro model that should permit objective characterization and development of specific treatments for these disease processes.
The authors thank Kenneth R. Brittian for technical assistance with immunohistochemical procedures.
Financial Support: DG is supported by National Institutes of Health grants HL-065270, HL-086662, and HL-083075, the Commonwealth of Kentucky Research Challenge for Excellence Trust Fund, and the Children’s Foundation Endowment for Sleep Research. LKG is supported by an investigator initiated grant from Merck Company.