Baseline Goblet Cell Mucin Secretion in the Airways Exceeds Stimulated Secretion over Extended Time Periods, and Is Sensitive to Shear Stress and Intracellular Mucin Stores

Abstract

Airway mucin secretion studies have focused on goblet cell responses to exogenous agonists almost to the exclusion of baseline mucin secretion (BLMS). In human bronchial epithelial cell cultures (HBECCs), maximal agonist-stimulated secretion exceeds baseline by ~3-fold as measured over hour-long periods, but mucin stores are discharged completely and require 24 h for full restoration. Hence, over 24 h, total baseline exceeds agonist-induced secretion by several-fold. Studies with HBECCs and mouse tracheas showed that BLMS is highly sensitive to mechanical stresses. Harvesting three consecutive 1 h baseline luminal incubations with HBECCs yielded equal rates of BLMS; however, lengthening the middle period to 72 h decreased the respective rate significantly, suggesting a stimulation of BLMS by the gentle washes of HBECC luminal surfaces. BLMS declined exponentially after washing HBECCs (t1/2 = 2.75 h), to rates approaching zero. HBECCs exposed to low perfusion rates exhibited spike-like increases in BLMS when flow was jumped 5-fold: BLMS increased >4 fold, then decreased within 5 min to a stable plateau at 1.5-2-fold over control. Higher flow jumps induced proportionally higher BLMS increases. Inducing mucous hyperplasia in HBECCs increased mucin production, BLMS and agonist-induced secretion. Mouse tracheal BLMS was ~6-fold higher during perfusion, than when flow was stopped. Munc13-2 null mouse tracheas, with their defect of accumulated cellular mucins, exhibited similar BLMS as WT, contrary to predictions of lower values. Graded mucous metaplasia induced in WT and Munc13-2 null tracheas with IL-13, caused proportional increases in BLMS, suggesting that naïve Munc13-2 mouse BLMS is elevated by increased mucin stores. We conclude that BLMS is, [i] a major component of mucin secretion in the lung, [ii] sustained by the mechanical activity of a dynamic lung, [iii] proportional to levels of mucin stores, and [iv] regulated differentially from agonist-induced mucin secretion.

Fig 1. Mucin secretion from HBECCs.

A. Measured mucins. Mucins were assessed by subunit ELISA for the accumulated mucus removed from HBECCs with a series of Careful Washes (Wash 1, 10 min, Wash 3–4, 1 h), and from the secretions released during 1 h baseline and agonist stimulation (100 μM ATPγS) periods. Data are expressed as the mean ± SE, n = 10 sets of HBECCs. B. Predicted mucin release. Mucin release predicted for a 24 h day from the data in Panel A. Predicted baseline was estimated by extended the 1 h baseline out to 24 h. Agonist-induced secretion over 24 h was estimated as just that released acutely (see text). Note the different scales, left and right.

Fig 2. Recovery of mucin stores in HBE cell cultures following agonist (ATPγS, 100 μM) induced discharge.

A. HBECCs examined by AB/PAS histology. HBECCs were fixed before (control) and periodically after exposure to agonist, and mucin stores were revealed by AB/PAS staining. B. Sample agarose Western blots. Whole cell extracts of HBECCs sampled at the indicated times were subjected to electrophoresis in agarose gels, vacuum blotted, and the blots probed for MUC5B with domain-specific antibodies. ‘Immature’, non-glycosylated mucin peptides were detected with an antibody to the PTS/mucin repeat domains, and ‘mature’, fully glycosylated mucins with antibodies to the Cys-rich domains within the glycosylated mucin domains. C. Time course of MUC5B peptides and mature mucins. Mucin quantitation was achieved by densitometry of agarose Western blots as in Panel B; all results are expressed relative to their respective controls. Left axis = MUC5B peptide as the immature, non-glycosylated form of the mucin, right axis = mature MUC5B glycoprotein. Time 0 = 45 min post-ATPγS, after cultures were washed to remove the secreted mucus; Controls (‘C’ in Panel B) = non-treated HBE cells. Data expressed as the mean ± SE (n = 7–9 sets).

Fig 3. Measured baseline secretion rates vary with period duration.

Following the Careful Wash protocol of Fig 1, HBECC mucin secretion rates were determined for three successive baseline periods, then the cultures were exposed to agonist (100 μM ATPγS). A. Uniform baseline periods. The baseline periods were all 1 hr in duration. B. Non-uniform baseline periods. The first and last baseline periods were 1 hr, but the middle one was 72 hr in duration. After the cultures were sampled for the B1 period, they were returned to the incubator for the 72 hr B2 period, and the mucins released were harvested by performing a 2^nd Careful Wash and combining the samples. n = 6; * p < 0.05 relative to period B1. Mucins in all samples were determined by subunit ELISA.

Fig 4. Time course of changes in HBECC baseline mucin secretion following a Careful Wash.

HBECCs were Carefully Washed and returned to the incubator under air:liquid luminal conditions. The cultures were then removed as a function of time and luminal mucus harvested with minimal perturbation. Cultures were sampled just once; different sets of cultures were used at the different time points (n = 4). Inset: Data plotted to show the total mucins harvested from the cultures over time. Mucins in all samples were determined by subunit ELISA.

Fig 5. Mucin secretory response of HBECCs to liquid shear stress.

HBECCs were perfused luminally at 100 μl/min during 2 h equilibration and 10 min baseline collection periods, then the perfusion rate was increased in a single step (arrow) by 2.5, 5, or 10 fold. Fractions collected during the baseline and increased flow periods were assessed for secreted mucins by the WGA ELLA. A. Time course showing the effects of a 5-fold increase (jump) in flow rate. Mucin secretion is plotted as mucin content/fraction, normalized to baseline (mean ± SE, n = 5). B. Relationship between the magnitude of the change in flow rate, and the normalized peak and integrated mucin secretory responses. The normalized peak and 30 min, integrated mucin secretory responses from the cultures subjected to different changes in flow-induced shear stress are plotted against the jump in flow rate. The abscissa shows the final flow rates, fold jump in flow, and the calculated, final steady-state shear stresses at each final flow rate used. Each point represents the mean ± SE of 5 or more experiments.

Fig 6. Time course of agonist-induced mucin secretion from human airways.

A. Comparing exocytosis (left axis) and secretions (right axis) in an individual perfused preparation. Analysis of the secretory response of a single, luminally perfused human nasal epithelial explant, circa 1993, measuring exocytic events in an individual goblet cell by microscopy, and mucin secretion from the same explant by ELISA (see [33]; shear stress ~66 mdyn/cm²). The time course of mucin secretion was corrected for the perfusion delay. Note the short duration of the exocytic burst, as measured by microscopy, and the relatively prolonged time course of mucin secretion determined by the ELISA. B. Time course of mucin secretion from HBECCs (non-perfused). Following determination of mucin secretion during a baseline period (blue bar; 30 min; n = 17), the mucins secreted in response to agonist (ATPγS, 100 μM) were determined by subunit ELISA as a function of time (green bars; n = 3–5 cultures/time point). Individual cultures were used for each time point. Note that the secretory response appears to be complete in the first 5 min.

Fig 7. Chronic effects of SMM (72 h) on HBECCs.

A. Effects on total mucins. Total mucins removed during the Careful Wash, and during the baseline and agonist-stimulated secretion periods, were determined individually, then summed. B. Effects on freshly secreted mucins. Mucins secreted at baseline and in response to ATPγS (100 μM, 45 min; * p < 0.05, n = 5)). Note the different ordinate scales in A and B. Mucins in all samples were determined by subunit ELISA.

Fig 8. Baseline mucin secretion from Munc13-2 mouse tracheas.

A. AB/PAS-stained sections of intrapulmonary bronchus, illustrating the degree of intracellular mucin accumulation by genotype. B. Mucins released during tracheal perfusion. Following a 30 min period of stopped-flow, perfusion was restarted and 5 min (50 μl) fractions collected. Pooled data are indicated in red. Note the line labeled ‘mean, perfused’, indicating the mean level of mucins released during Fractions 3–7: the area indicated in light gray, above the line for Fractions 1 and 2, represents mucins released during stopped-flow. Mucins in all samples were determined by subunit ELISA. C. Baseline mucin secretion during stopped-flow and perfusion. Data from (B), expressed as a rate. Despite the greater quantities of mucin released during stopped-flow (B), the corresponding rate of release is lower due to the longer duration of the 30 min period (vs. 5 min fractions for the perfused period). Note the similarities in stopped-flow and perfused baseline secretion between genotypes (NS; n = 5), despite the progressive increases in mucin stores in the Munc13-2 deficient mice (A).

Fig 9. Effects of ovalbumin-induced mucous metaplasia on baseline mucin secretion from WT mouse tracheas.

Tracheas from control and OVA-treated were equilibrated under continuous perfusion, and baseline mucin secretion was then determined under stopped-flow and following restoration of continuous perfusion, per Fig 2, using the subunit ELISA (* p < 0.05, n = 3). Insets at the top compare AB/PAS staining in sections of tracheas from control and OVA-treated mice.

Fig 10. Effects of graded, IL-13 induced mucous metaplasia on WT mouse mucin stores and secretion.

IL-13 (1 μg/instillation), or a sham, control solution, was instilled via the trachea into the airways of anesthetized mice, following Treatment protocols 0 (—), 1 (+), and 3 (+++) in Table 1. At the appropriate times following the last instillation, tracheas were harvested for mucin secretion experiments and the right lung lobe of each mouse was fixed for histology. A. IL-13 Effects on AB/PAS staining in the airways. Sections of the major intralobar bronchi were stained with AB/PAS and the degree of AB/PAS staining was quantified using ImageJ. Note the progressive mucous metaplasia and in Treatment 3 (+++) the presence of mucus plugs. The data are expressed as the integrated density per μm of basement membrane, using Box Plots (25^th, 50 ^th, 75 ^th percentiles). B. Relationship between AB/PAS staining and mucin secretion. Following IL-13 treatment, excised mouse tracheas were mounted for perfusion and equilibrated for 2 h, following which perfusion was stopped for the ‘Stopped-flow Baseline’, then restarted for the subsequent ‘Perfused Baseline’ and ‘Agonist Stimulated’ periods. The results are presented as scatter plots, with the measured mucin secretion rates, determined from subunit ELISAs, plotted against the AB/PAS Integrated Densities determined for the corresponding tissues from Panel A. Triangles, circles, and squares denote control and the 2 different IL-13 treatments, as indicated by the code, upper-right. Empty symbols denote secretions at baseline (stopped-flow and perfused) and filled symbols denote agonist-stimulated secretions. For the agonist-stimulated, IL-13+++ data (solid green squares), note the high value point that appears with the cluster of empty squares—to allow a simple visual comparison of the slopes, the scale for the agonist-stimulated data was not expanded to separate the data. All 3 slopes were significant, with correlation coefficients = 0.55, 0.61, and 0.59, and, with F ratio probabilities, p <0.005, <0.002, and <0.001.

Fig 11. Persistence of perfusion effects in P2Y2R null mouse.

After induction of mucous metaplasia with IL-13, tracheas from WT and P2Y2R null mice on the same 129S6 background were equilibrated under continuous perfusion, and baseline mucin secretion was then determined under stopped-flow and following restoration of continuous perfusion, from subunit ELISAs, per Fig 10 (* p < 0.05, n = 3 or 4).