High throughput screening of airway constriction in mouse lung slices

Abstract

The level of airway constriction in thin slices of lung tissue is highly variable. Owing to the labor-intensive nature of these experiments, determining the number of airways to be analyzed in order to allocate a reliable value of constriction in one mouse is challenging. Herein, a new automated device for physiology and image analysis was used to facilitate high throughput screening of airway constriction in lung slices. Airway constriction was first quantified in slices of lungs from male BALB/c mice with and without experimental asthma that were inflated with agarose through the trachea or trans-parenchymal injections. Random sampling simulations were then conducted to determine the number of airways required per mouse to quantify maximal constriction. The constriction of 45 ± 12 airways per mouse in 32 mice were analyzed. Mean maximal constriction was 37.4 ± 32.0%. The agarose inflating technique did not affect the methacholine response. However, the methacholine constriction was affected by experimental asthma (p = 0.003), shifting the methacholine concentration-response curve to the right, indicating a decreased sensitivity. Simulations then predicted that approximately 35, 16 and 29 airways per mouse are needed to quantify the maximal constriction mean, standard deviation and coefficient of variation, respectively; these numbers varying between mice and with experimental asthma.

Keywords: Airway smooth muscle; Asthma; Contraction; Mouse models; Precision-cut lung slices (PCLS).

Fig. 1

An example of an airway constricting in response to incremental concentrations of methacholine. Each picture represents the peak constriction captured by the physioLens at each concentration. The methacholine concentration is indicated above each picture. The percentage of constriction is also indicated underneath each picture. In this example, maximal constriction, measured in response to 10⁻⁴ M of methacholine, was 63.5%.

Fig. 2

Concentration–response curves, displaying airway constriction over increasing concentration of methacholine in mice exposed to either saline (blue) or house-dust mite (red) with their lungs inflated with either the traditional technique (darker colors) or with the needle technique (lighter colors). Results of the three-way ANOVA are shown below the graph. n = 8.

Fig. 3

Maximal airway constriction, showing the constriction in response to the highest concentration of methacholine tested (i.e., 10⁻⁴ M) in mice exposed to either saline (blue) or house-dust mite (red) with their lungs inflated with either the traditional technique or with the needle technique. Results of the two-way ANOVA are shown below the graph. n = 8.

Fig. 4

The EC50, showing the concentration of methacholine causing 50% of the maximal response in mice exposed to either saline (blue) or house-dust mite (red) with their lungs inflated with either the traditional technique or with the needle technique. Results of the two-way ANOVA are shown below the graph. n = 8.

Fig. 5

A representative trace of one iteration (black dots), showing the evolution of the running mean for the maximal airway constriction over 150 samples selected randomly from the dataset of one mouse (mouse 2, exposed to house-dust mite and with its lungs inflated with the traditional technique). The vertical light gray bars are the standard deviation for this specific iteration. Nine other traces are also shown in gray in the background. The blue horizontal dotted line is a landmark showing the final running mean (i.e., the running mean at n = 150). The blue lines in parallel represent the interval of tolerance, which was set in this example to be 20% above and below the final running mean. The red vertical dotted line is the point of stability, representing the minimum sample size (n = 26) for this specific iteration (in black) where all the subsequent running means never deviated from the final running mean by more than 20%. The orange and purple vertical dotted lines are landmarks showing the number of airways that were actually analyzed in this specific mouse (n = 41) and the average number of airways analyzed per mouse in this study (n = 45), respectively.

Fig. 6

Conformity over the sample size, showing how the percentage of conformity increases with increasing sample size at three different intervals of tolerance for each estimator (mean, SD, Coefficient of variation (CoV), and median) in each subgroup of mice. Every data point represents the mean ± SD of 8 mice. For clarity, only results of every fifth sample size are shown. The red horizontal dotted line in every graph is the set level of conformity used in the present study. The gray vertical dotted line in some graphs is aligned with the sample size where 80% conformity was achieved when the interval of tolerance was set at ± 20%. These latter sample sizes are also shown as the average of every subgroup of mice for each estimator in Table 2.