Content uploaded by Fatemeh Khadangi
Author content
All content in this area was uploaded by Fatemeh Khadangi on Jan 19, 2020
Content may be subject to copyright.
Airway smooth muscle adapting in dynamic conditions is refractory to the 1
bronchodilator effect of a deep inspiration. 2
3
4
5
6
Morgan Gazzola, Fatemeh Khadangi, Marine Clisson, Jonathan Beaudoin, Marie-Annick Clavel, 7
Ynuk Bossé 8
9
Laval University, Quebec, Canada 10
11
12
13
14
Corresponding author: 15
Ynuk Bossé 16
IUCPQ 17
Pavillon Mallet, M2694 18
2725, chemin Sainte-Foy 19
Québec, Qc, G1V 4G5 20
Phone: 418-656-8711(ext. 3489) 21
E-mail: ynuk.bosse@criucpq.ulaval.ca 22
23
24
25
26
Running head: ASM adaptations in static versus dynamic conditions 27
28
29
30
31
Word count: 4245 32
33
34
35
36
Sources of funding: 37
This work was supported by the Natural Sciences and Engineering Research Council of Canada 38
(NSERC). Morgan Gazzola and Ynuk Bossé were both supported by research scholars from 39
FRQS (Fonds de recherche du Québec – Santé). 40
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
Abstract (207 words) 41
42
Airway smooth muscle (ASM) is continuously strained during breathing at tidal volume. Whether 43
this tidal strain influences the magnitude of the bronchodilator response to a deep inspiration 44
(DI) is not clearly defined. The present in vitro study examines the effect of tidal strain on the 45
bronchodilator effect of DIs. ASM strips from sheep tracheas were mounted in organ baths and 46
then subjected to stretches (30% strain) simulating DIs at varying time intervals. In between 47
simulated DIs, the strips were either held at a fixed length (isometric) or oscillated continuously 48
by 6% (length oscillations) to simulate tidal strain. The contractile state of the strips was also 49
controlled by adding either methacholine or isoproterenol to activate or relax ASM, respectively. 50
Although the time-dependent gain in force caused by methacholine was attenuated by length 51
oscillations, part of the acquired force in the oscillating condition was preserved post-simulated 52
DIs, which was not the case in the isometric condition. Consequently, the bronchodilator effect 53
of simulated DIs (i.e., the decline in force post- versus pre-simulated DIs) was attenuated in 54
oscillating versus isometric conditions. These findings suggest that an ASM operating in a 55
dynamic environment acquired adaptations that make it refractory to the decline in contractility 56
inflicted by a larger strain simulating a DI. 57
58
59
Keywords: airway mechanics, contraction, strain, stress, bronchodilation, re-narrowing60
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
Introduction 61
The force generated by airway smooth muscle (ASM) is not only modulated by bronchoactive 62
molecules but also by strain (12, 15, 18, 19, 30, 45, 46). Being embedded in a dynamic organ 63
such as the lung, the ASM is strained continuously. The strain caused by breathing maneuvers 64
is highly variable and largely depends on the stiffness of the airway wall and the size of the 65
breaths. A deep inspiration (DI), for example, strains ASM by 15 to 30% (5). Many in vitro 66
studies have demonstrated that strain of this magnitude significantly decreases the contractile 67
capacity of ASM (15, 18, 20, 23, 26, 30-32, 46). This regulatory effect of DIs on ASM contraction 68
accounts, at least partially, for the bronchodilator effect of DIs post-bronchoconstriction (47). 69
70
The bronchodilator effect of the oscillating strain caused by breathing at tidal volume, hereafter 71
called tidal strain, is more controversial (12, 15, 20, 25). The effect of tidal strain on the 72
bronchodilator effect of a subsequent larger strain caused by a DI is also not clearly defined. 73
One can postulate that tidal strain pre-DI may soften ASM and, therefore, accentuate the strain 74
and the ensuing bronchodilator effect of a subsequent DI. Contrastingly, the results of our 75
recent in vitro study on ovine ASM strips demonstrated that tidal breathing is actually 76
decreasing the bronchodilator effect of a subsequent DI (17). This previous study was 77
conducted in force-controlled conditions, meaning that the distending force was controlled and 78
the resulting ASM length was measured. Therefore, the tidal strain was not controlled directly. 79
80
Herein, we used length-controlled experiments. We simulated DIs at different time intervals by 81
straining ASM by 30%. In between simulated DIs, the length of the ASM was either held fixed or 82
subjected to tidal strain. The contractile state was also controlled by adding either MCh to 83
activate the ASM or isoproterenol to relax the ASM. In combination, these experiments allowed 84
us to properly investigate the effect of strain at a magnitude simulating tidal breathing in vivo on 85
the bronchodilator effect of a subsequent simulated DI on both activated and relaxed ASM. 86
Based on our previous study (17), we hypothesized that a MCh-activated ASM subjected to tidal 87
strain would become refractory to the bronchodilator effect of a subsequent larger strain 88
simulating a DI. In other words, we expected that the decline in MCh-induced contraction 89
elicited by the simulated DI would be smaller for an ASM operating in a dynamic environment 90
than an ASM operating in a static environment. 91
92
93
94
Methods 95
Ovine tracheas 96
Dorsett hybrid sheep aged on average 6.1 ± 1.7 months were euthanized for other scientific 97
purposes. The protocol was approved by the Committee of Animal Care of Laval University in 98
accordance with the guidelines of the Canadian Council on Animal Care (Protocol 2018013). 99
The trachea was excised immediately after euthanasia, immersed into Krebs solution (pH 7.4, 100
111.9 mM NaCl, 5.0 mM KCl, 1.0 mM KH2PO4, 2.1 mM MgSO4, 29.8 mM NaHCO3, 11.5 mM 101
glucose, 2.9 mM CaCl2) and kept at 4 °C until further process. 102
103
Tracheal ASM strips 104
Two ASM strips from each trachea were isolated and mounted vertically in a 40-ml organ bath 105
at in situ length between 2 platinum electrodes (2 mm wide × 50 mm long) as previously 106
described (27). The bath was filled with Krebs solution maintained at 37°C by a jacketed bath 107
containing circulating heated water. The upper extremity of the strip was connected by a 108
surgical thread to a dual-mode servo-controlled lever arm system (model 300C; Aurora 109
Scientific Inc., Aurora, Canada). The latter not only measured force but also allowed changes of 110
length to be applied. The lower extremity of the strip was connected to a stationary hook at the 111
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
bottom of the bath. Before starting the protocol described below, the strips were subjected to a 112
period of preconditioning. Over such period, the strip was stimulated to contract every 5 min for 113
20 s with an electrical field (60 Hz, 20 volts, 2 ms) under isometric conditions. The 114
preconditioning lasted until the progressive increase in electrical field-induced force reached a 115
plateau. 116
117
Protocol 118
In each protocol, the strip was subjected to 4 large stretches (Figure 1). Each stretch strained 119
the ASM by 30% to simulate a DI. It was applied sinusoidally at a frequency of 0.1 Hz over 5 s 120
(ergo, a half-sine wave with no trough). The interval between simulated DIs was set to 2, 5, 10 121
and 30 min in a random order. During these intervals, the strips were either held fixed (i.e., 122
isometric) or strained continuously to simulate tidal breathing. More precisely, the tidal strain 123
consisted at oscillating the length of the strips sinusoidally at a magnitude of 3% (6% from 124
trough to peak) and at a frequency of 0.2 Hz. At all time, our system was thus controlling the 125
length while the force was measured. 126
127
The protocol was also performed under two contractile states: 1- During contraction elicited by 128
10-5 M of MCh; and 2- during relaxation elicited by 10-6 M of isoproterenol. The order of which 129
was randomized. Overall, each strip underwent 4 protocols: Contracted with MCh and relaxed 130
with isoproterenol with and without tidal strain in between simulated DIs. 131
132
A sequence of interventions was used between protocols to reset history and to accustom ASM 133
to both the contractile state (MCh versus isoproterenol) and the condition under which it had to 134
operate in the following protocol (i.e., with or without tidal strain in between simulated DIs) (29). 135
Briefly, it consisted of a period of 10 min flanked by a series of three consecutive DIs. 136
137
Data analyses 138
The cross-sectional area of the strip was estimated to then convert force (mN) into stress 139
(mN/mm2). The details were described previously (17). Briefly, it was assumed that the stress 140
generated by ASM in response to an electrical field stimulation after the conditioning period 141
when the strip was set at in situ length was 100 mN/mm2, which fairly corresponds to reported 142
values with similar preparations (9, 39). The area was then estimated as follows: area = 143
measured force / 100 mN/mm2. 144
145
All the data are presented as means ± standard deviations. The stresses shown are averages 146
computed across full simulated breaths, or across the corresponding periods when there was no 147
tidal strain. This was to ensure that the comparisons with and without tidal strain were all made 148
at the same average length. Seven variables were measured: 149
1- The stress during the last tidal breath before the simulated DIs. 150
2- The elastance during the last tidal breath before the simulated DIs. Elastance was 151
calculated using the traditional equation of motion that describes the behavior of a linear 152
system during a sinusoidal strain: namely σ = ε(E + iωR), where σ is stress, ε is strain, ω 153
is angular frequency, i is the imaginary number, E is elastance and R is resistance. 154
Obviously, these latter calculations could only be done when tidal strain was applied in 155
between simulated DIs. 156
3- The stiffness before the simulated DIs. The stiffness was calculated as the change in 157
stress from trough to peak during the last simulated tidal breath prior to the simulated 158
DIs divided by the excursion of length over the same period. Again, these latter 159
calculations could only be done when tidal strain was applied in between simulated DIs. 160
4- The elastance during the simulated DIs. The equation of motion described above could 161
not be used during the simulated DIs as the assumption on linearity is certainly violated 162
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
during such a large excursion of length. We therefore opted for a method requiring no 163
assumption on linearity. The method was previously expatiated (14). The math is below. 164
Briefly, it consists of measuring the area between the ascending and the descending 165
stress-strain curves during the simulated DIs, which is then used to sequentially deduce 166
resistance, the phase angle, hysteresivity and elastance. 167
168
• R = 4A/πω(ε2), where A is area, R is resistance, ω is angular frequency and ε is 169
strain. 170
171
• φ = arcsin ωRε/Δσ, where φ is the phase angle and Δσ is the change in stress. 172
173
• tan φ = η, where η is hysteresivity. 174
175
• E = ωR/η, where E is elastance. 176
177
5- The stiffness during the simulated DIs. As for the stiffness during a simulated tidal 178
breath, the stiffness during the DIs was calculated as the change in stress from trough to 179
peak divided by the excursion of length over the same period. 180
6- The stress during the first tidal breath immediately after the simulated DIs. 181
7- The decline in stress induced by the simulated DIs, which is also referred to as the 182
bronchodilator effect. The bronchodilator effect was calculated from the difference in 183
stress between the simulated tidal breaths immediately after versus immediately before 184
the simulated DIs. 185
186
The effects of changing the contractile state (MCh versus isoproterenol), the condition under 187
which ASM operated (with or without tidal strain) and the interval between simulated DIs (the 188
effect of time), as well as their paired interactions, on variables 1, 4, 5, 6 and 7 described above 189
were analyzed using repeated measures three-way ANOVAs. The variables 2 and 3 (elastance 190
and stiffness before the simulated DIs) were analyzed using repeated measures two-way 191
ANOVAs to measure the effects of changing the contractile state, the time interval and their 192
interaction. This is because these latter variables could not be measured in the absence of tidal 193
strain. Statistical analyses were performed using Prism 8 (Version 8.1.1, GraphPad Software, 194
San Diego, CA) and p ≤ 0.05 was deemed significant. 195
196
197
198
Results 199
ASM stress immediately before the simulated DIs is displayed in Figure 2. The stress was not 200
significantly affected by oscillating the length of ASM strips in between simulated DIs when 201
compared to the isometric condition. Prolonging the time interval between simulated DIs 202
significantly increased ASM stress. This effect was only seen in the contracted state, as 203
supported by a significant interaction between the contractile state and the time. Additionally, 204
the time-dependent increase in stress was more pronounced in isometric than in length 205
oscillating conditions, as supported by a significant interaction between the condition and the 206
time. 207
208
ASM elastance immediately before the simulated DIs is displayed in Figure 3. Prolonging the 209
interval between simulated DIs significantly increased ASM elastance. This effect was only seen 210
in the contracted state, as supported by a significant interaction between the contractile state 211
and the time. ASM stiffness before the simulated DIs was essentially the same (not shown), with 212
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
a significant effect for the contractile state (p < 0.0001), the time (p < 0.0001) and the interaction 213
(p < 0.0001). 214
215
ASM elastance during the simulated DIs is displayed in Figure 4. Unexpectedly, the contractile 216
state, the condition and the time had no overall effect. There was a trend for an interaction 217
between the contractile state and the condition, suggesting that MCh may increase elastance in 218
the condition of length oscillations. ASM stiffness during the simulated DIs was essentially the 219
same (not shown), with no overall effect for the contractile state, the condition and the time. 220
However, there was two exceptions. First, MCh slightly increased stiffness in the length 221
oscillating condition, as supported by a significant interaction between the contractile state and 222
the condition (p = 0.003). Second, prolonging the interval between simulated DIs increased 223
ASM stiffness but only in the length oscillating condition, as supported by a significant 224
interaction between the condition and the time (p = 0.03). 225
226
ASM stress immediately after the simulated DIs is displayed in Figure 5. Length oscillations 227
significantly affected the stress compared to the isometric condition. This effect was 228
predominant in the contracted state, as supported by a significant interaction between the 229
contractile state and the condition. More precisely, the stress generated by MCh was 230
substantially smaller in isometric than in length oscillating conditions. Prolonging the interval 231
between simulated DIs intricately affected the stress post-simulated DIs. In fact, a smile-shaped 232
curve emerged, meaning that the stress was settling at lower values when the elapsed times 233
between simulated DIs were of intermediate durations (5 and 10 min) as compared to when 234
they were of extreme durations (2 and 30 min). This was only seen in the contracted state, as 235
supported by a significant interaction between the contractile state and the time. Despite this 236
significant effect of time, the time-dependent increase observed pre-simulated DIs (Figure 2) 237
was to a great extent abolished by the simulated DIs in both isometric and length oscillating 238
conditions (Figure 5). 239
240
The bronchodilator effect of the simulated DIs (i.e., the decline in ASM stress caused by 241
simulated DIs) is displayed in Figure 6. The bronchodilator effect was significantly greater in 242
isometric than in length oscillating conditions. This effect was only seen in the contracted state, 243
as supported by a significant interaction between the contractile state and the condition. 244
Prolonging the interval between simulated DIs significantly increased the bronchodilator effect. 245
This effect was only seen in the contracted state, as supported by a significant interaction 246
between the contractile state and the time. In fact, the bronchodilator effect was virtually absent 247
with isoproterenol. The time-dependent increase in the bronchodilator effect of simulated DIs 248
was also more pronounced in isometric than in length oscillating conditions, as supported by a 249
significant interaction between the condition and the time. 250
251
252
253
Discussion 254
This study was specifically designed to investigate the effect of a continuous strain at a 255
magnitude simulating breathing at tidal volume on the bronchodilator effect of a simulated DI. 256
The results demonstrate that tidal strain attenuated the bronchodilator effect of simulated DIs 257
(Figure 6). In fact, while ASM stress post-simulated DIs with MCh was essentially similar to 258
ASM stress with isoproterenol in the isometric condition, ASM stress post-simulated DIs with 259
MCh was still substantially higher than with isoproterenol in the oscillating condition (Figure 5). 260
This means that while some of the acquired stress due to MCh was preserved post-simulated 261
DIs in the length oscillating condition, it completely vanished in the isometric condition. 262
Together, these results suggest that an activated ASM operating in a dynamic environment 263
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
becomes refractory to a larger strain and, therefore, exhibited a lower decline in contractility in 264
response to a simulated DI. 265
266
The contraction of ASM in response to a spasmogen is driven by a dynamic array of intracellular 267
molecular events. The downstream events not only include the myosin cross-bridge cycling that 268
provides the mechanical work for shortening, but also the formation of modular structures, such 269
as myosin filaments and adhesomes, that optimize contractility (36, 40, 50). The formation of 270
these modular structures requires the translocation of several proteins, their binding with 271
appropriate partners, the polymerization of actin and myosin, as well as an extensive 272
reorganization of the cell cytoskeleton to firmly link the contractile apparatus inside the cells to 273
the extracellular surroundings (36, 40, 49-51). Obviously, these post-translational events need 274
time to occur, so as the attainment of maximal contraction. 275
276
It is also important to understand that myosin filaments and adhesomes are sensitive to 277
mechanical strain. In fact, the regulatory effect of strain on ASM contraction is allegedly 278
attributed to the breakage of these modular structures (36, 50). A step change in ASM length, 279
for example, quickly reduces the content of myosin filaments, which occurs in conjunction with a 280
decrease in contractility (21, 36). However, if the muscle is stimulated to contract repeatedly at 281
the new length, both the content of myosin filaments and the contractility recover, sometimes 282
fully, over the course of approximately 30 min (21, 33, 36). This phenomenon was dubbed 283
‘length adaptation’ or ‘mechanical plasticity’ (6, 50). Therefore, as mentioned for a spasmogen 284
in the previous paragraph, the maximal contraction following a mechanical stimulus, such as a 285
change in length, takes several minutes to develop. To take into account the large influence of 286
time on ASM contraction, the present study examined the effect of prolonging the time interval 287
between simulated DIs (16, 28). 288
289
Consistent with the time needed for ASM to adapt to a new length, the stress generated in 290
response to a fix concentration of MCh slowly progressed over several minutes, both in 291
isometric and length oscillating conditions (Figure 2). This indicates that ASM undergoes some 292
sort of adaptations that slowly optimize its capacity to generate stress. Although this time-293
dependent increase in stress was greater in the isometric condition (Figure 2), which suggests a 294
beneficial effect of tidal strain on airway patency, the fact that these adaptations occurred in the 295
length oscillating condition is truly sensational. It implies that the molecular events driving these 296
adaptations can take place even when the length of ASM is oscillating. The time-dependent 297
increase in elastance and stiffness pre-simulated DIs (Figure 3) lends further support to the 298
notion that some sort of adaptations occurred in the oscillating condition and, as for length 299
adaptation, they may rely on the formation of modular structures that are reinforcing over time. 300
301
More surprising was the lack of overall effect of the contractile state (MCh versus isoproterenol) 302
on elastance and stiffness measured during the simulated DIs. This indicates that the elastance 303
and stiffness measured in the presence of MCh were the same as the ones measured in the 304
presence of isoproterenol (Figure 4). However, there was a significant interaction between the 305
contractile state and the condition for stiffness. This suggests that the effect of MCh on stiffness 306
during the simulated DIs depended on the condition in which ASM adapted pre-simulated DIs. 307
More precisely, the increased stiffness due to MCh observed pre-simulated DIs in the length 308
oscillation condition (Figure 3) was also perceived during a larger strain simulating a DI, which 309
was not the case in the isometric condition. A similar trend was observed for elastance. This 310
suggests that the adaptations acquired by activated ASM in dynamic conditions are different 311
from the ones acquired in static conditions. 312
313
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
Three aforementioned results also suggest that the adaptations acquired by activated ASM in 314
dynamic conditions are different from the ones acquired in static conditions: 1- The smaller time-315
dependent increase in stress pre-simulated DIs (Figure 2); 2- the preservation of part of the 316
stress acquired pre-simulated DIs after the simulated DIs (Figure 5); and 3- the smaller 317
bronchodilator effect of the simulated DIs (Figure 6). 318
319
The fact that the adaptations acquired in static versus dynamic conditions are different 320
represents the main finding of the present study. It implies that beyond the ability to reorganize 321
its cytoskeleton and contractile apparatus in order to optimize contractility, the modular 322
structures responsible for these adaptations can be built to operate over a certain level of strain. 323
Indeed, it seems that an ASM evolving in a dynamic environment is compelled to build different 324
structures, perhaps more elaborate or more flexible structures. These structures may require 325
more energy and take longer to form, as suggested by the slower time-dependent gain in ASM 326
stress in length oscillating versus isometric conditions (Figure 2). However, once they are 327
formed, these structures seem to stiffen ASM when measured over a larger stretch (not shown), 328
in addition to confer a certain refractoriness to the decline in contractility that is normally 329
afforded by such a stretch (Figure 5), resulting in an attenuated bronchodilator effect of the 330
simulated DIs (Figure 6). 331
332
It is certainly intriguing to note that different adaptations might take place in dynamic versus 333
static conditions. However, it is not very surprising. The proteins involved in the formation of 334
modular structures that are required to optimize ASM contraction, such as the adhesomes, are 335
also activated by strain (11, 41, 48). The mechanosensitivity of these proteins implies that they 336
should be differentially activated in static versus dynamic conditions, which may then differently 337
shape the various processes underlying these adaptations (50). 338
339
Alternative molecular mechanisms may explain our observations. Trepat and coworkers have 340
demonstrated an unifying relationship that describes the response to strain in a broad variety of 341
cells, including ASM cells (44). In particular, the extent of fluidization (viz., the magnitude by 342
which stiffness decreases) in response to a given strain, as well as the extent of the subsequent 343
re-solidification, seems to be dictated by how close the cell is from a solid-like state (or how far 344
is it from fluid-like state). According to these notions, an ASM operating in isometric conditions is 345
expected to solidify over time, and even more so when contracted. This phenomenon is 346
sometimes referred to as ‘physical ageing’ and is due to a slowly evolving network of non-347
specific physical forces (44). The facts that the stress pre-simulated DIs was slowly increasing 348
with time in our experiments and that it was almost significantly different between the isometric 349
and the length oscillation conditions (p = 0.07) are consistent with these notions. The implication 350
is that ASM exposed to tidal stain may not be adapting as mentioned above; i.e., by forming 351
more flexible structures that rely on highly coordinated sequences of signaling pathways. 352
Instead, tidal strain may simply retard ‘physical ageing’ and thereby keep ASM away for the 353
solid-like state, rendering it less susceptible to fluidization induced by a larger strain. At this 354
point, our study is observational. Consequently, we cannot determine the underlying molecular 355
mechanisms whereby ASM exposed to tidal strain develops refractoriness to the decline in 356
contractility induced by a larger strain. Studies investigating our observations at the molecular 357
scale are warranted. 358
359
In force-controlled experiments, a time-dependent increase is stiffness and elastance (Figure 3) 360
would translate into a progressively smaller strain (17). However, it is important to understand 361
that the opposite may occur in vivo. Indeed, previous experiments conducted in dogs in vivo 362
clearly predicted that the strain on the airway wall caused by swings in transpulmonary pressure 363
within the range of tidal breathing is greater on airways activated with MCh than on airways 364
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
relaxed with atropine (7). Other experimental and theoretical analyses have come to similar 365
conclusions; namely that an activated ASM is subjected to greater strain during tidal breathing 366
than a relaxed ASM (22). This phenomenon is mainly attributed to the non-linear stress-strain 367
curve of the airway wall. An airway wall with relaxed ASM tends to operate on a higher position 368
on its stress-strain curve and cannot be strained much further by the fluctuating stresses of 369
breathing. In contrast, the whole stress-strain curve is shifted to the right in an airway wall with 370
activated ASM. Assuming the same stress at end-tidal expiratory volume, this means that the 371
airway wall then operates on a lower position on its stress-strain curve and can thus be strained 372
substantially more by the same fluctuating stresses of breathing. Indirect in vivo evidences 373
actually suggest that ASM is strained more when activated. For example, different groups have 374
demonstrated that the tidal fluctuation of respiratory system resistance (Rrs) is near zero in non-375
asthmatic individuals (24, 34). The tidal fluctuation of Rrs is an indicator of the volume-376
dependence of resistance during tidal breathing. Assuming that there is no closure during 377
breathing at tidal volume, the tidal fluctuation of Rrs is also a proxy for the swings in airway 378
caliber and, therefore, of the strain on ASM during tidal breathing. In contrast to healthy 379
individuals, the tidal fluctuation of Rrs is quite large in asthmatic individuals (24, 34). The 380
volume-dependence of Rrs was actually shown to be the most predictive index of airflow 381
obstruction in wheezing toddlers (10). Additionally, the tidal fluctuation of Rrs is further 382
increased in both asthmatic and non-asthmatic individuals following an inhalational challenge 383
with MCh (24, 34). Altogether, these results suggest that an increase in ASM tone, caused 384
artificially by inhaling MCh or caused naturally in asthma, increases ASM strain during tidal 385
breathing. Testing the effect of length oscillations on the bronchodilator effect of simulated DIs 386
was thus relevant to further our understanding of the mechanisms underlying the impaired 387
bronchodilator effect of DIs in asthma (1, 4, 8, 13, 24, 34, 35, 37, 38). Our in vitro results 388
suggest that by adapting in the presence of tidal strain, the ASM becomes refractory to the 389
bronchodilator effect of DIs. They also offer a tentative explanation for the negative association 390
that was repeatedly reported in vivo between the tidal strain and the bronchodilator effect of a DI 391
(24, 34). 392
393
An important limitation of our study is that we applied a monotonous strain in between simulated 394
DIs. Recent studies have highlighted the importance of strain variability in modulating the signal 395
transduction that regulates the structure, microlocalization and function of organelles and 396
cytoskeletal elements involved in cell’s energetics (2). Based on these findings, it can be 397
predicted that a variable strain pattern would have significantly impact the time-dependent 398
observations we made in the present study, including the response to DIs. Definitely, the 399
incorporation of physiological levels of cycle-by-cycle variability in strain amplitude during 400
simulated tidal breathing should be considered in future studies, as it would emulate more 401
accurately real-life situations and thus generate results likely more relevant to in vivo physiology 402
(3, 42, 43). 403
404
Conclusions 405
The contraction of ASM in vivo is largely influenced by the concerted action of strain and the 406
great number of bronchoactive molecules found in the extracellular space. These stimuli can 407
acutely alter ASM contraction, but also instigate intracellular events that allow ASM to adapt 408
over longer time-scales in order to optimize its contractility. Herein, we demonstrate that these 409
adaptations were different when ASM is evolving in a dynamic environment compared to when it 410
is evolving in a static environment. More precisely, our results demonstrate that the adaptations 411
that occurred in length oscillating condition, although smaller in magnitude, underwent a smaller 412
decline in contractility in response to a simulated DI. This finding suggests that tidal strain may 413
attenuate the bronchodilator effect of DIs. We also believe that this finding is relevant to asthma 414
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
where, coincidentally, tidal strain is of greater magnitude and the bronchodilator effect of DIs is 415
impaired (24, 34). 416
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
References 417
418
1. Allen ND, Davis BE, and Cockcroft DW. Correlation between airway inflammation and 419
loss of deep-inhalation bronchoprotection in asthma. Ann Allergy Asthma Immunol 101: 413-420
418, 2008. 421
2. Bartolak-Suki E, Imsirovic J, Parameswaran H, Wellman TJ, Martinez N, Allen PG, 422
Frey U, and Suki B. Fluctuation-driven mechanotransduction regulates mitochondrial-network 423
structure and function. Nat Mater 14: 1049-1057, 2015. 424
3. Bartolak-Suki E, Noble PB, Bou Jawde S, Pillow JJ, and Suki B. Optimization of 425
Variable Ventilation for Physiology, Immune Response and Surfactant Enhancement in Preterm 426
Lambs. Frontiers in physiology 8: 425, 2017. 427
4. Black LD, Henderson AC, Atileh H, Israel E, Ingenito EP, and Lutchen KR. Relating 428
maximum airway dilation and subsequent reconstriction to reactivity in human lungs. J Appl 429
Physiol (1985) 96: 1808-1814, 2004. 430
5. Bosse Y. The strain on airway smooth muscle during a deep inspiration to total lung 431
capacity. J Eng Sci Med Diagn Ther 2: 1-21, 2019. 432
6. Bosse Y, Sobieszek A, Pare PD, and Seow CY. Length adaptation of airway smooth 433
muscle. Proc Am Thorac Soc 5: 62-67, 2008. 434
7. Brown RH, and Mitzner W. Effect of lung inflation and airway muscle tone on airway 435
diameter in vivo. J Appl Physiol 80: 1581-1588, 1996. 436
8. Burns CB, Taylor WR, and Ingram RH, Jr. Effects of deep inhalation in asthma: 437
relative airway and parenchymal hysteresis. J Appl Physiol 59: 1590-1596, 1985. 438
9. Chin LY, Bosse Y, Jiao Y, Solomon D, Hackett TL, Pare PD, and Seow CY. Human 439
airway smooth muscle is structurally and mechanically similar to that of other species. Eur 440
Respir J 36: 170-177, 2010. 441
10. Czovek D, Shackleton C, Hantos Z, Taylor K, Kumar A, Chacko A, Ware RS, Makan 442
G, Radics B, Gingl Z, and Sly PD. Tidal changes in respiratory resistance are sensitive 443
indicators of airway obstruction in children. Thorax 71: 907-915, 2016. 444
11. Desai LP, Wu Y, Tepper RS, and Gunst SJ. Mechanical stimuli and IL-13 interact at 445
integrin adhesion complexes to regulate expression of smooth muscle myosin heavy chain in 446
airway smooth muscle tissue. Am J Physiol Lung Cell Mol Physiol 301: L275-284, 2011. 447
12. Fairbank NJ, Connolly SC, Mackinnon JD, Wehry K, Deng L, and Maksym GN. 448
Airway smooth muscle cell tone amplifies contractile function in the presence of chronic cyclic 449
strain. Am J Physiol Lung Cell Mol Physiol 295: L479-488, 2008. 450
13. Fish JE, Ankin MG, Kelly JF, and Peterman VI. Regulation of bronchomotor tone by 451
lung inflation in asthmatic and nonasthmatic subjects. J Appl Physiol 50: 1079-1086, 1981. 452
14. Fredberg JJ, Bunk D, Ingenito E, and Shore SA. Tissue resistance and the contractile 453
state of lung parenchyma. J Appl Physiol (1985) 74: 1387-1397, 1993. 454
15. Fredberg JJ, Inouye D, Miller B, Nathan M, Jafari S, Raboudi SH, Butler JP, and 455
Shore SA. Airway smooth muscle, tidal stretches, and dynamically determined contractile 456
states. Am J Respir Crit Care Med 156: 1752-1759, 1997. 457
16. Gazzola M, Khadangi F, Clisson M, Beaudoin J, Clavel MA, and Bosse Y. The 458
elapsed time between simulated deep inspirations on the mechanics of airway smooth muscle 459
in ovine tracheal strips. In: American Thoracic Society. Dallas: Am J Respir Crit Care Med, 460
2019, p. A2161. 461
17. Gazzola M, Khadangi F, Clisson M, Beaudoin J, Clavel MA, and Bosse Y. The 462
shortening of airway smooth muscle is modulated by prolonging the time without simulated deep 463
inspirations in ovine tracheal strips. J Appl Physiol (1985) 2019. 464
18. Gunst SJ. Contractile force of canine airway smooth muscle during cyclical length 465
changes. J Appl Physiol 55: 759-769, 1983. 466
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
19. Harvey BC, Parameswaran H, and Lutchen KR. Can breathing-like pressure 467
oscillations reverse or prevent narrowing of small intact airways? J Appl Physiol (1985) 119: 47-468
54, 2015. 469
20. Harvey BC, Parameswaran H, and Lutchen KR. Can tidal breathing with deep 470
inspirations of intact airways create sustained bronchoprotection or bronchodilation? J Appl 471
Physiol (1985) 115: 436-445, 2013. 472
21. Herrera AM, McParland BE, Bienkowska A, Tait R, Pare PD, and Seow CY. 473
'Sarcomeres' of smooth muscle: functional characteristics and ultrastructural evidence. J Cell 474
Sci 118: 2381-2392, 2005. 475
22. Hiorns JE, Jensen OE, and Brook BS. Nonlinear compliance modulates dynamic 476
bronchoconstriction in a multiscale airway model. Biophys J 107: 3030-3042, 2014. 477
23. Ijpma G, Kachmar L, Matusovsky OS, Bates JH, Benedetti A, Martin JG, and 478
Lauzon AM. Human trachealis and main bronchi smooth muscle are normoresponsive in 479
asthma. Am J Respir Crit Care Med 191: 884-893, 2015. 480
24. Jensen A, Atileh H, Suki B, Ingenito EP, and Lutchen KR. Selected contribution: 481
airway caliber in healthy and asthmatic subjects: effects of bronchial challenge and deep 482
inspirations. J Appl Physiol 91: 506-515; discussion 504-505, 2001. 483
25. LaPrad AS, Szabo TL, Suki B, and Lutchen KR. Tidal stretches do not modulate 484
responsiveness of intact airways in vitro. J Appl Physiol 109: 295-304, 2010. 485
26. Lavoie TL, Krishnan R, Siegel HR, Maston ED, Fredberg JJ, Solway J, and Dowell 486
ML. Dilatation of the constricted human airway by tidal expansion of lung parenchyma. Am J 487
Respir Crit Care Med 186: 225-232, 2012. 488
27. Lee-Gosselin A, Pascoe CD, Couture C, Pare PD, and Bosse Y. Does the length 489
dependency of airway smooth muscle force contribute to airway hyperresponsiveness? J Appl 490
Physiol (1985) 115: 1304-1315, 2013. 491
28. Mailhot-Larouche S, and Bosse Y. Interval between simulated deep inspirations on the 492
dynamics of airway smooth muscle contraction in guinea pig bronchi. Respir Physiol Neurobiol 493
259: 136-142, 2019. 494
29. Mailhot-Larouche S, Lortie K, Marsolais D, Flamand N, and Bosse Y. An in vitro 495
study examining the duration between deep inspirations on the rate of renarrowing. Respir 496
Physiol Neurobiol 243: 13-19, 2017. 497
30. Noble PB, Jones RL, Needi ET, Cairncross A, Mitchell HW, James AL, and McFawn 498
PK. Responsiveness of the human airway in vitro during deep inspiration and tidal oscillation. J 499
Appl Physiol 110: 1510-1518, 2011. 500
31. Noble PB, McFawn PK, and Mitchell HW. Responsiveness of the isolated airway 501
during simulated deep inspirations: effect of airway smooth muscle stiffness and strain. J Appl 502
Physiol 103: 787-795, 2007. 503
32. Pascoe CD, Seow CY, Pare PD, and Bosse Y. Decrease of airway smooth muscle 504
contractility induced by simulated breathing maneuvers is not simply proportional to strain. J 505
Appl Physiol 114: 335-343, 2013. 506
33. Pratusevich VR, Seow CY, and Ford LE. Plasticity in canine airway smooth muscle. J 507
Gen Physiol 105: 73-94, 1995. 508
34. Salome CM, Thorpe CW, Diba C, Brown NJ, Berend N, and King GG. Airway re-509
narrowing following deep inspiration in asthmatic and nonasthmatic subjects. Eur Respir J 22: 510
62-68, 2003. 511
35. Scichilone N, Permutt S, and Togias A. The lack of the bronchoprotective and not the 512
bronchodilatory ability of deep inspiration is associated with airway hyperresponsiveness. Am J 513
Respir Crit Care Med 163: 413-419, 2001. 514
36. Seow CY. Myosin filament assembly in an ever-changing myofilament lattice of smooth 515
muscle. Am J Physiol Cell Physiol 289: C1363-1368, 2005. 516
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
37. Skloot G, Permutt S, and Togias A. Airway hyperresponsiveness in asthma: a problem 517
of limited smooth muscle relaxation with inspiration. J Clin Invest 96: 2393-2403, 1995. 518
38. Slats AM, Janssen K, van Schadewijk A, van der Plas DT, Schot R, van den 519
Aardweg JG, de Jongste JC, Hiemstra PS, Mauad T, Rabe KF, and Sterk PJ. Bronchial 520
inflammation and airway responses to deep inspiration in asthma and chronic obstructive 521
pulmonary disease. Am J Respir Crit Care Med 176: 121-128, 2007. 522
39. Stephens NL, Kroeger E, and Mehta JA. Force-velocity characteristics of respiratory 523
airway smooth muscle. J Appl Physiol 26: 685-692, 1969. 524
40. Tang DD. The Dynamic Actin Cytoskeleton in Smooth Muscle. Adv Pharmacol 81: 1-38, 525
2018. 526
41. Tang DD, and Gunst SJ. Selected contribution: roles of focal adhesion kinase and 527
paxillin in the mechanosensitive regulation of myosin phosphorylation in smooth muscle. J Appl 528
Physiol (1985) 91: 1452-1459, 2001. 529
42. Thammanomai A, Hamakawa H, Bartolak-Suki E, and Suki B. Combined effects of 530
ventilation mode and positive end-expiratory pressure on mechanics, gas exchange and the 531
epithelium in mice with acute lung injury. PLoS One 8: e53934, 2013. 532
43. Thammanomai A, Hueser LE, Majumdar A, Bartolak-Suki E, and Suki B. Design of a 533
new variable-ventilation method optimized for lung recruitment in mice. J Appl Physiol (1985) 534
104: 1329-1340, 2008. 535
44. Trepat X, Deng L, An SS, Navajas D, Tschumperlin DJ, Gerthoffer WT, Butler JP, 536
and Fredberg JJ. Universal physical responses to stretch in the living cell. Nature 447: 592-537
595, 2007. 538
45. Wang L, Chitano P, and Murphy TM. Length oscillation induces force potentiation in 539
infant guinea pig airway smooth muscle. Am J Physiol Lung Cell Mol Physiol 289: L909-915, 540
2005. 541
46. Wang L, Pare PD, and Seow CY. Effects of length oscillation on the subsequent force 542
development in swine tracheal smooth muscle. J Appl Physiol 88: 2246-2250, 2000. 543
47. Wong WD, Wang L, Pare PD, and Seow CY. Bronchodilatory effect of deep inspiration 544
in freshly isolated sheep lungs. Am J Physiol Lung Cell Mol Physiol 312: L178-L185, 2017. 545
48. Wu Y, Huang Y, and Gunst SJ. Focal adhesion kinase (FAK) and mechanical 546
stimulation negatively regulate the transition of airway smooth muscle tissues to a synthetic 547
phenotype. Am J Physiol Lung Cell Mol Physiol 311: L893-L902, 2016. 548
49. Zhang W, Bhetwal BP, and Gunst SJ. Rho kinase collaborates with p21-activated 549
kinase to regulate actin polymerization and contraction in airway smooth muscle. J Physiol 596: 550
3617-3635, 2018. 551
50. Zhang W, and Gunst SJ. Molecular mechanisms for the mechanical modulation of 552
airway responsiveness. J Eng Sci Med Diagn Ther 2: 1-8, 2019. 553
51. Zhang W, and Gunst SJ. Non-muscle (NM) myosin heavy chain phosphorylation 554
regulates the formation of NM myosin filaments, adhesome assembly and smooth muscle 555
contraction. J Physiol 595: 4279-4300, 2017. 556
557
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
Figure legends 558
Figure 1. Representative traces of length (A) and the corresponding stress (B) are depicted for 559
a strip maintained in a contracted state. In each protocol, the strip was subjected to a series of 560
four simulated deep inspirations intercalated by 2, 5, 10 and 30 min in a randomized order. In 561
this example, the strip was also subjected to tidal strain in between simulated deep inspirations, 562
which consisted of small length oscillations simulating tidal breathing. The oscillations cannot be 563
seen at the time-scale shown and appear as a thick line. A portion of the trace in B, which is 564
outlined 565
566
Figure 2. ASM stress immediately before the simulated deep inspirations. The values on the x-567
axis are the times elapsed since the previous simulated deep inspirations. In between simulated 568
deep inspirations, the length was either held fixed (isometric on the left side) or oscillated to 569
simulate tidal breathing (length oscillations on the right side). In each condition, the ASM was 570
either contracted with methacholine (MCh) at 10-5 M (solid symbols) or relaxed with 571
isoproterenol at 10-6 M (emptied symbols). The results of the three-way ANOVA are presented 572
in the table. n = 10 from 5 sheep 573
574
Figure 3. ASM elastance immediately before the simulated deep inspirations. See Figure 2 575
legend for further description. The results of the two-way ANOVA are presented in the table. n 576
= 10 from 5 sheep 577
578
Figure 4. ASM elastance during the simulated deep inspirations. See Figure 2 legend for further 579
description. The results of the three-way ANOVA are presented in the table. n = 10 from 5 580
sheep 581
582
Figure 5. ASM stress immediately after the simulated deep inspirations. See Figure 2 legend for 583
further description. The results of the three-way ANOVA are presented in the table. n = 10 from 584
5 sheep 585
586
Figure 6. The bronchodilator effect of simulated deep inspirations. The bronchodilator effect 587
was quantified by subtracting the stress pre-simulated deep inspiration (solid circles) by the 588
stress post-simulated deep inspiration (solid squares). The length of the gray double-ended 589
arrows thus represents the magnitude of the bronchodilator effect. Only the results of ASM 590
contracted with methacholine (MCh) at 10-5 M are shown. The results of ASM relaxed with 591
isoproterenol at 10-6 M are not shown, as there was almost no stress and therefore almost no 592
bronchodilator effect. See Figure 2 legend for further description. The results of the three-way 593
ANOVA are presented in the table. n = 10 from 5 sheep 594
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
Length trace
Time (s)
Length (mm)
5.5
6.0
6.5
7.0
7.5
8.0
0 500 1000 1500 2000 2500 3000
5 min 10 min 30 min 2 min
Length oscillations simulating tidal breathing
Stretch simulating a deep inspiration
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
Stress trace
Time (s)
Stress (mN/mm2)
0
50
100
150
200
250
0 500 1000 1500 2000 2500 3000
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
Zoomed stress trace
Time (s)
Stress (mN/mm2)
0
50
100
150
200
250
2840 2860 2880 2900 2920 2940 2960
●
●
●
●
●●●●●●●●●●●●●●●●●●●●●●●●
●
●●●●●●●●●●●●●●●●●●●● ● ●●
Length
(
mm
)
Corresponding length trace
5.5
6.0
6.5
7.0
7.5
8.0
2840 2860 2880 2900 2920 2940 2960
●
●
●
●
●●●●●●●●●●● ● ●●● ● ●●●● ● ● ● ●
●●●●● ● ●●●● ● ●● ● ● ● ● ●●●●●●●
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
2 5 10 30 2 5 10 30
0
30
60
90
120
Time (min)
Stress (mN/mm2)
Stress pre-DI
Isometric Length Oscillations
MCh
Isoproterenol
Contractile state <0.0001
Condition 0.07
Time <0.0001
Contractile state x Condition 0.09
Contractile state x Time <0.0001
Condition x Time 0.0003
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
2
5
10
30
2
5
10
30
0
20
40
60
80
100
120
Time (min)
Elastance (mN / mm3)
Elastance pre-DI
MCh
Isoproterenol
Contractile state <0.0001
Time <0.0001
Contractile state x Time <0.0001
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
2 5 10 30 2 5 10 30
0
25
50
75
100
125
Time (min)
Elastance (mN/mm3)
Elastance during DI
Isometric Length Oscillations
MCh
Isoproterenol
Contractile state 0.66
Condition 0.13
Time 0.95
Contractile state x Condition 0.06
Contractile state x Time 0.36
Condition x Time 0.21
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
2 5 10 30 2 5 10 30
0
5
10
15
20
25
Time (min)
Stress (mN/mm2)
Stress post-DI
Isometric Length Oscillations
MCh
Isoproterenol
Contractile state 0.0004
Condition <0.0001
Time 0.005
Contractile state x Condition 0.0002
Contractile state x Time 0.02
Condition x Time 0.41
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.
2 5 10 30 2 5 10 30
0
30
60
90
120
Time (min)
Stress (mN/mm2)
Bronchodilator Effect
Isometric Length Oscillations
MCh Pre-DI
MCh Post-DI
Contractile state 0.0001
Condition 0.003
Time <0.0001
Contractile state x Condition 0.007
Contractile state x Time <0.0001
Condition x Time 0.0001
Downloaded from www.physiology.org/journal/ajplung at Univ of Nebraska Lincoln (129.093.016.003) on January 10, 2020.