Porcine tracheas from both female and male pigs were obtained from a local abattoir, with exemption from Institutional Animal Care and Use Committee (IACUC) approval. During transport to the lab, tracheas were kept in ice-cold physiologic saline solution (composition in mM: 118.99 NaCl, 1.17 MgSO₄, 1.18 KH₂PO₄, 4.7 KCl, 2.5 CaCl₂, 0.03 EDTA, 5.5 dextrose, 25 HEPES, and pH 7.4). The ASM layer from each trachea was dissected under a binocular microscope, and eight strips (∼0.8 × 4 mM) were cut as previously described (Fredberg et al., 1996; Sieck et al., 1998, 2001, 2019; Jones et al., 1999a; Dogan et al., 2017). The ASM strips were separated into two groups: one group exposed to TNFα (100 ng/ml; Cell Sciences, MA, United States) for 24 h at room temperature and a second group exposed to physiological saline for the same period (control). Four of the ASM strips (two TNFα treated and two controls) were used for duplicate measurements of simultaneous [Ca²⁺]_cyt and force responses to ACh stimulation (EC₅₀; 2.6 μM for control and 1.3 μM for TNFα). The second set of four ASM strips (two TNFα treated and two controls) were used to examine ACh-induced changes in rMLC₂₀ phosphorylation.

To measure [Ca²⁺]_cyt, the ASM strips were incubated for 4 h (at 25°C) in physiological saline solution containing the acetoxymethyl ester of fura-2 (fura-2 AM; 5 μM) dissolved in dimethyl sulfoxide (DMSO) as well as 0.02% Pluronic (F-127; Molecular Probes) to facilitate efficient cytosolic loading of fura-2 AM. The ASM strips were then mounted using stainless steel microforceps between a micropositioner (Mitutoyo America Corp., Aurora, IL, United States) to adjust muscle length and an isometric force transducer (KG4; Scientific Instruments, GmbH, Heidelberg, Germany) in a 0.2-ml quartz cuvette (Guth Muscle Research System; Scientific Instruments GmbH, Heidelberg, Germany) (Guth and Wojciechowski, 1986; Sieck et al., 2019) that was continuously perfused with physiological saline solution aerated with 95% O₂ and 5% CO₂. The ASM strips were perfused for 30 min to wash out any excess fura-2 AM. The length of the ASM strip was then adjusted over a 1-h period to achieve a sustained preload force of 10 mN.

Fura-2 fluorescence ion the ASM strip was excited using a mercury high-pressure lamp (75 W) as a light source at alternating (every 2 ms) wavelengths of 340 and 380 nm restricted by using a rotating filter wheel as previously described (Lorenz et al., 1999; Han et al., 2010). Emitted fluorescence was detected at 510 nm (restricted using a bandpass filter) using a photomultiplier tube. The ratio of emitted fluorescence at 340 and 380 nm excitation wavelengths (F₃₄₀/F₃₈₀) was used to determine [Ca²⁺]_cyt based on a calibration equation described by Grynkiewicz et al. (1985); Todoroki-Ikeda et al. (2000), Han et al. (2010), and Sieck et al., 2019. Duplicate measurements of isometric [Ca²⁺]_cyt and force responses of ASM strips to ACh concentration (EC₅₀) were digitized (1 kHz sampling rate) using LabChart8 (AD Instruments, Colorado Springs, CO, United States). The duplicate measurements were subsequently averaged for each trachea.

As previously described (Sieck et al., 2019), rMLC₂₀ phosphorylation was detected using a Phos-tag^TM sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gel with Zn²⁺ (Takeya et al., 2008; Walsh et al., 2011; Sieck et al., 2019; Han et al., 2021) (Wako Chemicals Inc., Richmond, VA, United States) and standard western blotting. In each of the TNFα treated and control groups, one ASM strip was not exposed to ACh and was thus used to determine the baseline level of rMLC₂₀ phosphorylation. The second ASM strip in each group was stimulated with ACh concentration (EC₅₀) for 30 s, a time period previously found to induce maximum rMLC₂₀ phosphorylation (Parkman et al., 2001). The ASM strips were then flash-frozen in 10% trichloroacetic acid/10 mM dithiothreitol in pre-chilled acetone. Electrophoresis of the Phos-tag^TM gels was performed at 20 mA for 1 h 50 min in running buffer (pH 7.4; 100 mM Tris-base, 100 mM 3-morpholinopropane-1-sulfonic acid, 0.1% SDS, 5 mM Sodium Bisulfate) at the room temperature. The gel was immersed in 25 mM Tris, 192 mM Glycine, 10% methanol (v/v) containing 10 mM EDTA to remove Zn²⁺ for 30 min and proteins were transferred to polyvinylidene difluoride membrane, which was fixed with 0.5% formaldehyde in PBS for 45 min (Takeya et al., 2008) and blocked with 5% dry milk in TBST. Using a standard western blotting technique, the membrane was incubated with primary (1:2000 dilution, rabbit polyclonal anti-rMLC₂₀; sc-15370; Santa Cruz, CA, United States) and secondary antibody (1:10,000 dilution, goat anti-rabbit IgG-HRP conjugate; Santa Cruz, CA, United States). Unphosphorylated rMLC₂₀ and phosphorylated rMLC₂₀ (p-rMLC₂₀) were detected by enhanced chemiluminescence (SuperSignal West Dura Extended Duration Substrate; Thermo Fisher Scientific, Rockford, IL, United States) and imaged on ChemiDoc MP Image System (Bio-Rad Laboratories, Hercules, CA, United States) and analyzed using an Image-Lab-Software (version 6.0.1, Bio-Rad Laboratories, Hercules, CA, United States) (Figure 5C).

Four ASM strips were used from each animal (n = 6) for the analysis of [Ca²⁺]_cyt and force responses induced by ACh stimulation with (two strips) and without (two strips) 24-h exposure to TNFα. The two independent measurements of [Ca²⁺]_cyt and force responses to ACh were then averaged to provide a single measurement per treatment per animal. In a separate set of four ASM strips from each animal (n = 6), baseline (no ACh stimulation) and ACh-induced changes in rMLC₂₀ phosphorylation were analyzed with (two strips – one baseline and one ACh stimulated) and without (two strips – one baseline and one ACh stimulated) 24-h exposure to TNFα. Results were analyzed using a Student t-test (JMP Pro software; JMP, RRID: SCR_014242). Data are summarized as means ± SD, with significance established at p < 0.05.

In each of six animals (n = 6), the duplicate measurements of ASM [Ca²⁺]_cyt and force responses to ACh stimulation were comparable (<10% variance between the two measurements) and averaged to obtain a single value for summary analysis. In each ASM strip, ACh stimulation induced a biphasic [Ca²⁺]_cyt response reaching an initial peak by ∼5 s and then declining to a sustained plateau after ∼2–3 min. The isometric force response to ACh stimulation was delayed and slower with maximum specific force reached after ∼1–2 min. Compared to control (untreated) ASM strips, 24-h exposure to TNFα (100 ng/ml) increased the peak force response to ACh by ∼18% compared to control (Figures 2A,C; p < 0.05), whereas the peak [Ca²⁺]_cyt response to ACh was unaffected by TNFα (Figures 2A,B). The ratio of maximum specific force to maximum [Ca²⁺]_cyt increased by ∼18% in ASM strips exposed to 24-h TNFα (p < 0.01; Figure 2D). These results indicate that static Ca²⁺ sensitivity of force generation in ASM is increased by TNFα.

In each ASM strip, [Ca²⁺]_cyt and isometric force responses to ACh stimulation were simultaneously recorded (Figure 3A). The ACh-induced elevation of [Ca²⁺]_cyt always preceded the force response by ∼3–5 s (Figure 3A). The fura-2 fluorescence signal of [Ca²⁺]_cyt was noisier affecting the signal to noise level and detection of the onset of the ACh-induced response. Yet, the signal to noise ratio of the [Ca²⁺]_cyt responses was less than 10%, which was sufficient to determine the onset of the [Ca²⁺]_cyt response within 10 ms with a 1 kHz sampling rate. The noise level of the mechanical force signal was much lower, and thus did not affect the detection of the onset of the ACh force response. The temporal delay between the onset of ACh-induced [Ca²⁺]_cyt and force responses was ∼4–5 s corresponding to the time to reach a peak [Ca²⁺]_cyt response. Importantly, compared to control (untreated) ASM strips, 24-h TNFα exposure had no impact on the temporal delay between the onset of ACh-induced [Ca²⁺]_cyt and force responses (Figure 3B).

The temporal characteristics of the ACh-induced [Ca²⁺]_cyt and force responses in ASM strips were determined by calculating the first derivatives of [Ca²⁺]_cyt and force responses (i.e., d[Ca²⁺]_cyt/dt and dForce/dt) (Figure 4A). The maximum ACh-induced d[Ca²⁺]_cyt/dt was ∼40 nM [Ca²⁺]_cyt s^–1. Compared to control (untreated) ASM strips, 24-h TNFα exposure had no significant effect on the rate of rise in ACh-induced [Ca²⁺]_cyt response (Figure 4B). The maximum ACh-induced dForce/dt was much slower at ∼0.6 N cm^–2 s^–1. Although peak force was higher in TNFα treated ASM strips, the maximum dForce/dt was not different than untreated controls (Figure 4B).

The dynamic coupling of ACh-induced [Ca²⁺]_cyt and force responses in ASM strips was evaluated using phase-loop plots (Figure 5A). The ascending limb of the phase loop plot showed that little force developed as [Ca²⁺]_cyt increased to a peak value that was similar between 24-h TNFα treated and untreated control ASM strips. The rising rate of the phase-loop plots of ACh-induced [Ca²⁺]_cyt and force responses was similar between TNFα and control ASM strips (Figure 5B). Subsequently force increased as [Ca²⁺]_cyt decreased. Importantly, the ACh-induced force response in TNFα treated ASM strips was greater compared to untreated control ASM strips. The relaxation rate of the phase-loop plots of ACh-induced [Ca²⁺]_cyt and force responses was also similar between TNFα and control ASM strips (Figure 5C). To provide an overall assessment of the dynamic coupling of ACh-induced [Ca²⁺]_cyt and force responses, the integral of the phase-loop plots was calculated. Compared to control (untreated) ASM strips, TNFα significantly increased the integral of the phase-loop plot (p < 0.05, Figure 5D).

In a separate set of ASM strips, the extent of rMLC₂₀ phosphorylation (ratio of p-rMLC₂₀ to total rMLC₂₀) induced by 30 s ACh stimulation was examined in ASM strips using western blot analysis (Figure 6). The 30 s time period for ACh stimulation was selected as we have previously shown that the maximum extent of rMLC₂₀ phosphorylation is reached by this time (Dogan et al., 2017). ACh stimulation significantly increased the extent of rMLC₂₀ phosphorylation in both 24-h TNFα treated and untreated control ASM strips (Figures 6B,C), compared to the baseline (no ACh stimulation). Compared to control ASM strips, 24-h TNFα exposure had no significant effect on the extent of ACh-induced rMLC₂₀ phosphorylation (Figures 6B,C), although the ACh-induced force generated by TNFα treated ASM strips after 30 s was significantly greater than that of control ASM strips (p < 0.05; Figure 6A). Thus, the increase in ACh-induced ASM force following TNFα exposure was unrelated to the extent of rMLC₂₀ phosphorylation. These results are consistent with previous studies showing that TNFα increases ASM force generation due to increased contractile protein content (greater number of contractile units) and enhanced cytoskeletal remodeling (actin polymerization) resulting in increased tethering of contractile elements to the plasma membrane and force translation to the extracellular matrix (ECM) rather than changes in rMLC₂₀ phosphorylation (Jones et al., 1999a; Zhang et al., 2015; Dogan et al., 2017; Sieck et al., 2019).

It is commonly thought that an increase in Ca²⁺ sensitivity of force generation in ASM is mediated by increasing the extent of rMLC₂₀ phosphorylation via agonist-induced activation of rho kinase (ROCK)-dependent inhibition of myosin light chain phosphatase (MLCP) (de Lanerolle and Paul, 1991; Jones et al., 1994a, 1999a; Somlyo and Somlyo, 1994; Karaki et al., 1997; Sieck et al., 1998; Todoroki-Ikeda et al., 2000; Sakamoto et al., 2003; Sakai et al., 2017; Alvarez-Santos et al., 2020). However, we and others previously reported that the extent of rMLC₂₀ phosphorylation induced by ACh stimulation reaches a maximum after ∼30 s or less and does not change thereafter (Jones et al., 1999a; Mehta and Gunst, 1999; Zhang et al., 2015; Dogan et al., 2017; Sieck et al., 2019; Seow and An, 2020; Han et al., 2021). This timeframe does not match the time for force generation in ASM, which reaches a maximum only after 60–120 s. Moreover, the extent of rMLC₂₀ phosphorylation does not change with increasing ACh concentration, even though force increases (Dogan et al., 2017; Sieck et al., 2019; Han et al., 2021). Thus, while [Ca²⁺]_cyt-dependent rMLC₂₀ phosphorylation does mediate cross-bridge recruitment and the onset of a force response in ASM, it does not appear to regulate the longer-term time course or amplitude of the force response. In the present study, even at the lower EC₅₀ concentration of ACh (1.3 vs 2.6 μM in TNFα treated vs untreated control, respectively), the force response in TNFα treated ASM was greater while the extent of rMLC₂₀ phosphorylation was unaffected. In agreement, there was also no effect of TNFα exposure on the temporal delay between the onset of [Ca²⁺]_cyt and force responses or on their rates of rise after ACh stimulation. These results indicate that neither the extent nor the rate of rMLC₂₀ phosphorylation in ASM is affected by TNFα exposure. Thus, we conclude that ACh-dependent rMLC₂₀ phosphorylation is not the primary mechanism through which Ca²⁺ sensitivity of force generation increases after 24-h TNFα exposure.

In ASM, muscarinic stimulation induces a biphasic elevation of [Ca²⁺]_cyt with a peak level reach in ∼5 s followed by a slow decline to steady state after ∼120–180 s. The initial elevation of [Ca²⁺]_cyt in response to ACh stimulation is due to several sources of Ca²⁺ including inositol triphosphate receptor (IP₃R) and ryanodine receptor (RyR) mediated sarcoplasmic reticulum (SR) Ca²⁺ release as well as extracellular Ca²⁺ influx via L-type, voltage-gated Ca²⁺ channels (Takuwa et al., 1987; Jones et al., 1994b). Both the IP₃R and RyR channels are Ca²⁺ sensitive resulting in Ca²⁺-induced SR Ca²⁺ release (CICR) and a more rapid accelerating elevation of [Ca²⁺]_cyt (Kannan et al., 1997; Sieck et al., 1997; Prakash et al., 2000; Pabelick et al., 2001; Bai et al., 2009). ACh stimulation activates phospholipase C and IP₃ production, and IP₃ mediates the opening of IP₃R channels in the SR. In contrast, there is no second messenger agonist to mediate the opening of RyR channels in the SR. Instead, with the initial elevation of [Ca²⁺]_cyt due to extracellular Ca²⁺ influx or IP₃R mediated SR Ca²⁺ release, CICR mediates the opening of RyR channels in ASM. However, ACh-induced production of cyclic ADP ribose increases the sensitivity of RyR to CICR (Kannan et al., 1996; Hsuan et al., 1999; White et al., 2000; Deshpande et al., 2004). In porcine ASM, we previously showed that ACh stimulation induces localized constant amplitude [Ca²⁺]_cyt oscillations that propagate through the cell (Prakash et al., 1997; Pabelick et al., 1999; Sieck et al., 2001). The global elevation of [Ca²⁺]_cyt in response to muscarinic stimulation reflects the spatial-temporal integration of these propagating [Ca²⁺]_cyt oscillations (Prakash et al., 1997; Pabelick et al., 1999; Sieck et al., 2001). While the amplitude of localized [Ca²⁺]_cyt oscillations is constant, both [Ca²⁺]_cyt oscillation frequency and propagation velocity are increased by increasing ACh concentration. The time course of the spatial/temporal summation of ACh-induced [Ca²⁺]_cyt oscillations in ASM cells is ∼700 ms, roughly matching the short-term delay in rMLC₂₀ phosphorylation and cross-bridge recruitment (Figure 1; Prakash et al., 1997; Pabelick et al., 1999; Sieck et al., 2001). Measurements of the time delay between the onset of [Ca²⁺]_cyt and force responses as well as the first derivatives of these responses (i.e., d[Ca²⁺]_cyt/dt and dForce/dt) reflect the short-term dynamic coupling mediated by rMLC₂₀ phosphorylation and cross-bridge recruitment (Figures 3A,B). Importantly, 24-h TNFα exposure did not affect these measures of short-term coupling of ACh-induced [Ca²⁺]_cyt and force responses.

Unlike the [Ca²⁺]_cyt response, the ACh-induced force response in ASM is not biphasic but develops at a much slower rate reaching a plateau in force only after 60–120 s. As shown in Figure 1, short-term force generation in ASM depends on [Ca²⁺]_cyt/calmodulin-mediated activation of MLCK and rMLC₂₀ phosphorylation and cross-bridge recruitment. The phosphorylation of rMLC₂₀ is also affected by MLCP and dephosphorylation; however, we and others have shown that the maximum level of ACh-induced rMLC₂₀ phosphorylation in ASM is reached and sustained after ∼30 s or less, well before the plateau in ACh-induced force (Washabau et al., 1991; Dogan et al., 2017; Sieck et al., 2019; Han et al., 2021). Importantly, ACh-induced force generation in ASM increases by the addition of de novo contractile units due to [Ca²⁺]_cyt-dependent polymerization of actin and myosin, and to the tethering of filamentous actin to the cortical cytoskeleton for force translation to the ECM (Figure 1; Gunst and Tang, 2000; Sieck et al., 2001; Murphy and Rembold, 2005; Ijpma et al., 2010, 2011; Sieck and Gransee, 2012; Delmotte et al., 2020; Seow and An, 2020; Wang et al., 2020). It can be assumed that by ∼30 s after ACh stimulation, when the maximum extent of rMLC₂₀ phosphorylation is achieved, the recruitment of available cross-bridges does not account for additional force generation. Thus, from 30 to ∼120 s after ACh stimulation when ASM force plateaus, the addition of de novo contractile units or the tethering of actin filaments to the cortical cytoskeleton are the major contributors to force generation.

The cycling of cross-bridges in ASM is associated with ATP hydrolysis, and is dependent on load (Jones et al., 1999a; Delmotte et al., 2020). The tethering of actin filaments to the cortical cytoskeleton via dense bodies allows ASM force transmission to the ECM (Jones et al., 1999b; Mehta and Gunst, 1999; Gunst and Fredberg, 2003; Gunst et al., 2003; Gunst and Zhang, 2008; Zhang and Gunst, 2008; Chitano et al., 2017; Dogan et al., 2017; Tang, 2018; Seow and An, 2020; Wang et al., 2020). At the same time, the tethering of contractile units increases internal loading and slows cross-bridge cycling rate (Dogan et al., 2017; Sieck et al., 2019; Han et al., 2021). In permeabilized ASM strips maximally activated by Ca²⁺ (pCa 4.0), we simultaneously measured force and ATP hydrolysis rate using an NADH-linked fluorescence technique (Delmotte et al., 2020). The maximal increase in ATP hydrolysis rate occurred during the initial 30–60 s with the rMLC₂₀ phosphorylation-mediated recruitment of cross-bridges and the de novo addition of contractile units. Thereafter, the increase in ATP hydrolysis rate slowed reflecting the tethering of contractile units to the cortical cytoskeleton and internal loading with peak ATP hydrolysis rate reached after ∼120 s (Delmotte et al., 2020). Thereafter, ATP hydrolysis rate decreased with further tethering of contractile units and increased internal loading until a plateau was reached after ∼300–400 s.

The polymerization of monomeric globular (G) actin to filamentous (F) actin is Ca²⁺ dependent and increases with ACh stimulation (Hirshman and Emala, 1999; Jones et al., 1999b; Mehta and Gunst, 1999; Sieck et al., 2019). Zhang et al. reported that ACh-induced actin polymerization in ASM is regulated by RhoA activation, which enhances the assembly of the adhesome within the cortical cytoskeleton (Zhang et al., 2012; Seow and An, 2020). In support, we found that the extent of actin polymerization and tethering to the cortical cytoskeleton increases with time and with ACh concentration (Brandenburg et al., 2019). Furthermore, inhibiting actin polymerization greatly reduces ASM force generation (Jones et al., 1999b; Mehta and Gunst, 1999; Sieck et al., 2019; Delmotte et al., 2020; Han et al., 2021), reflecting the disruption of both the contribution of additional contractile units and force translation due to the tethering of contractile elements to the cortical cytoskeleton.

In the heart, phase-loop plots of [Ca²⁺]_cyt and force responses evoked by electrical stimulation have been used to assess Ca²⁺ sensitivity of cardiomyocytes based on the [Ca²⁺]_cyt at which 50% relaxation occurs (Schaible et al., 2016; Tveita et al., 2019). This assessment reflects the off-rate of Ca²⁺ binding to troponin-C (TnC) and thin filament regulation of cross-bridge attachment/detachment in cardiomyocytes. In contrast, the coupling of [Ca²⁺]_cyt and force responses in ASM is far more complex; however, phase-loop plot of ACh-induced [Ca²⁺]_cyt and force responses still provides valuable physiological insight. The rising phase of the phase-loop plot most likely reflects rMLC₂₀ phosphorylation and cross-bridge recruitment, while the relaxation rate most likely reflects rMLC₂₀ dephosphorylation. Importantly, both the rising phase and the relaxation rate of the phase-loop plots of ACh-induced [Ca²⁺]_cyt and force responses were not affected by 24-h TNFα exposure. In contrast, the integral of the ACh-induced [Ca²⁺]_cyt and force phase-loop plot increased significantly after 24-h TNFα.