Non‐muscle myosin II regulates aortic stiffness through effects on specific focal adhesion proteins and the non‐muscle cortical cytoskeleton

Abstract Non‐muscle myosin II (NMII) plays a role in many fundamental cellular processes including cell adhesion, migration, and cytokinesis. However, its role in mammalian vascular function is not well understood. Here, we investigated the function of NMII in the biomechanical and signalling properties of mouse aorta. We found that blebbistatin, an inhibitor of NMII, decreases agonist‐induced aortic stress and stiffness in a dose‐dependent manner. We also specifically demonstrate that in freshly isolated, contractile, aortic smooth muscle cells, the non‐muscle myosin IIA (NMIIA) isoform is associated with contractile filaments in the core of the cell as well as those in the non‐muscle cell cortex. However, the non‐muscle myosin IIB (NMIIB) isoform is excluded from the cell cortex and colocalizes only with contractile filaments. Furthermore, both siRNA knockdown of NMIIA and NMIIB isoforms in the differentiated A7r5 smooth muscle cell line and blebbistatin‐mediated inhibition of NM myosin II suppress agonist‐activated increases in phosphorylation of the focal adhesion proteins FAK Y925 and paxillin Y118. Thus, we show in the present study, for the first time that NMII regulates aortic stiffness and stress and that this regulation is mediated through the tension‐dependent phosphorylation of the focal adhesion proteins FAK and paxillin.

genes (MYH9, MYH10 and MYH14, respectively), which are located on three different chromosomes. 7,8 Although all three NMII isoforms contain similar domain structures and exhibit 60%-80% amino acid sequence identity, 9 they have different expressions and intracellular localizations in different cell and tissue types. NMIIA and NMIIB isoforms are expressed at a high level in many smooth muscle tissues, whereas NMIIC expression is low in SM tissue and high in neuronal tissue. 7 Ablation of NMIIA and NMIIB in the whole mouse results in lethality at an embryonic stage, 2,10,11 whereas ablation of NMIIC causes no known phenotypic change. 12 In addition to the classical roles in cell adhesion, cell migration, and cytokinesis, a few studies have reported NMII involvement in the regulation of the biomechanics of smooth muscle contraction. 6,[13][14][15] However, the subcellular molecular mechanisms by which NMII regulate smooth muscle biomechanics remain unclear and are likely to be tissue-specific.
Stiffness is an important biomechanical property of the aorta.
Aortic stiffness increases with age and is an independent predictor of negative cardiovascular outcome, including hypertension, stroke, kidney disease and vascular dementia. [16][17][18][19] Recent studies have shown that vascular smooth muscle cells (VSMCs) contribute significantly to the total vascular wall stiffness. 20,21 There are three dynamic components of the VSM cell that have been shown to contribute to VSMC stiffness: first, the cyclic attachment of cross-bridges in contractile filament; second, the remodelling and transmission of force and stiffness through a non-muscle actin cytoskeleton 22,23 ; and third, remodelling of focal adhesion complexes connected to the non-muscle actin cytoskeleton. 24 Although NMII has been shown to be involved in the cell-mediated extracellular matrix reorganization and the resulting changes to the matrix stiffness, 25 its possible role in the regulation of VSM cell stiffness has not been investigated.
The goal of this study was to determine whether NMII plays a role in the biomechanical properties of the aortic wall, and if so, to determine the molecular mechanisms. Here we show that NMII is, indeed, involved in the regulation of both aortic stress and stiffness.
Furthermore, we show that the molecular mechanisms by which NMII regulates aortic stiffness include tension-dependent alterations of the phosphorylation state of smooth muscle focal adhesion proteins and the cortical non-muscle actin cytoskeleton of the smooth muscle cell.

| Animals and aortic tissue preparation
C57BL/6J adult male mice (~3-4 months old) were used in this study. All the procedures were performed according to a protocol approved by the Boston University IACUC. Mice were killed by inhalation with an overdose of isoflurane in a closed chamber. Aortic tissue was quickly excised, rinsed and placed in an ice-cold oxygenated (95% O 2 -5% CO 2 ) physiological saline solution (PSS; in mmol/L: 120 NaCl, 5.9 KCl, 11.5 Dextrose, 25 NaHCO 3 , 1.2 NaH 2 PO 4 , 1.2 MgCl 2 , 2.5 CaCl 2 ; pH, 7.4). Axial rings (5 mm length) were cut from the thoracic aorta for biomechanics. At the end of the protocol, rings were quick-frozen in a dry-ice, acetone, 10 mmol/L dithiothreitol, 10% TCA slurry and stored at −80°C.

| Measurement of aortic geometry and biomechanics
Axial length, diameter, and wall thickness of aortic tissue were quantitated to determine the cross-sectional area and used to calculate aortic stress and stiffness. Ex vivo aortic stress and stiffness were measured as previously described. 20,26,27 Briefly, thin triangular pieces of wire (0.01-inch diameter) were threaded through the lumen of the aortic ring and suspended in an organ bath (50 ml) containing 37°C oxygenated PSS (95% O 2 -5% CO 2 ). The upper triangle was connected to a computer-controlled motorized lever arm (Dual-Mode Lever Arm System, Model 300C, Aurora Scientific), which also functions as a force transducer. Aortic rings were stretched in the circumferential direction to the previously determined optimum length (1.8 × slack length).
Aortic stiffness was determined by high-frequency (40 Hz), small-amplitude (1%) sinusoidal length perturbations. Aortic tissue stiffness is defined as the change in stress divided by the change in strain in response to the sinusoidal oscillations, (ΔF/A)/ΔL/L 0 ) where ΔF is the amplitude of force output during cyclic stretches, A is the cross-sectional area, ΔL is the amplitude of cyclic length changes, and L 0 is the optimal length. Stress was calculated as force ΔF divided by the cross-sectional area (A = 2hι where h is the wall thickness and ι is the axial length of the ring).

| Isolation of single differentiated smooth muscle cells from mouse aorta
To avoid any cytoskeletal changes that could occur during cell culture, freshly dissociated VSMCs (dVSMCs) were used in this study.
Contractile dVSMCs were enzymatically isolated from mouse aorta tissue according to a previously published method 22

| A7r5 cell culture and siRNA transfection
A7r5 rat aortic SMCs (ATCC) were cultured in growth medium composed of Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% foetal bovine serum and 1% glutamine and maintained under humidified conditions at 37°C and 5% CO 2 .

Expression of NMII isoforms was inhibited by ON-TARGETplus
SMARTpool siRNAs specific for NMIIA and NMIIB, purchased from Horizon discovery. ON-TARGETplus Non-targeting Control siRNA was used as a control. Cells were seeded at a density of 1 × 10 5 cells/ mL and were transfected using DharmaFECT (Horizon Discovery), at a final concentration of 25 nmol/L in an antibiotic-free Opti-MEM medium according to the manufacturer's instructions. siRNA knockdown was analysed after 72 hours of transfection by Western blot using antibodies specific to NMIIA and NMIIB. Cells were lysed on ice, and samples were stored at −80°C.

| Western blotting
Frozen aortic tissue was quickly homogenized and total protein measured as described previously by our group. 29 Bands were normalized to GAPDH.
Equal volume of G and F-actin samples was resolved by SDS-PAGE and immunoblotting with anti-actin primary antibody (Cytoskeleton).

| Statistical analysis
All results are presented as mean ± SEM Analysis was carried out using the GraphPad Prism 8 software. Two-tailed student's t tests were used when comparing values between two experimental groups. The significance of differences between two individual datasets was taken at P < .05. We used two-way ANOVA to compare parameters among treatment groups followed by Turkey's multiple comparisons test.

| Ex vivo inhibition of non-muscle myosin II in the mouse inhibits alpha agonist-activated aortic stress and stiffness
To elucidate the role of NMII in vascular function, we used a small molecule inhibitor of NMII, blebbistatin (Bleb). Blebbistatin binds to the myosin ADP-Pi complex of the myosin ATPase cycle and interferes with the release of phosphate; therefore, it blocks the myosin II head in the actin-detached state and prevents actin-myosin interactions. 31 Based on in vitro and ex vivo characterization, blebbistatin has been shown to be a cell permeable selective inhibitor of NMII (IC 50 ~ 0.5-5 μmol/L) and has a low binding affinity towards smooth muscle myosin (SMII) (IC 50 ~ 15-80 μmol/L). 4,[31][32][33][34][35][36][37][38] Only the active (−)-enantiomer of blebbistatin is known to inhibit NMII activity, and only this enantiomer was used in the present study. 32,33,39 To determine the effective concentration of blebbistatin required to alter the biomechanics of mouse aorta tissue, we per- To investigate the role of NMII specifically in aortic stiffness, we used biomechanical techniques to quantitate agonist-induced aortic stiffness in the presence and absence of blebbistatin. 30 As seen in Figure 1C, blebbistatin significantly decreases phenylephrine-induced increases in aortic stiffness at 5 and 10 μmol/L concentrations (n = 8) as compared to control tissues treated with vehicle only.
However, similar to the aortic stress, no statistically significant effect was observed on aortic stiffness at 2 μmol/L concentration of blebbistatin as compared to control tissue.

| Blebbistatin also decreases depolarizationinduced increases in aortic stress and stiffness
Multiple signalling pathways coexist in SM cells, which contribute to SM contractility and stiffness. 40 Membrane depolarization activates SM by a signalling mechanism involving the activation of voltage gated Ca 2+ channels, Ca 2+ influx, Ca 2+ binding to calmodulin, Ca 2+ -calmodulin complex to activate myosin light chain kinase, F I G U R E 1 Effect of blebbistatin on biomechanical properties of aortic tissue activated by an alpha agonist. A, Isometric force development in control vehicle-treated aorta tissue (upper panel) and blebbistatin-treated aorta tissue (lower panel). Aortic rings were treated for 30 min with 2, 5 and 10 μmol/L blebbistatin after PE stimulation. B, Blebbistatin significantly inhibited PE-induced aortic stress at 5 and 10μmol/L as compared to control vehicle-treated tissues (n = 8). C, Blebbistatin significantly inhibited PE-induced aortic stiffness at 5 and 10 μmol/L as compared to only vehicle-treated tissues. No statistically significant inhibition was found at 2 μmol/L on aortic stress and stiffness as compared to vehicle-treated tissues (n = 8). Values in graphs are normalized to PE-induced control stress/stiffness. Control stress/stiffness refers to the stress/stiffness measured in the presence of PE before blebbistatin/vehicle treatment. ** and **** indicate P < 0.005 and <0.0001, respectively. Two-way ANOVA **** **** **** ** A B C phosphorylation of myosin light chains, and actin-myosin crossbridge cycling and contraction.
To determine whether NMII is involved in the depolarization-mediated pathway of contraction, aorta tissues were treated with 10 μmol/L blebbistatin for 30 minutes followed by treatment with a depolarizing physiological saline solution containing 51 mmol/L KCl.
Blebbistatin significantly decreases the KCl-induced increase in aortic stress as compared to only vehicle-treated tissues (Figure 2A,B).
KCl-induced aortic stiffness was also significantly decreased by blebbistatin ( Figure 2C). These results indicate that NMII affects parts of signalling pathways common to both KCl and PE activation, either in the contractile filaments, focal adhesions or non-muscle actin cytoskeleton. To study whether NMII localizes with the contractile filaments in the core of the cell or with the non-muscle cytoskeleton adjacent to the plasmalemma, we enzymatically dissociated single dVSMCs from the mouse aorta and imaged the cells with high-resolution deconvolution microscopy as previously described. 22 Our laboratory has published that α-smooth muscle actin associates with contractile filaments in the core of the VSMC, whereas γ-cytoplasmic actin associates with the sub-plasmalemmal cell cortex and β-cytoplasmic actin is located in punctate dense bodies in ferret VSMC. 43  We then costained for the NMII isoforms to allow comparison with the specific actin isoforms in the aortic smcs. We found that NMIIA colocalizes with α-SMA filaments ( Figure 3B, upper row), presumably in the contractile filaments in the core of the cell ( Figure 3B upper row, enlarged view). Then, we costained for NMIIA and the cortical γ-CYA ( Figure 3B, lower row). The merged image shows that NMIIA also colocalizes with cortical γ-CYA ( Figure 3B, lower row, enlarged view).
Next, we investigated the localization of the NMIIB isoform ( Figure 3C). First, we compared it with that of α-SMA in the contractile filaments ( Figure 3C,

| NM myosin II is not required for the endocytic recycling of focal adhesion proteins
FAs in contractile dVSM cells are dynamic in nature. Prior studies from our laboratory and those of others have shown that redistribution of FA proteins occurs 24,50,51 in response to the addition of vasoconstrictors, and the cyclic remodelling of FA proteins requires an endocytic recycling pathway in vascular smooth muscles. 24 Thus, we speculated that the endocytic recycling of the focal adhesion proteins might be NM myosin II-dependent. To test this hypothesis, we first determined whether endosomes have any role in the regulation of aortic stiffness in VSM. We inhibited endosome function by treating the tissue for 1 hour with 150 μmol/L primaquine, an inhibitor of endosome budding. As seen in Figure 6A-C, both PE-induced aortic stress and stiffness decreased in the primaquine treated tissues as compared to control tissues.
Then, we asked whether endosomes are necessary for the regulation of phosphorylation of FA proteins observed here. Thus, we examined the phosphorylation of FAK and paxillin in the presence or absence of primaquine and found that PE increases the phosphorylation of FAK and paxillin at Y925 and Y118 sites, respectively; however, no significant changes were detectable in the phosphorylation of FAK and paxillin in the presence of primaquine ( Figure 6D-E).
These data indicate that the endosomes are not necessary for the role of NMII in regulating FA phosphorylation and that the phosphorylation of FA proteins, which is regulated by NMII, must occur, temporally, before entry into the endosomes.
Additionally, the question arises as to how the endosome regulates aortic stiffness in VSM. Previous studies have shown that actin polymerization increases in the presence of agonists and that it is also a key regulator of vascular contractility and stiffness. 23,30 Thus, we tested whether endosomes regulate aortic stiffness via regulation of actin polymerization and whether actin polymerization is sensitive to primaquine treatment. Aortic tissues were treated in the presence and absence of primaquine. As shown by filamentous/ globular actin ratios in Figure 6F, the PE-induced increase in actin polymerization was significantly decreased in primaquine-treated tissues. Thus, these data indicate that endosomes regulate aortic stiffness in VSMCs via an effect on actin polymerization and that this mechanism does not involve an effect on NMII.

| D ISCUSS I ON
The present study demonstrates that NMII regulates stress and stiffness in aortic SMC. Further, we show here that NMII not only localizes to the cell cortex and the contractile filaments, but also regulates aortic stress and stiffness via effects on FA signalling and the activity of the contractile filaments. Our data support the hypothesis that in addition to the SMII, NMII regulates tonic force maintenance and contributes to smooth muscle contractility as well as aortic stiffness, which therefore may provide a novel therapeutic target for the treatment of vascular disease. Of note, another group 39 has reported an apparently low level of NMII in rabbit smooth muscles, but that blebbistatin concentration (~4-5 μmol/L) inhibited the tonic contractions of adult rabbit femoral and saphenous arteries, suggesting either a significant functional effect of the low abundance NMII in this tissue or an action of blebbistatin separate from its inhibition of NMII. Therefore, the role of NMII in smooth muscle contraction may be organ-specific and species-specific.
In this study, we inhibited NMII with blebbistatin to determine its contribution in vascular contractility and stiffness in mouse aorta.
Blebbistatin is known to selectively inhibit NMII (IC 50 ~0.5-5 μmol/L) and has a low affinity towards smooth muscle myosin II (IC 50 ~80 μmol/L) as well as myosin classes I, V and X. [31][32][33] However, one group has suggested that blebbistatin can also inhibit some smooth Importantly, our results are first to show that blebbistatin inhibits agonist-induced aortic stiffness which is a recognized risk factor associated with ageing-induced vascular dementia, as well as kidney and heart disease.
Although NMII isoforms share considerable similarity in their amino acid sequence and molecular structure, they exhibit differential expression, function and subcellular localization in different tissues. 7,37,41,42 The NMIIA isoform is predominantly expressed in adult bladder tissue, whereas the NMIIB isoform is highly expressed in the aorta tissue. 13,55 Our results in freshly isolated dVSMCs from mouse aorta demonstrate that NMIIA isoforms colocalize with α-SMA-containing contractile filaments as well as with γ-CYA at the cell cortex ( Figure 3B). Moreover, we show that NMIIB colocalizes with α-SMA contractile filaments. However, we did not observe colocalization of NMIIB with γ-CYA a marker of cortical actin ( Figure 3C).
NMMII isoforms have been implicated in actin and focal adhesion remodelling in many different cell types. 56

ACK N OWLED G EM ENTS
This research was supported by NIH NIA AG053274.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the finding of this study are available from the corresponding author upon reasonable request.

R E FE R E N C E S
F I G U R E 7 NM myosin II isoform function in the regulation of aortic stress and stiffness [Colour figure can be viewed at wileyonlinelibrary.com]