Uni-axial stretch induces actin stress fiber reorganization and activates c-Jun NH₂ terminal kinase via RhoA and Rho kinase in human bladder smooth muscle cells

Bladder wall remodeling such as smooth muscle hypertrophy/hyperplasia occurs under conditions of bladder outlet obstruction including benign prostate hyperplasia. Although these etiologies are not well understood, an excessive mechanical overload might be a prior factor. In the human obstructed bladder, sustained stretch stress is believed to cause bladder wall remodeling such as a change in the ratio of extracellular matrix and smooth muscle cells [1, 2]. In vivo animal models with a partial urethral ligature have revealed smooth muscle hypertrophy in the bladder wall, which are quite similar to the obstructed bladder in humans [3]. Although in vitro devices do not completely mimic the bladder wall overload, stretch devices that enable the stimulation of cultured bladder smooth muscle cells were used to reveal the pathological mechanisms under condition of mechanical overload. Park et al. indicated that angiotensin release induced by mechanical stretch acts as mitogen in bladder smooth muscle cells [4, 5].

Some studies have identified intracellular signaling pathways that mediate the biological effects evoked by mechanical stimuli and ultimately lead to nuclear events [5]. Of these pathways, the mitogen activated protein kinases (MAPKs), which constitute a family of serine/threonine kinases, are known to mediate signals that are activated by external stimuli and that regulate cell growth and differentiation. One MAPK family member, c-Jun NH₂-terminal kinase (JNK), has been reported to be activated by mechanical stretch in vascular smooth muscle cells [6] and cardiac myocytes [7]. In addition, Nguyen et al. indicated that cyclic stretch activates JNK in bladder smooth muscle cells [8]. We also showed that stretch stimulation activated JNK in rat bladder smooth muscle cells by the influx of Ca²⁺ through a stretch activated ion channel [9].

RhoA is a member of the Rho family of 20 to 30 kDa GTPase proteins that cycle between an active GTP-bound form and an inactive GDP-bound form. One of the important roles of RhoA is to act as a regulator of actin stress fibers [10]. RhoA is involved in cell division, movement, polarization and morphological changes via reorganization of actin stress fibers. The Rho-associated coiled-coil forming protein kinase (ROCK) is a molecule of RhoA that acts as a serine/threonine kinase and phosphorylates various substrates. Actin stress fiber reorganization was recently reported to be mediated by the ROCK pathway [11]. The physical deformation of cells appears to be caused by reorganization of actin stress fibers in order to adapt to their extracellular environments.

Previous evidence suggested that the RhoA/ROCK pathway is involved in the pathogenesis of obstructed bladder [12] and in the change in Ca²⁺ sensitization due to agonist stimulation [13]. Poley et al. indicated that quick stretch of rabbit bladder smooth muscle sufficient to induce calcium entry and stimulate a myogenic contraction does not activate the ROCK, and that basally active ROCK is necessary for stretch induced myogenic contraction [14]. We undertook to identify the roles of the RhoA/ROCK pathway in the early signalling events evoked by mechanical stimuli in human bladder smooth muscle cells (HBSMCs).

Mechanical overload is possibly involved in remodeling of the bladder wall, and such mechanical stress may cause morphological changes in HBSMCs. To understand the intracellular events caused by mechanical stretch, we focused on the RhoA/ROCK signaling pathway. The results demonstrated the following: 1) Uni-axial stretch activated JNK1 in HBSMCs. 2) Uni-axial stretch strongly evoked the reorganization of actin stress fibers and coordinated their orientation so that they became more perpendicular to the stretch direction. 3) The inhibitor experiments showed that the RhoA/ROCK pathway had pivotal roles in transduction of the intracellular signals induced by mechanical stretch. The stretch parameters used in this experiment were chosen to better emphasize the reaction of actin stress fiber reorganization. It should be noted that the stretch cycle used differs from the physiological condition and further assessment is needed to reflect the actual in vivo status.

RhoA mediates various extracellular signals that are induced by stimulation with agonists such as noradrenaline [17] and acetylcholine [16] in smooth muscle. RhoA was also shown to play significant roles in the signal transduction induced by mechanical stretch in endothelial cells [18] and cardiac myocytes [19]. Similarly, our data showed that RhoA appeared to have critical roles in the signaling induced by mechanical stretch in HBSMCs. However, the molecules that lead to RhoA activation by mechanical stimuli remain unclear. It is possible that scaffold proteins such as integrins or mechanosensitive ion channels [9, 20] may participate in the activation of RhoA by sensing physical stimuli via the extracellular matrix. However, further experiments are required to clarify this mechanism.

Our results suggested the existence of signal cross-linkage between RhoA/ROCK and JNK pathways in HBSMCs. Previous papers have also reported that RhoA induces JNK activation in cardiac myocytes by mechanical stretching [15]. Furthermore, stimulation of NIH 3T3 cells with an agonist such as lysophosphatidic acid (LPA) [21] or of vascular smooth muscle cells with Angiotensin II [22] causes signaling cross-talk between the ROCK pathway and JNK. There was no clear indication as to which molecules in JNK cascades are stimulated by mechanical stretch activation of RhoA/ROCK. However, some previous reports showed that ROCK activation by LPA in NIH3T3 cells induced activation of mitogen-activated protein kinase kinase 4 (MKK4), which is an upstream kinase of JNK and leads to the phosphorylation of JNK [21]. In addition, ROCK activation by activin in keratinocytes induced activation of mitogen-activated protein kinase kinase kinase 1 (MEKK1), which is an upstream kinase of MKK4, and subsequently leads to JNK activation [23]. The activation of JNK by mechanical stretch was previously shown to promote the expression of immediate early genes in vascular smooth muscle cells [6] and cardiomyocytes [7]. Although our data indicated that the JNK inhibitor SP600125 inhibited JNK activation by mechanical stretch, it should be noted that a recent paper suggested a limitation to the specificity of SP600125 and suggested the possibility that SP600125 might also inhibit other kinases (e.g., phosphatidylinositol 3-kinase) [24].

Actin stress fibers are composed of from 20 to 30 parallel actin filaments that are bundled by a bridge protein, α-actinin [10]. Actin filaments are temporally connected to bundles of polymerized type II myosin and generate tension by utilizing the actomyosin bond. The ends of actin stress fibers connect to focal adhesions where cell membranes attach to the extra-cellular matrix [10]. Under static conditions, actin stress fibers generate an isometric contraction in order to maintain a constant length with scaffolding at both ends. Cells are thought to sense extra-cellular physical signals using this isometric contraction.

In the stretch experiments of the present study, actin stress fibers were reorganized by mechanical stretch, such that stronger actin stress fibers were formed as a result of the stretch stimuli and the stress fibers were organized so that their orientation came close to being perpendicular to the stretch direction. It is unclear whether actin stress fiber angle change induces alteration of cell contractility in BSMCs. Moreover, there is little evidence that actin fiber orientation change enhances smooth muscle contractility in other types of muscle cells. Many of the cells formed projections such as filopodia in the lateral side of their cell membrane, where actin stress fibers are bound. Reorganized fibers may cause isometric tension in order to maintain the cell area. These morphological changes appear to be an adaptation response for the conservation of intracellular configurations because the formation of actin stress fibers perpendicular to the direction of stretch may restrict the movement of cell organelles. In addition, the shape of some cells was also elongated perpendicular to the direction of stretch. We also conjecture that cells may try to change their polarity by using actin stress fibers as a frame in order to reduce mechanical stress. However, there was no significant difference in the overall horizontal-vertical ratio of the cell length of stretched cells vs. unstretched cells due to their wide variety of cell shapes. The morphological changes of the stress fibers may be induced via the RhoA/ROCK pathway because the C3 exoenzyme and Y-27632 inhibited their reorganization. ROCK has been reported to participate in the induction of prominent actin stress fiber formation [11]. One of the mechanisms underlying the induction of stress fiber formation by ROCK is the phosphorylation of LIM-kinase, which is an effector of cofilin. It is considered that phosphorylated cofilin is subsequently rendered inactive and that thereby actin polymerization is ultimately stabilized [25]. Moreover, Na⁺/H⁺ exchanger 1, which is another substrate of ROCK, is activated by ROCK and subsequently induces the connection between actin filaments and cell surface proteins [26, 27]. Furthermore, mDia, which is an effector of RhoA, is reported to induce actin stress fiber reorganization together with ROCK [28]. It is reported that stress fiber orientation change due to mechanical stretch needs mDia activation in endothelial cells [18]. ROCK has also been reported to induce actinomyosin contraction via phosphorylation of at least four substrates including myosin light chain, all of which lead to increased myosin phosphorylation and increased actomyosin contractility [10].

Our results indicated that the JNK inhibitor SP600125 did not suppress actin stress fiber reorganization, therefore the JNK pathway may not participate in this reorganization. An interesting finding was reported by Kaunas et al. that a 90° change in the direction of stretch caused re-activation of JNK, when JNK activity had disappeared after the initial stretch. In addition, JNK activity was prolonged by cytochalasin D which also suppresses actin stress fiber reorganization [29]. Hence, actin cytoskeleton contraction appears to act as some sort of trigger for JNK activation. Their concept was that deformation of the actin cytoskeleton could induce biochemical signals since the stretch-induced JNK activation subsides if stress fibers are able to organize in a configuration that minimizes the perturbation by stretch.

Bladder smooth muscle cells appear to have elaborate mechanisms for sensing mechanical stress and for adapting to mechanical stress overload by cytoskeletal remodeling and by activating cell growth signals such as JNK via RhoA/ROCK pathways.