Amino Terminus

More than half of the TRPV4 protein is formed by the intracellularly located amino terminus that most prominently harbors three ankyrin repeat domains (ARD1–3) within an ankyrin repeat region from aa235 to aa367, a cluster of four protein kinase C (PKC)–phosphorylation sites and a cAMP-dependent–phosphorylation site upstream of the ankyrin repeat region, and a cluster of two PKC sites within and downstream of ARD3 (Figure 8.1A). A functionally identified Src family dependent tyrosine phosphorylation site has been described at aa253 within ARD1. Other potentially functionally important motifs are an N-myristoylation site at aa24–29, a bipartite nuclear targeting sequence that overlaps with part of ARD3, and a PKC phosphorylation site.

FIGURE 8.1

Schematic overview of TRPV4’s predicted structural and functional components. Shown are schematic representations of TRPV4’s amino terminus (A), its central region with the membrane-spanning domains and the pore loop (B), and the channel’s (more...)

It has been hypothesized that increased PKC activity in response to mechanical stress could be a factor in TRPV4 activation. Activation of PKC with phorbol esters leads to opening of TRPV4 and increase of intracellular Ca²⁺ concentration when the experiment is conducted at 37°C [4,9]. At room temperature, PKC-activating phorbol esters have no or very little effect on TRPV4 gating. The sites of PKC action on TRPV4 are unknown. Candidates for direct phosphorylation are the six PKC phosphorylation sites in the amino terminus and one motif just downstream of TM2 (Figure 8.1A–B). Another motif that has been implicated in regulating TRPV4 is a single Src family tyrosine phosphorylation site within ARD1 that is phosphorylated in response to hypotonic stress [19]. A Y253F point mutant exhibited nearly abolished hypotonicity responses [19], a result that recently has been challenged [20,21].

Membrane-Spanning Core Region and Pore Loop

The 242-aa-long central domain of TRPV4 consists of TM1–6, which feature, between TM5 and TM6, a short hydrophobic stretch that is the putative pore region or pore loop (PL, Figure 8.1B). TRPV4 does display only a few predicted short extracellular domains, 25 aa between TM1 and TM2, 4 aa between TM3 and TM4, and 34 aa and 12 aa upstream and downstream of the PL. The channel appears to be posttranslationally modified by glycosylation [22], and a bona fide Asn glycosylation site within the extracellular stretch between TM5 and the PL has been shown to be glycosylated in heterologously expressed TRPV4 [23]. A PKC phosphorylation site downstream of TM2 is potentially involved in the above-mentioned PKC regulation of TRPV4 activation.

By analogy to the bacterial KcsA K⁺ channel [12,13], the PL is probably composed of a short highly hydrophobic pore helix immediately followed by an ion selectivity segment, the channel selectivity filter, that has been postulated to encompass aa672–682 [16]. Several individual aas affect the selectivity and permeability; when mutated (notably aspartate residues D672 and D682, and, somewhat surprisingly, M680), they markedly impair calcium permeation, results are consistent with the purported selectivity filter of TRPV4 (see below).

Carboxyl Terminus

TRPV4’s carboxyl terminal tail appears to be the docking site for at least two interacting proteins. One of these sites has been narrowed down to a region between aa812 and aa831 and confers binding to calmodulin [24]. Mutations within this region resulted in a loss of Ca²⁺-dependent calmodulin binding and a loss of Ca²⁺-dependent potentiation of TRPV4 currents. This observed effect adds to the complex Ca²⁺ dependence of TRPV4 activity that also involves inhibitory effects [1,25].

Immediately upstream of the calmodulin-binding region, an interaction site with microfilament-associated protein 7 (MAP7) has been proposed. Coexpression of TRPV4 with MAP7 in CHO cells apparently increases the amount of TRPV4 protein associated with the plasma membrane, which could be a method employed by cells to control the density of TRPV4 in the plasma membrane [26].

Mechanosensitivity

The TRPV4 channel was initially cloned based on its sensitivity to hypoosmotic cell swelling, implicating the channel as a potential mechanosensitive channel [1–3] as mentioned. It was subsequently shown that application of fluid shear stress (or fluid flow) across the apical surface of TRPV4-expressing HEK293 epithelial cells and M-1 renal collecting duct cells activated the channel, confirming the mechanosensitive nature of activation [4,27]. The response to mechanical stimulation, however, was shown to be highly sensitive to temperature [1,4], implicating a temperature-induced augmentation or sensitization to mechanical stimuli. The mechanosensitive nature of TRPV4, or a TRPV4 complex, was most clearly demonstrated in a seminal study of C. elegans using osm-9 mutant worms lacking expression of a worm TRPV homologue, OSM-9, which underlies osmotic (hyperosmolarity) and mechanical (nose touch) avoidance behavior. Transgenically targeting expression of mammalian TRPV4 to the ASH amphid sensory neuron in the mutants (a normal site of OSM-9 expression) rescued the osmotic and mechanical avoidance responses, providing compelling evidence that in the in vivo setting, TRPV4 functions as an osmosensitive and mechanosensitive channel or sensory complex [28].

The mechanism by which the TRPV4 channel is activated by mechanical stresses is currently not fully understood. Two, not mutually exclusive, principal mechanisms have been proposed: direct mechanical activation and indirect activation through other transduction pathways. Cell swelling can lead to activation of both the phospholipase C (PLC)/diacylglycerol (DAG) transduction pathway [29,30] and the phospholipase A2 (PLA2)/arachidonic acid (AA) transduction pathway [31–33], two pathways that can modulate, or regulate, TRPV4 activation (see below). However, Nilius and coworkers demonstrated, in a detailed series of studies, that swelling-induced production of AA and its downstream epoxyeicosatrienoic acid (EET) metabolites (5,6-EET and 8,9-EET) may be responsible, at least in part, for hypotonic-induced activation of TRPV4 [10,11,21]. Inhibition of PLA2 or cytochrome P450 epoxygenase activity reduced the swelling-induced activation of TRPV4, while application of the EET metabolites or inhibition of EET hydrolysis enhanced both the swelling and AA-induced activation of TRPV4. TRPV4 knockout mice demonstrated a reduced swelling and AA-induced activation of calcium entry in mouse aortic endothelial cells [10]. Hence, the PLA2–cytochrome P450 pathway may reflect a dominant pathway and mechanism by which mechanical stress may activate the channel. However, direct application of membrane stretch per se on channel activation, such as recently demonstrated for TRPC1 [34], without input from arachidonic acid and its metabolites, or from DAG and its metabolites, has not been systematically tested and remains to be determined, particularly at physiologically relevant temperatures.

The mechanosensory functions of TRPV4 in mammals are slowly being elucidated as described in more detail previously [35,36] and in other chapters of this volume. As noted, TRPV4 has been shown to be activated by cell swelling [1–3], implicating a potential role in cell volume regulation [37], and in sensing shear stress, where the channel may play a key sensory role in flow-sensitive tissues such as vascular endothelial cells and renal tubular epithelial cells [4,6,27]. The channel has been strongly implicated as a mechanosensor in cilia of oviductal epithelial cells, showing an increased channel activity and accompanying Ca²⁺ influx, following an increase in fluid viscosity (i.e., increased mechanical load) [38]. In addition, hypotonicity-induced nociception has also been described as TRPV4-mediated [39]. Further, studies employing TRPV4-deficient knockout mice have implicated TRPV4 in specific functions: in sensing plasma osmolarity by the sensory circumventricular organs of the hypothalamus, which, in turn, regulate secretion of antidiuretic hormone and control of extracellular fluid osmolarity; [40,41] in sensing tail pressure by dorsal root ganglia sensory neurons, implicating TRPV4 as a high threshold mechanoreceptor; [7] and in sensing hypotonic cell volume in mouse aortic endothelial cells [10]. Other functions of TRPV4 are continuing to emerge, and many more are likely to be forthcoming [35,36].

Thermosensitivity

The survival of animals, especially warm-blooded animals, relies on the animals’ abilities to sense and respond to changes in temperature. TRP channels may be central components of this sensing ability. The TRPV channel members, TRPV1–4, have been shown to be activated by warm to noxious temperatures and have been implicated in temperature sensing by the body. The TRPV channels display defined temperature thresholds for activation: TRPV1, 42°C; [42,43] TRPV2, 52°C; [44] TRPV3, 31°C; [45–47] and TRPV4, near 27–35°C [5,6]. The channels have recently been dubbed “thermoTRPs” because of this feature [48]. These investigators demonstrated that the mechanism of temperature dependency may relate to a temperature-dependent shift of the channels’ voltage dependency (normally not a notable feature of TRP channels) to more physiologically relevant voltage ranges. That is, elevating the temperature for warm/hot TRPVs provides a left shift in the voltage sensitivity, making the voltage dependency apparent at more “physiological” voltages, thereby activating the channel. (Cool/cold-activated TRPs displayed a complementary right shift in the voltage sensitivity with cooling.) TRPV4, however, does not specifically fit this model, as it could not be demonstrated to be temperature sensitive when studied in patch-clamp studies where the membrane was detached from the cell [48]. Other cellular factors, yet to be identified, may be critical for the TRPV4 channel to display temperature-induced activation.

The physiological function of TRPV4 in thermosensation is still emerging [48,49]. Besides functioning as a general sensor of altered cell/tissue temperature, TRPV4 expression in skin keratinocytes has been implicated as a sensor of “warm” temperatures [50]. TRPV4 knockout studies clearly implicate TRPV4 in selecting “preferred” footpad temperatures and as being essential in thermal hyperalgesia [51].

Diacylglycerol and Phorbol Esters

The first ligand discovered to activate TRPV4 was the synthetic phorbol ester, 4α-phorbol 12,13-didecanoate (4α-PDD) [8]. It was observed that 4α-PDD could potently activate TRPV4, even at relatively modest concentrations (<1 μM), and it appeared to be relatively specific for TRPV4 because it does not appear to activate other TRPV family members. Since 4α-PDD is a non-protein kinase C–activating phorphol ester that is not metabolizable, it may activate TRPV4 by directly binding to the channel or an associated subunit. However, recent studies by Xu and coworkers [9] have demonstrated that other inactive 4α isomers can activate the channel with a potency that appears to be related to their chemical hydrophobicity. This may indicate a potential effect of the phorbol isomers on the lipid environment in modulating, or activating, the channel. The physiological relevance of such a lipid-modulated mechanism of activation may have broad implications on the function and mechanisms of TRPV4 gating as noted for other mechanogated channels [52,53].

The PKC-activating phorbol esters were originally considered only to activate TRPV4 weakly based on studies done at room temperature [8]. It was subsequently demonstrated that elevation of temperature to the physiological range resulted in a potent augmentation, or sensitization, of the channel to phorbol esters and mechanical stimuli [4]. At 37°C, PMA in the submicromolar range activated TRPV4 in a PKC-dependent manner, approaching the potency of 4α-PDD [4,9]. Other PKC-activating phorbol esters (phorbol 12,13-didecanoate, PDD; phorbol 12,13-dibutyrate, PDBu) [9] similarly activated the channel in a PKC-dependent manner, implicating PKC as an important modulator of TRPV4 activity. These studies raise the exciting possibility that signaling pathways underlying induction of PKC, such as G-protein-coupled receptors or tyrosine kinase receptors linked to activation of phospholipase C, may reflect emerging, as yet undiscovered, pathways for TRPV4 activation.

Arachidonic Acid and Epoxygenase Metabolites

The discovery that arachidonic acid (AA) and its downstream metabolites may activate TRPV4 provided the first evidence for endogenous agonists of the channel. It was initially shown by Nilius and coworkers [11] that the endocannabinoid ananda-mide (AEA) and its lipoxygenase (fatty acid amidohydrolase) metabolite, AA, could activate TRPV4. A nonmetabolizable analogue of AEA (e.g., methanandamide) was without effect, while an inhibitor of fatty acid amidohydrolase, phenylmethylsulfonyl fluoride, abolished the AEA-induced activation of TRPV4. It was subsequently shown that generating EET metabolites of AA via the cytochrome P450 epoxygenase pathway, notably 5,6-EET and 8,9-EET, could activate TRPV4 [10,21]. Other studies have shown that the 11,12-EET may similarly activate TRPV4 [54]. Furthermore, activation by AA metabolites appears to be distinct from the 4α-PDD pathway of activation because mutation of a tyrosine residue near the third transmembrane domain (Y555, Figure 8.1B) to an alanine impairs activation by 4α-PDD, but not by AA or cell swelling, implicating a potential role for tyrosine phosphorylation in the activation pathway for 4α-PDD [21]. Recent studies in mouse vascular endothelial cells, which express endogenous TRPV4, further demonstrate that inhibiting EET hydrolysis also leads to an enhanced response of AA or EET analogues, while in cells isolated from animals lacking TRPV4 (TRPV4^−/−), all TRPV4 agonists are without effect [10]. Hence, accumulating evidence points to the AA pathway and the epoxygenase metabolites as playing key roles in regulating TRPV4 activity in the endogenous setting.

The Molecular Basis of Selectivity

The molecular basis of the TRPV4 channel selectivity filter has not been fully elucidated, although charged residues in the purported filter segment, aa672–682, have been implicated. In particular, two aspartate residues (D672 and D682) appear to cooperatively affect the pore’s Ca²⁺ permeability and channel rectification [55]. Nevertheless, these two aas are not strictly linked as mutations to neutralize D672 alone does not have any strong effects, and neutralization of D682 but not D672 almost abolishes the blocking effect of ruthenium red [20,55].

Nilius and colleagues [16,20] have hypothesized that the structural design of the TRPV4 pore is comparable to that of the bacterial KcsA K⁺ channel [12,13]. Based on this assessment, the PL is characterized by a short pore helix followed by the selectivity filter (Figure 8.1B). This notion was spurred by the weak but noticeable similarity of the TIGMGD stretch of aa677–aa682 (Figure 8.1B) within the PL with the TXGYGD selectivity filter residues of the K⁺ channel signature sequence [12,13]. The aforementioned mutations of D672 and D682, as well as M680, which impair Ca²⁺ permeation and reduce whole cell current amplitude, circumstantially support this hypothesis and clearly demonstrate involvement of this segment of the PL in ion selectivity and permeability [20,55].

Emerging Paradigms of Gating

The conformational changes in channel structure that lead to functional transitions from closed to open conducting states (i.e. “gating”) are not well understood for most channels. Again, based on structural analysis of bacterial K⁺ channels [12–15], potential TRPV4 gating mechanisms can be suggested [16]. The tetrameric subunit arrangement provides for fourfold symmetry around the central pore region [17,18], where an inner transmembrane helical segment from each of the four subunits each arranges concentrically to form the channel barrel or pore. The transmembrane helices are tilted inward to form a cone shape of the barrel with narrowing toward the cytoplasmic face. This narrowing of the transmembrane segments toward the cytoplasmic face forms the main activation gate [17,18,56]. Hinging movements of these segments can widen the pore, thereby opening the gate, or can narrow the pore, thereby closing the gate. The overall structure of the TRPV channels, including TRPV4, would be consistent with such a gating structure, as speculated by others [16], although this model for TRPV4 gating has not been explicitly evaluated.

Other components of the structure of P-loop channels, notably the pore helix, may also play a role in the gating process. Rotation or movement of the pore helix can modulate gating, or perhaps act as a secondary gate, in some P-loop channels [57,58]. Recent studies of TRPV5 gating, using the cysteine-accessibility method, have shown that inhibiting the channel by internal hydrogen is associated with a clockwise rotation of the pore helix along its long axis [59]. It is implicit in these findings that such rotational movements of the pore helix may be critical for gating other TRPV channels, such as TRPV4. Therefore, it is reasonable to conclude that, at a minimum, both hinging movements of the helices forming the activation gate at the cytoplasm face and rotation of the pore helices of the pore loop regions are components of TRPV channel gates. For multimodal channels, such as TRPV4, it may be that other structural components to gating may also be at play. Hence, future studies should provide exciting new insights into the molecular mechanisms of gating of TRPV4 and other TRP channels.