Loukin SH et al. (2015), A channelopathy mechanism revealed by direct ca...

XB-ART-51476

Proc Natl Acad Sci U S A 2015 Jul 28;11230:9400-5. doi: 10.1073/pnas.1510602112.

Show Gene links Show Anatomy links

A channelopathy mechanism revealed by direct calmodulin activation of TrpV4.

Loukin SH , Teng J , Kung C .

???displayArticle.abstract???
Ca(2+)-calmodulin (CaM) regulates varieties of ion channels, including Transient Receptor Potential vanilloid subtype 4 (TrpV4). It has previously been proposed that internal Ca(2+) increases TrpV4 activity through Ca(2+)-CaM binding to a C-terminal Ca(2+)-CaM binding domain (CBD). We confirmed this model by directly presenting Ca(2+)-CaM protein to membrane patches excised from TrpV4-expressing oocytes. Over 50 TRPV4 mutations are now known to cause heritable skeletal dysplasia (SD) and other diseases in human. We have previously examined 14 SD alleles and found them to all have gain-of-function effects, with the gain of constitutive open probability paralleling disease severity. Among the 14 SD alleles examined, E797K and P799L are located immediate upstream of the CBD. They not only have increase basal activity, but, unlike the wild-type or other SD-mutant channels examined, they were greatly reduced in their response to Ca(2+)-CaM. Deleting a 10-residue upstream peptide (Δ795-804) that covers the two SD mutant sites resulted in strong constitutive activity and the complete lack of Ca(2+)-CaM response. We propose that the region immediately upstream of CBD is an autoinhibitory domain that maintains the closed state through electrostatic interactions, and adjacent detachable Ca(2+)-CaM binding to CBD sterically interferes with this autoinhibition. This work further supports the notion that TrpV4 mutations cause SD by constitutive leakage. However, the closed conformation is likely destabilized by various mutations by different mechanisms, including the permanent removal of an autoinhibition documented here.

???displayArticle.pubmedLink??? 26170305
???displayArticle.pmcLink??? PMC4522800
???displayArticle.link??? Proc Natl Acad Sci U S A
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: trpv1 trpv4

???displayArticle.disOnts??? bone development disease

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Exogenous Ca2+-CaM activates wild-type TrpV4 but not TrpV4 with a key mutation that prevents Ca2+-CaM binding. (A) A representative patch of low activity from oocytes expressing wild-type TRPV4 was excised and clamped at a test potential of 80 mV in ionically symmetric solutions containing 0.1 mM CaCl2 Methods). Little activity was observed at the start (region â€œiâ€ in trace and all-points histogram below). At the point indicated in this continuous recording, 1Î¼M bovine CaM was added to and mixed in the bath (mixing artifacts are shown, but grayed out in this and all subsequent traces) resulting in a substantial increase in TrpV4 activity (â€œiiâ€). Chelation of Ca2+ with 0.5 mM K+ BAPTA suppressed (iii) and readdition of 1 mM CaCl2 (â€œivâ€) restored this Ca2+-CaM-dependent activation of the wild-type TrpV4. Zero current levels are indicated by horizontal carats at left of traces. (B) Ca2+-CaM presentation as in (A) to a representative patch with higher basal activity excised from an oocyte expressing the CaM-nonbinding W822A TrpV4 (9). No CaM activation can be discerned. Note that the initial basal Pos for the exemplar traces illustrated here were low for wild type, so that unitary activity could still be discerned after robust activation, yet higher for W822A, so that lack of activation could be demonstrated. We did not observe a systematically lower basal Po for W822A overall.
	Fig. 2. W822A TrpV4 maintains activation by mechanical stimuli. (A) A continuous recording showing that the wild-type TrpV4 opened repeatedly in response to membrane stretch pulses in inside-out patches excised from oocytes. (B) All-point histograms corresponding to each regions from this trace before (Upper) and during suction (Lower). (C and D) A continuous recording showing that the CaM-nonresponding W822A TrpV4 channels maintain their mechanosensitivity. Recording conditions are identical to the initial conditions described in Fig. 1 and in Methods except that the test potential was 60 mV. Zero current/pressure levels are indicated by horizontal carats at the left in this and all subsequent figures.
	Fig. 3. TrpV4 with disease-causing mutations upstream of the CaM-binding domain resembles a constitutive Ca2+-CaM activated state. (A) Oocytes expressing reduced levels of TrpV4 with P799L mutation, examined as in Fig 1 (except at a test potential of 60 mV) show strong constitutive activity, but in the majority of cases Ca2+-CaM activation cannot be discerned (but see text and Fig. S1). (B) All-points histograms of the traces in A before (i) and after (ii) Ca2+-CaM addition. C and D are arranged as A and B for the same examination of TrpV4 with the E797K mutation. Continuous recording from P799L (E and F) and E797K channels (G and H) showed that they were not generally defective in opening, being repeatedly activatable by membrane stretches. All point histograms on the right (F and H) correspond to the labeled regions in the traces with suction being present in the lower but not the upper two histograms in each case.
	Fig. S1. Modest Ca2+- CaM activation can sometimes be observed for E797K and P799L. 30â40% of the time, modest CaM activation can be detected for both P799L (Upper) as well as E797K (Lower). Experimental conditions are identical to those described in Fig. 3A). Such modest activation contrasts to the robust activation of wild-type channels (Fig. 1A) and non-AID mutants (see Fig. 2A). It also explains why the basal activities of the point mutants are high, but less than that of the AID deletion mutant or the double 797/8K mutant (Fig 4).
	Fig. S2. TrpV4 channels with disease-causing mutations not in proximity to the CaM-binding domain remain activatable by Ca2+-CaM. (A) Patches excised from oocytes expressing a SD-causing allele, V620I, located in S5, remained readily activated by Ca2+-CaM. Experimental conditions were as described in Fig. 1. (B) Other SD-causing mutants not near the CaM binding site including T89I (amino terminal), D333G and Î´333â€“7 (ankyrin repeat domain), and R616Q (S4â€“S5 linker) likewise remain activatable by Ca2+-CaM. (C) Maximal fractional increase in nPo observed for the various mutants. Conditions are as in Fig. 3 Aâ€“D.
	Fig. S3. Ca2+-CaM does not activate AID deletion mutant in an exemplar patch. Conditions are as described in Fig. 1 except that patches were excised from oocytes expressing the AID deletion mutant â795â804.
	Fig. S4. N-terminal basic domain mutants remain robustly activatable by Ca2+-CaM. Excised patches from the charge-neutralized âKRAâ mutant (A) as well as the more severe charge-reversed â5xKDâ mutant (B) were exposed to bovine Ca2+-CaM as in Fig. 1. Test potential was 60 mV. See text.
	Fig. 4. Basal activity dramatically increased by deletion of the AID but not other key mutations. (A) Whole-oocyte currents were recorded with a two-electrode voltage clamp from oocytes injected with either water or a small amount (0.1 ng) of wild-type or mutant TRPV4 cRNA. Only very low levels of TrpV4-specific basal current could be detected in wild-type injectants compared with water-injected oocytes (Left, inset traces have a 20-fold magnified current scale), whereas a dramatic increase in currents was observed after 5-min exposure to 1 Î¼M GSK in the wild-type-TRPV4 cRNA-injected but not the water-injected oocytes (Right). As we have shown (5), oocytes expressing TRPV4-E797K and -P799L have much higher basal current densities with relatively small increases after GSK exposure due to this (note that there was greater inactivation of these mutant currents here than previously observed, which we attribute to a change in our protocol designed to prevent inward Ba2+ flux through TrpV4, see Methods. TrpV4 with a 10-aa deletion (Î´795â€“804) removing the E797-P799 region, had high basal activity and likewise only nominal further stimulation by GSK. (B) Basal open probability (Po) estimated by the relative levels of leak-subtracted peak currents at 40 mV before and after GSK exposure from oocytes expressing the TRPV4 alleles shown in A, and five more: the CaM-nonbinding W822A, the putative N-terminal binding domain mutants KRA and 5xKD (see text), the E77K/D798K double mutant and the CBD deletion mutant Î´CBD. (Inset) Pos of the low-basal Po mutants on an expanded y axis aligned to their identities on the main axis below (mean Â± SEM, n > 6).
	Fig. 5. Autoinhibitory and CaM-binding domains in carboxyl tail. (A) Topological cartoon illustrating key regions of the carboxyl domain of TrpV4 considered in this work including the CaM-binding domain (CBD, green) and the proposed AID (blue). Also shown are the locations of the human SD mutations E799K and P799L as well as the engineered CaM-binding mutation W822A. (B) Alignment of the entire carboxyl termini of TrpV1 (Upper) and V4 (Lower). Identical residues (red) indicated conservation through but not beyond the AID. The extent of the known cryo structure of TrpV1 is indicated by the vertical dotted line (in A as well). Extended Î²-sheet (arrows) secondary structures predicted using PSSpred (Y. Zhang, zhanglab.ccmb.med.umich.edu/PSSpred) are indicated above and below the primary sequences for TrpV1 and TrpV4, respectively, with the red arrow highlighting the likely undefined Î² sheet from the cryo structure. Point mutants examined in this study are indicated by yellow boxes, the blue bar indicates the extent of the AID deletant and the green to the CBD deletant.
	Fig. S5. An AID model of Ca2+-CaM activation of TrpV4. (A) Basic autoinhibitory model of Ca2+-CaM activation of TrpV4 (see text). AID and CBD are colored as in (Fig. 5A) and Ca2+-CaM is represented as black dumbbell structure on right. Transitional arrows are weighted to approximate equilibrium biases. The purple arrow between states a and b reflects transitions altered by AID mutations, whereas the red arrows representing transitions effected by non-AID mutations. (B) Order-of-magnitude estimates of equilibrium constants corresponding to AID model of Ca2+-CaM activation shown in A for wild type (Left) AID deletant (Center) and non-AID mutants (â€œotherâ€, Right) are listed at Top. Transitions altered by AID mutation are highlighted in purple, whereas those altered by the other gating mutants are highlighted in red. (Middle) State populations based on 1,000 channels residing in state â€œaâ€. (Bottom) ICalculated Po in the absence (Upper) or presence (Lower) of Ca2+-CaM. Note that wild-type channels have their observed low basal Po (0.001) and strong (âˆ¼50-fold) CaM activation, both the mutants have high basal Pos of âˆ¼0.1, but only the non-AID mutant is significantly stimulated by Ca2+-CaM, opening with over 80% of the remaining closed channels being opened on average in its presence.

References [+] :

Adelman, SK channels and calmodulin. 2016, Pubmed