From Xenopus Interacts with the Penta-EF-Hand Protein Peflin

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 586 (2012) 4276–4281

journal homepage: www.FEBSLetters.org

Transient receptor potential N (TRPN1) from Xenopus interacts with the penta-EF-hand protein peﬂin ⇑ Dominik Wiemuth a,1, Lena van de Sandt a,1, Rudolf Herr b, Stefan Gründer a,

a Department of Physiology, RWTH Aachen University, 52074 Aachen, Germany b Department of Physiology, University of Tübingen, 72076 Tübingen, Germany

article info abstract

Article history: TRPN1 is a candidate mechanotransduction channel in Drosophila and Caenorhabditis elegans, also Received 2 October 2012 present in hair cells of lower vertebrates. At its N-terminal cytoplasmic tail it contains 28 ankyrin Accepted 26 October 2012 repeats. We performed a yeast two-hybrid screen with the N-terminal ankyrin repeats of Xenopus Available online 15 November 2012 TRPN1 as bait and identified the Penta-EF-hand protein peflin as a putative interaction partner. We confirmed this interaction by GST pulldown assays and by co-localization in an epithelial cell Edited by Gianni Cesareni model. Collectively, our study identifies peflin as an interaction partner of TRPN1 in vitro.

Keywords: Structured summary of protein interactions: TRPN1 Peflin TRPN1 physically interacts with Peflin by two hybrid (View Interaction: 1, 2, 3) PEF TRPN1 and Peflin colocalize by fluorescence microscopy (View Interaction: 1, 2) Mechanotransduction TRPN1 physically interacts with Peflin by pull down (View Interaction: 1, 2, 3) Hair cell Ó 2012 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. TRP channel

1. Introduction A unique feature of TRPN1 is the presence of 28–29 ankyrin (ANK) repeats at its cytoplasmic N-terminal tail, comprising 1 Transient receptor potential N (TRPN1) is a good candidate for a 000 amino acids. Crystallization of twelve ANK repeats of human mechanosensitive transduction channel in Drosophila [1,2] and ankyrinR revealed that they fold into a superhelix [12], resembling Caenorhabditis elegans [3]. It was initially identified in a genetic a nanospring. Atomic-force microscopic measurements indeed re- screen for touch-insensitive Drosophila mutants [4,5] and dubbed vealed that tandem ANK repeats are elastic and behave as a linear no mechanoreceptor potential C (nompC). Recently, Drosophila and reversible spring [13] with a spring constant comparable to the NOMPC was localized to the distal tips of mechanosensory cilia gating spring of hair cells (1 mN/m) [14]. Thus, the ankyrin re- in sensory neurons [6–8] and the C. elegans NOMPC orthologue peats of TRPN1 may transmit mechanical forces to the channel. has been shown to be a pore forming channel [3]. Here we show that in vitro the terminal ANK repeats of xlTRPN1 TRPN1 is also present in mechanosensory hair cells of lower interact with the Penta-EF-hand protein peflin, a putative Ca2+ vertebrates, such as zebrafish [9] and the clawed frog, Xenopus lae- binding protein of unknown function. vis [10]. In zebrafish, morpholino-mediated removal of TRPN1 function eliminates microphonic potentials of hair cells and results 2. Materials and methods in larval deafness and imbalance [9]. Xenopus TRPN1 (xlTRPN1) localizes to the tips of kinocilia [10], which have recently been 2.1. DNA constructs shown to mediate mechanosensitivity in developing zebrafish hair cells [11], arguing for a role of TRPN1 in transduction also in lower For yeast two-hybrid assays, residues 1–209 and residues vertebrates. 1395–1521 of xlTRPN1 were amplified and cloned into the pGBKT7 GAL4 DNA-binding vector (Clontech, Palo Alto, CA, USA) to yield clones TRPN1_ANK1–6 and TRPN1_C, respectively. Peflin was Abbreviations: ANK, ankyrin; GST, glutathione-S-transferase; NHB, N-terminal cloned into the pACT2 GAL4 DNA-activating vector (Clontech). hydrophobic; PEF, penta-EF-hand; TRP, transient receptor potential For GST fusion proteins, xlTRPN1 fragments were cloned into ⇑ Corresponding author. Address: Department of Physiology, RWTH Aachen University, Pauwelsstrasse 30, D-52074 Aachen, Germany. Fax: +49 241 80 82434. pGEX-2TK (GE Healthcare). For fusions with GFP, xlTRPN1 and E-mail address: [email protected] (S. Gründer). peflin were cloned into pEGFP-C1 (Clontech) and for fusions with 1 These authors contributed equally to this work. mCherry, the sequence coding for GFP in TRPN1-pEGFP-C1 was

0014-5793/$36.00 Ó 2012 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.febslet.2012.10.046 D. Wiemuth et al. / FEBS Letters 586 (2012) 4276–4281 4277 replaced by the sequence coding for mCherry. For details, see Sup- plementary material online.

2.2. Yeast two-hybrid assay

The yeast two-hybrid assay was conducted using the MATCH- MAKER GAL4 Two-Hybrid System 3 and a Xenopus laevis oocyte Matchmaker cDNA library (Clontech), according to the manufacturer’s instructions. For details, see Supplementary material online.

2.3. GST fusion protein binding assays

GST fusion protein binding assays were conducted as described previously [15]. For Western Blot analysis, an anti-GFP antibody (1:2500) (Sigma–Aldrich, Deisenhofen, Germany) and an HRP- coupled anti-mouse antibody (1:5000) (Sigma–Aldrich) were used. For details, see Supplementary material online.

2.4. Cell culture, transfection and immunoﬂuorescence microscopy

COS-7 cells were maintained and transfected as described previously [15]. LLC–PK1–CL4 cells (CL4 cells; kindly provided by Dr. Jim R. Bartles, Department of Cell and Molecular Biology, North- western University, Chicago, Illinois, USA) were grown in Minimal Essential Medium Alpha Medium (Invitrogen), supplemented with 10% bovine serum albumin and 5% penicillin/streptomycin. Cells were transfected with 1.5 lg of each cDNA construct using Lipo- Fig. 1. TRPN1 fragments interact with peflin in a yeast two-hybrid assay. (A) fectamine 2000™ (Invitrogen) according to the manufacturer’s Schematic topology of xTPRN1 with the 28 ANK repeats depicted as grey ellipses. protocol. For details on immunofluorescence microscopy, see Sup- ANK1-6 and the C-terminal tail that were used as baits in (B) are highlighted by plementary material online. dashed squares. (B) In yeast, ANK1-6 (left two quadrants) and the C-terminal tail of xlTRPN1 (right two quadrants) interact with peflin (upper two quadrants), but not with syntaxin (lower two quadrants). Depicted is growth of co-transformed yeast À À À 3. Results and discussion on selective plates (His /Leu /Trp ).

3.1. TRPN1 interacts with the Penta-EF-hand protein peﬂin

Using a yeast two-hybrid assay, we screened a Xenopus laevis cDNA library with the six N-terminal ankyrin (ANK) repeats of TRPN1 as bait (residues 1–209; Fig. 1A) and isolated a full-length sequence of the Xenopus orthologue of peflin (penta-EF-hand protein with long N-terminal hydrophobic domain) as a putative interaction partner (Fig. 1B). Peflin is a 30 kDa protein with unknown function [16]. It belongs to the penta-EF-hand (PEF) protein family, members of which contain five EF-hands in their C-terminal domain and an N-terminal hydrophobic domain (NHB domain) [17] (Fig. 2A). Peflin has an unusually long NHB domain that contains nine repeats of a nonapeptide (consensus sequence: A/ PPGGPYGGP) [16]. The closest relative of peflin within the PEF family is ALG-2 (apoptosis-linked gene-2), which has recently been shown to interact with the TRP channel TRPML1 [18], a TRP channel expressed in lysosomes and mutated in mucolipidosis type IV. In contrast to TRPN1, TRPML1 lacks ANK repeats and interacts via a patch of charged residues at its cytoplasmic N-terminal tail with ALG-2 [18]. We then asked whether peflin also interacts with the C-terminal tail of TRPN1. This was indeed the case, as peflin also interacted with the cytoplasmic C-terminal tail of TRPN1 in the yeast two- Fig. 2. The N-terminal ankyrin repeats of xlTRPN1 interact with peflin in GST hybrid assay (Fig. 1B). This situation is reminiscent of TRPV1 and pulldown assays. (A) Schematic representation of peflin. The nine nonapeptide TRPV4, TRP channels that also interact with both their N-terminal repeats in the N-terminal NHB domain are depicted as light grey boxes, the 2+ consensus loops of the five EF hands in the C-terminal PEF domain as dark grey ANK repeats and their C-terminal tail domains with the Ca - boxes. (B) The expression of the GFP-peflin fusion (calculated molecular mass, binding protein calmodulin [19–22]. 57 kDa) is shown in the upper panel and represents 5% of the input used for the We next confirmed the interaction of TRPN1 with peflin in GST pulldown assays (lanes 1–6). Cell lysates were incubated with glutathione- pulldown assays. The six N-terminal ANK repeats (ANK1–6; resi- Sepharose beads carrying GST (lane 1), GST-TRPN1_C-trunc (lane 2), GST- TRPN1_ANK1–2 (lane 3), GST-TRPN1_ANK3–6 (lane 4), or GST-TRPN1_ANK1–6 dues 1–209), corresponding to the bait used in the initial yeast (lane 5) and bound proteins analyzed by Western blotting using an anti-GFP two-hybrid screen, and the carboxyl-terminal cytoplasmic tail antibody. None of the GST fusion proteins interacted with GFP alone (calculated (residues 1395–1521) were synthesized in bacteria as GST fusion molecular mass, 27 kDa); shown is an example for GST-TRPN1_ANK1–2 (lane 6). 4278 D. Wiemuth et al. / FEBS Letters 586 (2012) 4276–4281 proteins. Both fusion proteins were relatively insoluble and were We then performed GST pulldown assays with ANK1–2 and the recovered only in small amounts. We therefore created additional fragments and the Ca2+-binding mutant of peflin fused to GFP. fusions of smaller TRPN1 fragments; we fused ANK1–2 (residues Fig. 3A shows one example of such an experiment and a quantita- 1–73), ANK3–6 (residues 71–209), and a truncated fragment of tive analysis of a total of five similar experiments, in which the the C-terminal tail (residues 1448–1521) to GST. These shorter amount of interacting protein was normalized to full-length wild- proteins were still not readily soluble but better recovered from type peflin. While ANK1–2 pulled down the Ca2+-binding mutant bacteria than ANK1–6 and the complete C-terminal tail and all truncated forms of peflin in significantly (P < 0.01) smaller (Supplementary Fig. S1). Following purification, GST fusion amounts than full-length wildtype peflin (Fig. 3), the interaction proteins were incubated with cell lysates containing peflin fused was almost abolished for DNHB and DNHBEF1 + 2, lacking the to the C-terminus of GFP. GST and GFP alone were used as negative NHB domain (<10% of protein pulled down in comparison to wt controls. These pulldown assays revealed that ANK1–6, ANK1–2, peflin; P << 0.01). These results suggest that peflin has several ANK3–6, and the truncated carboxyl-terminal tail specifically interaction sites with the ANK repeats of TRPN1 and that the interacted with peflin (Fig. 2B), confirming the interaction NHB domain and EF5, which mediates dimerization, are especially between TRPN1 and peflin. Moreover, these results indicate that important for binding ANK repeats. The interaction of ANK1–2 several ANK repeats in the N-terminal tail of TRPN1 interact with with the Ca2+ binding mutant of peflin was also strongly reduced peflin. (31 ± 9% of protein pulled down in comparison to wt peflin; P < 0.01), suggesting that the ability to bind Ca2+ is also important 3.2. The NHB domain of peflin and the ability to bind Ca2+ are crucial for binding TRPN1 ANK repeats. for the interaction with TRPN1 We did similar pulldown assays with the C-terminal tail of TRPN1. As shown in Fig. 3B, the C-terminal tail pulled down the We next wanted to identify the domains of peflin that mediate Ca2+-binding mutant of peflin equally well as wt peflin and it the interaction with TRPN1. The Ca2+-binding protein calmodulin pulled down DNHB even better than wt (P < 0.01). In contrast, has four EF-hands, EF1–EF2 and EF3–EF4, which form spatially sep- the three truncations lacking one or several EF-hands were pulled arated subdomains, the N- and C-lobes, respectively, that function down in significantly (P < 0.05) smaller amounts than wt peflin (almost) independently [23]. Compared to calmodulin, PEF pro- (Fig. 3B). This reduction was especially prominent for DEF5 and teins have seven-residue-deletions between EF-2s and EF-3s in a DEF3–5 (<20% of protein pulled down in comparison to wt peflin; region that links N- and C-lobes in calmodulin, indicating that P << 0.01), suggesting that EF5 and the ability of peflin to dimerize the two lobes may not be independent in peflin [16]. The additional is particularly important for its interaction with the C-terminal tail EF hand of peflin, EF5, is unpaired and does not bind Ca2+ but medi- of TRPN1. These results suggest that different surfaces of peflin ates heterodimerization with ALG-2 [24]. Therefore, we created interact with the ANK repeats and the C-terminal tail. While the peflin fragments, in which we deleted 1) EF5 (DEF5), 2) the NHB presence of the NHB domain is important for the interaction with domain (DNHB), 3) EF3–EF5 (DEF3–5), or 4) the NHB domain to- ANK1–2, it is dispensable and may even hinder the interaction gether with NHB/EF1-2 (DNHB/EF1-2Fig. 3, lanes 3 to 6, respec- with the C-terminal tail. tively). Moreover, we asked whether the ability of peflin to bind Ca2+ is important for the interaction with TRPN1. It is unclear 3.3. Peflin co-localizes with TRPN1 in a hair cell model whether peflin indeed binds Ca2+ but peflin translocates to mem- branes in the presence of Ca2+ [16,25] and heterodimerization with Next, we investigated interaction of TRPN1 and peflin in CL4 ALG-2 occurs only in the absence of Ca2+, suggesting peflin does cells, an epithelial cell model in which certain hair cell plasma bind Ca2+. EF-hands 1 and 3 of peflin best conform to the EF-hand membrane proteins are more efficiently targeted to the plasma consensus sequence [16] and both contain a conserved glutamate membrane than in other cell lines [27]. When expressed in CL4 residue (E137 and E204, respectively) at position –z that is critical cells, mCherry-TRPN1 localized intracellularly in a reticular pattern for Ca2+ coordination [26]. We therefore created a mutant of peflin (Fig. 4A). Staining of TRPN1-transfected CL4 cells with a polyclonal in which we substituted E137 and E204 by alanine (Fig. 3, lane 2). anti-TRPN1 antibody [10] yielded a similar pattern (not shown),

Fig. 3. Truncations of peflin and mutations in EF hands strongly decrease interaction with the ankyrin repeats and the C-terminal domain of xlTRPN1. (A) Left, schematic diagram of peflin and its truncations used in GST pulldown assays. Asterisks indicate the E-to-A exchanges in the first and third EF hands and numbers indicate the amino acids of peflin present in the constructs, respectively. All constructs were used as C-terminal fusions to GFP (calculated molecular mass, 39–57 kDa). Middle, lysates from cells transiently expressing GFP-peflin, truncations and the Ca2+-binding mutation of peflin were incubated with glutathione-Sepharose beads carrying GST-TRPN1_ANK1–2 and bound proteins analyzed by Western blotting using an anti-GFP antibody. The expression of the proteins is shown in the left panel and represents 5% of the input used for the pulldown assay. Right, scatter dot plot illustrating quantification of five similar GST pulldown experiments; each dot represents one experiment, bars indicate mean and S.E.M., respectively; n = 5. For quantification, the ratios of the intensities of bands in the pulldown and the input were calculated and normalized to these ratios for wt-peflin. (B) as in (A) but with glutathione-Sepharose beads carrying GST-TRPN1_C-trunc. n =5.⁄, P < 0.05, ⁄⁄, P < 0.01; ⁄⁄⁄, P < 0.001 (Student’s t test). D. Wiemuth et al. / FEBS Letters 586 (2012) 4276–4281 4279

A A Peflin TRPN1 Peflin TRPN1_ANK1-2

B B Peflin TRPN1 Merge Peflin TRPN1_ANK1-2 MERGE

C Peflin ΔANK1-6 Merge Fig. 5. TRPN1_ANK1–2 and peflin co-localize in CL4 cells. (A) Subcellular distribution of RFP-ANK1–2 and GFP-peflin in CL4 cells shown by confocal microscopy. (B) When coexpressed, GFP-peflin redistributed RFP-ANK1–2 to a punctate pattern. RFP-ANK1–2 is in red and GFP-peflin in green. Scale bars, 10 lm.

respectively, revealed predominant expression of mCherry-TRPN1 Peflin ΔC Merge in the ER (Supplementary Fig. S2). GFP-peflin had a different localization than mCherry-TRPN1: fluorescence was mainly associated with a few large intracellular punctae (Fig. 4A). Co-localization studies revealed no co-localization with marker proteins for the ER and different vesicles (Supplementary Fig. S2), suggesting that the punctae after overex- pression of peflin originate from aggregates in the cytoplasm. Peflin ΔANK1-6/C Merge Expression of the peflin fragments in CL4 cells revealed that they formed similar punctae as wt peflin with the exception of DNHB and DNHBEF1–2, which were diffusely distributed in the cytoplasm and the nucleus (Fig. 6A), suggesting that the NHB domain mediates aggregation of peflin, presumably via hydrophobic interactions. When co-expressed with mCherry-TRPN1, GFP-peflin changed its distribution and no longer localized in a punctate pattern but in a reticular pattern, similar to TRPN1 (Fig. 4B). This mCherry- 100 D TRPN1-dependent distribution of GFP-peflin overlapped with the PDI staining (Supplementary Fig. S2C), suggesting that in the presence of TRPN1 GFP-peflin was associated with the ER. Thus, co- 50 expression of TRPN1 and peflin revealed that both proteins tightly interacted in CL4 cells. When we co-expressed deletion mutants of TRPN1 that either truncated ANK1–6 (DANK1–6) or the C-terminal

% colocalization tail (DC) or both (DANK1–6/C), peflin was no longer redistributed and localized in its typical punctate pattern (Fig. 4C), showing that C both ANK1–6 and the C-terminal domain are necessary for the Δ tight interaction of peflin with TRPN1 in living cells. TRPN1 ANK1-6 Δ These results contrast with the GST pulldown assays where pef- ANK1-6/C Δ lin interacted with ANK1–2 and the C-terminal tail in the absence of the other domain. Therefore, we asked whether peflin interacts Fig. 4. TRPN1 and peflin co-localize in CL4 cells. (A) Subcellular distribution of GFP- peflin and mCherry-TRPN1 in CL4 cells shown by confocal microscopy. (B) When with the ANK repeats of TRPN1 also in CL4 cells and fused ANK1–2 co-expressed, mCherry-TRPN1 redistributed GFP-peflin from the typical punctate to RFP. RFP-ANK1–2 was diffusely distributed in the cytoplasm of pattern to the TRPN1-like pattern. (C) GFP-peflin did not colocalize with deletions of CL4 cells (Fig. 5A). After co-expression of RFP-ANK1–2 with GFP- TRPN1, which either truncated ANK1–6 or the C-terminal tail (residues 1395-1521) peflin, RFP-ANK1–2 adopted the staining pattern of GFP-peflin: in or both. mCherry-TRPN1 is in red and GFP-peflin in green. Scale bars, 10 lm. (D) Quantification of the co-localization of GFP-peflin with mCherry-TRPN1 and its addition to the diffuse distribution, RFP fluorescence co-localized truncations. Quantification was done using the co-localization tool of the ZEN with the GFP fluorescence to large intracellular punctae. This result software (Carl Zeiss, Jena, Germany). ⁄⁄⁄, P < 0.001 (Student’s t test), n = 18. suggests that peflin interacts with the isolated two N-terminal ANK repeats of TRPN1 also in CL4 cells. Together these results suggest that ANK1–2 tightly interact with peflin but when fused to the indicating that the fusion to mCherry did not change the distribu- TRPN1 backbone this interaction needs the support of the tion of TRPN1. Co-localization studies with protein disulfide isom- C-terminal tail. Similarly, the interaction of peflin with the erase (PDI) and giantin, marker proteins for the ER and the Golgi, C-terminus fused to the TRPN1 backbone needs the support of 4280 D. Wiemuth et al. / FEBS Letters 586 (2012) 4276–4281

A B

Peflin Ca2+ Peflin Ca2+ TRPN1 Merge

ΔEF5 ΔEF5 TRPN1 Merge

ΔNHB ΔNHB TRPN1 Merge

ΔEF3-5 ΔEF3-5 TRPN1MERGE MergeMERGE

ΔNHB/EF1-2 ΔNHB/EF1-2 TRPN1 Merge

C 100 n.s. n.s.

50 % colocalization

2+ EF5 Δ NHB Peflin Δ EF3-5 Δ

Peflin Ca NHB/EF1-2 Δ

Fig. 6. Truncations of peflin and mutations in EF hands strongly decrease co-localization with TRPN1 in CL4 cells. (A) Subcellular distribution of truncations of peflin and mutations in EF hands in CL4 cells shown by confocal microscopy. (B) When co-expressed, truncations of peflin and mutations in EF hands did not co-localize with TRPN1 in CL4 cells. Note that the peflin fragments lacking the NHB domain, DNHB and DNHB/EF1–2, localize diffusely in the cytoplasm in the presence and absence of TRPN1. GFP- peflin is in green and mCherry-TRPN1 in red. Scale bars, 10 lm. (C) Quantification of the co-localization of mCherry-TRPN1 with GFP-peflin and its mutants. Quantification was done using the co-localization tool of the ZEN software (Carl Zeiss). ⁄⁄⁄, P < 0.001 (Student’s t test), n = 18. Note that co-localization of DNHB and DNHB/EF1–2 was not significantly reduced; these two truncations localized diffusely in CL4 cells, however.

the N-terminal ANK repeats. In agreement, neither of the peflin evidence suggests that peflin binds Ca2+ [16,25], the impact of pef- fragments was redistributed by TRPN1 in CL4 cells, as expected lin on TRPN1 may relate to its capacity to bind Ca2+. The transduc- from the loss of their interaction with ANK1–2 in pulldown assays tion channel of hair cells is Ca2+-permeable and Ca2+ contributes to (Fig. 6B and C). transduction channel adaptation. Fast adaptation is thought to be Because the function of peflin is unknown and xlTRPN1 has so initiated by Ca2+-binding to the transduction channel itself or a far not been functionally expressed, we can only speculate on the protein in direct contact with it. Peflin is a candidate for a Ca2+- possible impact of its interaction with TRPN1. Since preliminary binding protein in direct contact with xlTRPN1, an ortholog of D. Wiemuth et al. / FEBS Letters 586 (2012) 4276–4281 4281 the fly transduction channel, potentially implicating peflin in fast [10] Shin, J.B., Adams, D., Paukert, M., Siba, M., Sidi, S., Levin, M., Gillespie, P.G. and transducer channel adaptation. Gründer, S. (2005) Xenopus TRPN1 (NOMPC) localizes to microtubule-based cilia in epithelial cells, including inner-ear hair cells. Proc. Natl. Acad. Sci. USA In conclusion, we have identified peflin as a protein interacting 102, 12572–12577. in vitro with TRPN1. Peflin may thus be a newly identified compo- [11] Kindt, K.S., Finch, G. and Nicolson, T. (2012) Kinocilia mediate nent of the transduction apparatus of hair cells of lower mechanosensitivity in developing zebrafish hair cells. Dev. Cell 23, 329–341. [12] Michaely, P., Tomchick, D.R., Machius, M. and Anderson, R.G. (2002) Crystal vertebrates. structure of a 12 ANK repeat stack from human ankyrinR. EMBO J. 21, 6387– 6396. Acknowledgements [13] Lee, G., Abdi, K., Jiang, Y., Michaely, P., Bennett, V. and Marszalek, P.E. (2006) Nanospring behaviour of ankyrin repeats. Nature 440, 246–249. [14] Howard, J. and Hudspeth, A.J. (1988) Compliance of the hair bundle associated We thank M. Wirtz and A. Oslender-Bujotzek for expert techni- with gating of mechanoelectrical transduction channels in the bullfrog’s cal assistance, I. Kadurin for advice on the yeast two-hybrid assay, saccular hair cell. Neuron 1, 189–199. [15] Wiemuth, D., Lott, J.S., Ly, K., Ke, Y., Teesdale-Spittle, P., Snyder, P.M. and O. Staub for the gift of the Xenopus oocytes cDNA library, and J.R. McDonald, F.J. (2010) Interaction of serum- and glucocorticoid regulated Bartles for the gift of LLC-PK1-CL4 cells. This work was supported kinase 1 (SGK1) with the WW-domains of Nedd4-2 is required for epithelial by a Grant of the Deutsche Forschungsgemeinschaft to S.G. (Grant sodium channel regulation. PLoS ONE 5, e12163. # GR 1771/6-1). [16] Kitaura, Y., Watanabe, M., Satoh, H., Kawai, T., Hitomi, K. and Maki, M. (1999) Peflin, a novel member of the five-EF-hand-protein family, is similar to the apoptosis-linked gene 2 (ALG-2) protein but possesses non-apeptide repeats Appendix A. Supplementary data in the N-terminal hydrophobic region. Biochem. Biophys. Res. Commun. 263, 68–75. [17] Maki, M., Kitaura, Y., Satoh, H., Ohkouchi, S. and Shibata, H. (2002) Structures, Supplementary data associated with this article can be found, functions and molecular evolution of the penta-EF-hand Ca2+-binding in the online version, at http://dx.doi.org/10.1016/j.febslet.2012. proteins. Biochim. Biophys. Acta 1600, 51–60. [18] Vergarajauregui, S., Martina, J.A. and Puertollano, R. (2009) Identification of 10.046. the penta-EF-hand protein ALG-2 as a Ca2+-dependent interactor of mucolipin-1. J. Biol. Chem. 284, 36357–36366. References [19] Numazaki, M., Tominaga, T., Takeuchi, K., Murayama, N., Toyooka, H. and Tominaga, M. (2003) Structural determinant of TRPV1 desensitization interacts with calmodulin. Proc. Natl. Acad. Sci. USA 100, 8002–8006. [1] Effertz, T., Wiek, R. and Göpfert, M.C. (2011) NompC TRP channel is essential [20] Strotmann, R., Schultz, G. and Plant, T.D. (2003) Ca2+-dependent potentiation for Drosophila sound receptor function. Curr. Biol. 21, 592–597. of the non-selective cation channel TRPV4 is mediated by a C-terminal [2] Effertz, T., Nadrowski, B., Piepenbrock, D., Albert, J.T. and Göpfert, M.C. (2012) calmodulin binding site. J. Biol. Chem. 278, 26541–26549. Direct gating and mechanical integrity of Drosophila auditory transducers [21] Lishko, P.V., Procko, E., Jin, X., Phelps, C.B. and Gaudet, R. (2007) The ankyrin require TRPN1. Nat. Neurosci. 15, 1198–1200. repeats of TRPV1 bind multiple ligands and modulate channel sensitivity. [3] Kang, L., Gao, J., Schafer, W.R., Xie, Z. and Xu, X.Z. (2010) C. elegans TRP family Neuron 54, 905–918. protein TRP-4 is a pore-forming subunit of a native mechanotransduction [22] Phelps, C.B., Wang, R.R., Choo, S.S. and Gaudet, R. (2010) Differential regulation channel. Neuron 67, 381–391. of TRPV1, TRPV3, and TRPV4 sensitivity through a conserved binding site on [4] Kernan, M., Cowan, D. and Zuker, C. (1994) Genetic dissection of the ankyrin repeat domain. J. Biol. Chem. 285, 731–740. mechanosensory transduction: mechanoreception-defective mutations of [23] Kretsinger, R.H. (1997) EF-hands embrace. Nat. Struct. Biol. 4, 514–516. Drosophila. Neuron 12, 1195–1206. [24] Kitaura, Y., Satoh, H., Takahashi, H., Shibata, H. and Maki, M. (2002) Both ALG- [5] Walker, R.G., Willingham, A.T. and Zuker, C.S. (2000) A Drosophila 2 and peflin, penta-EF-hand (PEF) proteins, are stabilized by dimerization mechanosensory transduction channel. Science 287, 2229–2234. through their fifth EF-hand regions. Arch. Biochem. Biophys. 399, 12–18. [6] Lee, J., Moon, S., Cha, Y. and Chung, Y.D. (2010) Drosophila TRPN(=NOMPC) [25] Kitaura, Y., Matsumoto, S., Satoh, H., Hitomi, K. and Maki, M. (2001) Peflin and channel localizes to the distal end of mechanosensory cilia. PLoS ONE 5, ALG-2, members of the penta-EF-hand protein family, form a heterodimer that e11012. dissociates in a Ca2+-dependent manner. J. Biol. Chem. 276, 14053–14058. [7] Cheng, L.E., Song, W., Looger, L.L., Jan, L.Y. and Jan, Y.N. (2010) The role of the [26] Babu, A., Su, H., Ryu, Y. and Gulati, J. (1992) Determination of residue TRP channel NompC in Drosophila larval and adult locomotion. Neuron 67, specificity in the EF-hand of troponin C for Ca2+ coordination, by genetic 373–380. engineering. J. Biol. Chem. 267, 15469–15474. [8] Liang, X., Madrid, J., Saleh, H.S. and Howard, J. (2011) NOMPC, a member of the [27] Zheng, L., Zheng, J., Whitlon, D.S., Garcia-Anoveros, J. and Bartles, J.R. (2010) TRP channel family, localizes to the tubular body and distal cilium of Targeting of the hair cell proteins cadherin 23, harmonin, myosin XVa, espin, Drosophila campaniform and chordotonal receptor cells. Cytoskeleton and prestin in an epithelial cell model. J. Neurosci. 30, 7187–7201. (Hoboken) 68, 1–7. [9] Sidi, S., Friedrich, R.W. and Nicolson, T. (2003) NompC TRP channel required for vertebrate sensory hair cell mechanotransduction. Science 301, 96–99.