Cloning and Functional Expression of Human Short TRP7, a Candidate Protein for Store-operated Ca2+ Influx.

The regulation and control of plasma membrane Ca2+ fluxes is critical for the initiation and maintenance of a variety of signal transduction cascades. Recently, the study of transient receptor potential channels (TRPs) has suggested that these proteins have an important role to play in mediating capacitative calcium entry. In this study, we have isolated a cDNA from human brain that encodes a novel transient receptor potential channel termed human TRP7 (hTRP7). hTRP7 is a member of the short TRP channel family and is 98% homologous to mouse TRP7 (mTRP7). At the mRNA level hTRP7 was widely expressed in tissues of the central nervous system, as well as some peripheral tissues such as pituitary gland and kidney. However, in contrast to mTRP7, which is highly expressed in heart and lung, hTRP7 was undetectable in these tissues. For functional analysis, we heterologously expressed hTRP7 cDNA in an human embryonic kidney cell line. In comparison with untransfected cells depletion of intracellular calcium stores in hTRP7 expressing cells, using either carbachol or thapsigargin, produced a marked increase in the subsequent level of Ca2+ influx . This increased Ca2+ entry was blocked by inhibitors of capacitative calcium entry such as La3+ and Gd3+. Furthermore, transient transfection of an hTRP7 anti-sense expression construct into cells expressing hTRP7, eliminated the augmented store operated Ca2+ entry. Our findings suggest that hTRP7 is a store-operated calcium channel, a finding in stark contrast to the mouse orthologue, mTRP7, which is reported to enhance Ca2+ influx independently of store depletion, and suggests that and mouse TRP7 channels may fulfil different physiological roles.

enhance Ca 2+ influx independently of store depletion, and suggests that human and mouse TRP7 channels may fulfil different physiological roles.

Introduction
Numerous G-protein-coupled receptors signal through the phospholipase C pathway to liberate intracellular Ca 2+ stores and elicit the activation of a variety of cellular signal transduction pathways. Intrinsic to the integrity of this system is the process of refilling of these Ca 2+ stores which requires a regulated Ca 2+ influx across the plasma membrane. Two broad mechanisms appear to be responsible for activating this Ca 2+ influx and consequent store refilling. In a number of systems, store depletion itself seems to activate refilling. This can be experimentally demonstrated by passively depleting Ca 2+ stores with Ca 2+ ATPase inhibitors such as thapsigargin 1;2 . As no obvious chemical messenger is involved this mechanism it has been dubbed capacitative Ca 2+ entry (CCE) or store operated Ca 2+ (SOC) entry. The second key mechanism operates independently of Ca 2+ release from intracellular stores, whereby downstream products of the phospholipase Cβ cascade directly activate Ca 2+ influx [3][4][5][6][7] .
The initial leads in the hunt to identify the molecular basis of the store refilling pathway were provided by study of phototransduction in Drosophila melanogaster. In photoreceptor cells three related genes have been isolated whose products have been reported to function as calcium permeable channels: the transient receptor potential (trp) gene 8;9 , the transient receptor potential-like (trpl) 4 gene 10 and the transient receptor potential gamma (trpγ) gene 11 . Expression of trp in a baculovirus system produced currents when intracellular Ca 2+ stores were depleted with thapsigargin 12;13 , a result indicative of a role as a store-operated channel mediating CCE. In contrast however, heterologous expression of trpl and trpγ resulted in currents that were constitutively active irrespective of Ca 2+ store status and therefore were not classified as store operated channels 10;14;15 .
Thus far, seven short mammalian cDNA homologues of drosophila TRP channels have been identified, termed TRP1-7 [16][17][18][19] . A common characteristic of these channels is that they are Ca 2+ -permeable, non-selective cation channels. Analysis of the channel activation properties of some mammalian TRP channels has lead to different results being obtained dependent on the expression system used. TRP1, however, clearly produced a current activated by Ca 2+ store depletion both in recombinant 20 and native systems. Less is known about the TRP2 gene product; although it has also been reported to have store-operated properties 21 , the human orthologue of TRP2 appears to be a pseudogene 22 . TRP4 and 5 have been implicated in both receptor-and store-operated channel function depending on the cell line in which they are expressed and the source of the channel (i.e. human or mouse) [23][24][25][26][27] . Genetic deletion 28 or anti-sense knockdown 29

Molecular Cloning and Sequence Analysis of hTRP7
As part of a genomics-based programme to identify novel ion channels we Analysis of hTRP7 using the TMPRED algorithm 40 that predicts transmembrane domains (TMs), predicted a protein having six TMs and a pore region between TM5 and TM6, similar to those predicted for other TRP subtypes (Fig. 1B). The predicted protein sequence of hTRP7, when aligned with other members of TRP families, revealed that hTRP7 shares 98% overall identity with the mouse TRP7 protein, and 81% and 75% overall identity with hTRP3 and hTRP6 respectively.
With regard to Drosophila, the most closely related TRP channel is TRPγ, with which hTRP7 shares 41% overall identity. The overall identity with both TRP and TRPL is about 38%. Taken together with the lack of an hydrophobic N-terminal signal sequence, this suggests that hTRP7 is a membrane protein with a core of six transmembrane segments and a pore region, with intracellular flanking N-and C-terminal domains 41;42 . Fig. 1C depicts a dendrogram of short TRP channels, constructed using the CLUSTAL W program 43 . It is clear that hTRP7 is a member of the short transient receptor potential channel family and falls into the TRP3/6/7 subfamily of short TRP channels.
The genomic structure of the hTRP7 gene was determined by comparison of the cDNA sequence with the genomic sequences (AC008661, AC026299 and AC063980; Table 1). hTRP7 gene appears to consist of 12 exons, although exon 6 is inferred from comparison with the genomic structures of TRP3 and TRP6 30 and exon 1 is inferred from an alignment of the longer public mouse TRP7 sequence (AF139923) to the human genomic sequence (AC063980). The genomic sequences of human TRP7 contain a number of STS markers which have been mapped to the chromosomal region 5q31 (Ensembl ContigView and NCBI Map Viewer). One of these markers is the microsatellite marker locus D5S393.
BLAST searching with the nucleotide sequence of D5S393 (Z16468) against the genomic sequence AC063980, indicates that it is situated ~300 bp upstream of the region predicted to be exon 1 of hTRP7. Thus, D5S393 provides a useful genetic marker for screening hTRP7 as a candidate gene for genetic diseases mapping to this chromosomal region.

Analysis of hTRP7 mRNA Expression in Human Tissues and Cell Lines
An extensive study of mRNA distribution was carried out in human CNS regions, peripheral tissues and a panel of cell lines using the TaqMan real-time RT-PCR technique 36;38;39 . These studies indicate that hTRP7 is widely expressed in the CNS with the highest levels present in the nucleus accumbens and somewhat lower expression in the putamen, striatum, hypothalamus, caudate nucleus, locus coeruleus and medulla oblongata. Lower levels of mRNA expression were observed in all other CNS regions studied ( Fig. 2A). In peripheral tissues a high level of hTRP7 expression was detected in pituitary gland and kidney, with lower levels found in intestine, prostate and cartilage ( Fig. 2A).
In order to identify which cell line would be most suitable for over-expression of hTRP7, we analysed the expression of hTRP7 mRNA in a panel of cell lines. The highest level of expression was detected in the monkey kidney fibroblast cell lines COS-1 and COS-7, with lower expression in human kidney HK-2 cells (Fig. 2B).
No mRNA expression was detected in the other cell lines examined, including HEK-293 cells.

Expression and localization of hTRP7 in HEK-293 cells
To perform a functional characterisation of hTRP7, a HEK-293 cell-derived stable line was generated expressing an N-terminus FLAG-tagged hTRP7 cDNA.
Since HEK-293 cells lack detectable mRNA expression of endogenous hTRP7 ( Fig 2B), they are ideal for studying the functional properties of this protein 27;44 .
TaqMan real time RT-PCR performed on HEK-293 clones expressing hTRP7 showed high levels of hTRP7 mRNA expression (data not shown).
The cellular localisation of the FLAG-tagged hTRP7 protein was examined with standard immunofluorescence techniques. Fig. 2C shows a confocal image of cells stably transfected with hTRP7 and exposed to an anti-FLAG antibody.
Consistent with the data obtained from TaqMan analysis, considerable immuno- being an integral membrane protein.

Functional characterisation of hTRP7
To investigate the functional role of hTRP7, carbachol (CCh) was used to activate the phospholipase C pathway in the hTRP7 cell line via stimulation of the endogenous muscarinic acetylcholine receptor 46 (Fig. 4A). There was no significant difference in this initial increase between control and hTRP7-expressing cells.
When These data demonstrate an association between hTRP7 and SOC activity, and show that hTRP7 is responsible for enhancement of CCE through a receptorindependent store depletion-induced pathway.

Effect of Mn 2+ in hTRP7-transfected cells
Mn 2+ a quencher of Fura-2 fluorescence 48 can enter cells through certain types of Ca 2+ influx channels 49 . This property can thus be used to directly monitor cation influx through such channels. In the presence of 1 mM Mn 2+ , basal Mn 2+ influx in hTRP7-transfected cells was faster than in control cells (Fig. 5A). Mn 2+ entry was also determined during CCh-stimulated Ca 2+ entry by adding 1mM Mn 2+ 30 sec before depletion of the intracellular Ca 2+ stores with CCh (Fig. 5B). CCh stimulation increased Mn 2+ entry in both control and hTRP7 expressing cells, however a higher rate of entry was in the hTRP7 cells. Therefore, hTRP7 expression increases resting Mn 2+ entry, suggesting that these channels may exhibit a degree of constitutive activity.

Effect of Ca 2+ channel blockers in hTRP7-transfected cells
In addition we investigated the ability of the lanthanide ions gadolinium (Gd 3+ ) and lanthanum (La 3+ ), which are both non-specific Ca 2+ channel blockers, as Ca 2+ entry in hTRP7-transfected cells (Fig. 6C). These data indicate that hTRP7 may be directly involved in mediating Ca 2+ entry and might be a useful tool in identifying the type of Ca 2+ -influx channel present 45 .

Expression of anti-sense hTRP7 cDNA in cells expressing hTRP7 decreases the SOC activity
To provide further evidence that hTRP7 was involved in a SOC entry mechanism, the hTRP7 stable cell line was transiently transfected with hTRP7 cDNA in the anti-sense orientation 26 which was used to monitor [Ca 2+ ] i (Fig. 7A). To determine the effect of the hTRP7 anti-sense construct the level of Ca 2+ entry was compared between cells expressing and lacking RFP fluorescence. Thapsigargin stimulated internal Ca 2+ release was not altered in cells transfected with the anti-sense construct (Fig. 7B).
However, the thapsigargin stimulated Ca 2+ influx (second peak of [Ca 2+ ] i increase) was significantly reduced compared with that observed in hTRP7 cells not expressing the anti-sense construct, and was similar to that obtained in HEK-293. (Fig. 7B). Therefore, transient expression of hTRP7 anti-sense cDNA in the hTRP7 cell line decreases the expression of hTRP7 mRNA and reduces the functional activation of SOC entry by thapsigargin.

Discussion
In this report, we describe the cloning of a novel human short TRP channel, hTRP7, that shows a high sequence identity to mouse TRP7 31 , and substantial homology to human TRP3 52 and TRP6 30 . Sequence analysis of hTRP7 revealed an amino acid change from the GenBank sequence, with a proline replacing a leucine at position 111, that was not a result of a cloning artefact. This residue is located in a region highly conserved among the short TRP family proteins and is C-terminal to the second predicted ankyrin domain. It has been hypothesised that TRP interactions with the cytoskeleton may be mediated through one or more of these ~33-amino acid ankyrin repeats 19 and that alteration or stabilisation of such interactions may regulate the function of a variety of ion channels [53][54][55][56] . It is Future studies using mutagenesis, functional analysis and yeast two hybrid screening, will help to address the importance of this amino acid substitution.
mRNA distribution analysis of hTRP7 demonstrated another important difference between the human and mouse orthologues of this channel. hTRP7 was widely expressed in tissues of the CNS, with a more restricted pattern of expression in peripheral tissues. In contrast, the mouse homologue appears poorly expressed in the CNS at the mRNA level, with only low levels of expression detected in the hind brain (predominantly Purkinje cells of the cerebellum) 31 . Peripheral expression of mTRP7 is also different from that of hTRP7, with highest levels of mRNA observed in heart and lung but with little or no expression in the kidney 31 .
These differences suggest that mouse and human TRP7 channels may have quite distinct physiological roles despite their high sequence homology.
Functionally hTRP7 was also demonstrated to behave quite differently to mTRP7.
Following agonist (CCh) stimulation in the absence of Ca 2+ , Ca 2+ entry into hTRP7 transfected cells was markedly increased compared to control cells following re-addition of extracellular Ca 2+ . Furthermore, hTRP7 expressing cells displayed significantly increased thapsigargin stimulated Ca 2+ influx compared to control cells. This increased capacitative entry was abolished by transfection with an anti-sense hTRP7 expression construct. This data suggests following emptying of intracellular Ca 2+ stores, hTRP7 channel has a role in both receptor dependent CCE elicited by the agonist driven G q -coupled phospholipase C pathway, as well by a receptor independent store depletion induced pathway 21 . This is in clear contrast to the functional role previously described for mTRP7, which when expressed in a similar HEK-293 cell line, demonstrated receptor-activated behaviour independent of Ca 2+ store depletion 31 .
Despite the 98% protein identity between the two channels, sequence alignment of human and mouse TRP7 reveals differences that could potentially affect channel function. In particular, amino acid differences located in the C-terminal region could affect the interaction of the channels with components of signal transduction machinery such as calmodulin, Na + /H + exchanger regulatory factor (NHRF) and inositol 1,4,5 triphosphate receptors [61][62][63] . Studies with chimeric channels and mutagenesis experiments may help to clarify the role that these domains play in channel function.
It is important to note that these channels do not function in isolation. Thus the activity of a recombinantly expressed protein may also be influenced by the endogenous TRP channels as well as associated regulatory mechanisms present in the cells in which these channels are expressed. It has previously been reported that TRP channels can form both homomeric or heteromeric complexes, with either TRP proteins or other unidentified proteins. For example, interactions have been reported for hTRP1 with both hTRP3 64 and hTRP5 65 . In addition, heteromultimers have been reported between Drosophila TRP and TRPL 46 , and TRPL and TRPγ 11 . Such protein-protein interactions could certainly influence the characteristics and regulation of TRP channel activity in various cells. Clearly, further studies are required for a more complete understanding of TRP channel function and an important aspect of these studies will be to determine the presence and co-localisation of the short TRP proteins in different tissues and cell types.
In conclusion, we have identified hTRP7, the human orthologue of mTRP7 31 .
This channel appears to be functionally distinct from its rodent counterpart in terms of its distribution profile and its role in receptor-independent SOC entry.
These studies highlight the importance of investigating species orthologues and suggest that hTRP7 should be considered as functionally distinct to its closest relatives TRP3 and TRP6.