Dynamics and calcium sensitivity of the Ca2+/myristoyl switch protein hippocalcin in living cells

Hippocalcin signaling via site-specific translocation in hippocampal neurons

Hippocalcin is a Ca(2+)-binding protein, which belongs to the family of neuronal Ca(2+) sensors. It is highly expressed in the hippocampus but molecular mechanisms underlying its action in this part of the brain have not been investigated in detail. To study whether intrinsic neuronal activity could result in hippocalcin-mediated signal transduction we examined spontaneous and action potential (AP)-dependent changes in fluorescence of yellow fluorescent protein-tagged hippocalcin (HPCA-YFP) in transiently transfected hippocampal cultured neurons. In 6-12 DIV neurons HPCA-YFP spontaneously translocated longitudinally to specific sites within diffusionally confined domains of neuronal processes. The translocations to these sites were expressed as fast, reversible increases in HPCA-YFP fluorescence coincided with a decrease in adjacent sites indicating genuine protein translocation. Physiologically relevant neuronal stimulation with short trains of action potentials also resulted in fast, simultaneous, reversible, and [Ca(2+)](i)-dependent translocations of HPCA-YFP to several sites synchronizing hippocalcin signaling in different parts of neuronal processes. The amount of translocated protein increased with the number of action potentials in a train decoding the number of APs into the amount of translocated protein. We conclude that hippocalcin may signal within diffusionally restricted domains of neuronal processes in which it might play a physiological role in Ca(2+)-dependent local activation of specific molecular targets.

A novel role of hippocalcin in bFGF-induced neurite outgrowth of H19-7 cells

Hippocalcin is a Ca2+-binding protein that is expressed mainly in pyramidal nerve cells of the hippocampus. However, its functions and mechanism in the brain remain unclear. To elucidate the role of hippocalcin, we used a conditionally immortalized hippocampal cell line (H19-7) and showed that bFGF treatment increased the expression of hippocalcin during bFGF-induced neurite outgrowth of H19-7 cells. Overexpression of hippocalcin dramatically elongated neurites and increased the expression of basic helix-loop-helix transcription factor, that is, NeuroD without bFGF stimulation. Treatment of the cells with hippocalcin siRNA completely blocked bFGF-induced neurite outgrowth and NeuroD expression. bFGF stimulation resulted in activation of phospholipase C-gamma (PLC-gamma) and an increased level of intracellular Ca2+. Hippocalcin expression by bFGF stimulation was fully blocked by both the PLC-gamma inhibitor U73122 and BAPTA-AM, a chelator of intracellular Ca2+, suggesting that hippocalcin expression by bFGF is dependent on PLC-gamma and Ca2+. Moreover, both U73122 and BAPTA-AM completely blocked bFGF-induced neurite outgrowth and NeuroD expression. Taken together, these results suggest for the first time that bFGF induces hippocalcin expression in H19-7 cells through PLC-gamma activation, which leads to neurite outgrowth.

Hippocalcin gates the calcium activation of the slow afterhyperpolarization in hippocampal pyramidal cells

In the brain, calcium influx following a train of action potentials activates potassium channels that mediate a slow afterhyperpolarization current (I(sAHP)). The key steps between calcium influx and potassium channel activation remain unknown. Here we report that the key intermediate between calcium and the sAHP channels is the diffusible calcium sensor hippocalcin. Brief depolarizations sufficient to activate the I(sAHP) in wild-type mice do not elicit the I(sAHP) in hippocalcin knockout mice. Introduction of hippocalcin in cultured hippocampal neurons leads to a pronounced I(sAHP), while neurons expressing a hippocalcin mutant lacking N-terminal myristoylation exhibit a small I(sAHP) that is similar to that recorded in uninfected neurons. This implies that hippocalcin must bind to the plasma membrane to mediate its effects. These findings support a model in which the calcium sensor for the sAHP channels is not preassociated with the channel complex.

Hippocalcin: a new solution to an old puzzle

In hippocampal pyramidal neurons, calcium entry following an action potential burst results in a slow afterhyperpolarization (sAHP) that critically regulates subsequent excitability. Although this potassium current was described two decades ago, the mechanism whereby the rise in intracellular calcium generates the sAHP was, until now, not known. In this issue of Neuron, Tzingounis et al. now show that calcium binding to hippocalcin, a member of the NCS family, is one of the necessary steps involved in production of the sAHP.

Neuronal calcium sensor proteins: generating diversity in neuronal Ca2+ signalling

In neurons, intracellular calcium signals have crucial roles in activating neurotransmitter release and in triggering alterations in neuronal function. Calmodulin has been widely studied as a Ca(2+) sensor that has several defined roles in neuronal Ca(2+) signalling, but members of the neuronal calcium sensor protein family have also begun to emerge as key components in a number of regulatory pathways and have increased the diversity of neuronal Ca(2+) signalling pathways. The differing properties of these proteins allow them to have discrete, non-redundant functions.

Hippocalcin protects hippocampal neurons against excitotoxin damage by enhancing calcium extrusion

Hippocalcin, which is a member of the neuronal calcium-sensor protein family, is highly expressed in hippocampal pyramidal cells. Recently, it was demonstrated that hippocalcin deficit caused an increase in neuronal cell death in the field CA3 of Ammon's horn (CA3) region of the hippocampus following the systemic injection of kainic acid. Treatment with kainic acid results in seizure-induced cell death in CA3. In the present study, we injected quinolinic acid, which is an N-methyl-d-aspartate receptor agonist, into the hippocampal field CA1 of Ammon's horn (CA1) region in hippocalcin-knockout (-/-) mice, a procedure which mimics transient ischemia. Although significant pyknotic changes were observed at the injected site in wild-type (+/+) mice 24 h after injection, the area of pyknotic cells extended throughout the hippocampus in -/- mice. The quantification of cell numbers in Nissl-stained sections indicated that the cell damage in -/- mice was more severe than that in +/+ mice. The density of terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-biotin nick-end labeling-positive cells roughly paralleled that of Nissl-stained pyknotic cells. Primary cultures of hippocampal neurons showed that the number of surviving neurons from -/- mice after 7 days in culture was smaller than the number from +/+ mice. The measurement of intracellular calcium concentrations in single cells revealed that the calcium extrusion from -/- neurons was slower than that from +/+ neurons. The involvement of hippocalcin in the upkeep of calcium extrusion was confirmed using hippocalcin-expressing COS7 cells. These results suggest that hippocalcin plays an important role in calcium extrusion from neurons and, in turn, helps to protect them against calcium-dependent excitotoxin damage in the hippocampus.

Hippocalcin increases phospholipase D2 expression through extracellular signal-regulated kinase activation and lysophosphatidic acid potentiates the hippocalcin-induced phospholipase D2 expression

We have previously isolated a 22 kDa protein from a rat brain which was found to be involved in activating phospholipsae D (PLD), and identified the protein as hippocalcin through sequence analysis. Nevertheless, the function of hippocalcin for PLD activation still remains to be resolved. Here, we proposed that hippocalcin was involved in extracellular signal-regulated kinase (ERK)-mediated PLD2 expression. To elucidate a role of hippocalcin, we made hippocalcin transfected NIH3T3 cells and showed that the expression of PLD2 and basal PLD activity were increased in hippocalcin transfected cells. We performed PLD assay with dominant negative PLD2 (DN-PLD2) and hippocalcin co-transfected cells. DN-PLD2 suppressed increase of basal PLD activity in hippocalcin transfected cells, suggesting that increased basal PLD activity is due to PLD2 over-expression. Hippocalcin is a Ca2+-binding protein, which is expressed mainly in the hippocampus. Since it is known that lysophosphatidic acid (LPA) increases intracellular Ca2+, we investigated the possible role of hippocalcin in the LPA-induced elevation of intracellular Ca2+. When the intracellular Ca2+ level was increased by LPA, hippocalcin was translocated to the membrane after LPA treatment in hippocalcin transfected cells. In addition, treatment with LPA in hippocalcin transfected cells markedly potentiated PLD2 expression and showed morphological changes of cell shape suggesting that increased PLD2 expression acts as one of the major factors to cause change of cell shape by making altered membrane lipid composition. Hippocalcin-induced PLD2 expression potentiated by LPA in hippocalcin transfected cells was inhibited by a PI-PLC inhibitor, U73122 and a chelator of intracellular Ca2+, BAPTA-AM suggesting that activation of hippocalcin caused by increased intracellular Ca2+ is important to induce over-expression of PLD2. However, downregulation of PKC and treatment of a chelator of extracellular Ca2+, EGTA had little or no effect on the inhibition of hippocalcin-induced PLD2 expression potentiated by LPA in the hippocalcin transfected cells. Interestingly, when we over-express hippocalcin, ERK was activated, and treatment with LPA in hippocalcin transfected cells significantly potentiated ERK activation. Specific inhibition of ERK dramatically abolished hippocalcin-induced PLD2 expression. Taken together, these results suggest for the first time that hippocalcin can induce PLD2 expression and LPA potentiates hippocalcin-induced PLD2 expression, which is mediated by ERK activation.

Novel frequenin-modulated Ca2+-signaling membrane guanylate cyclase (ROS-GC) transduction pathway in bovine hippocampus

Frequenin is a member of the neuronal Ca(2+) sensor protein family, implicated in being the modulator of the neurotransmitter release, potassium channels, phosphatidylinositol signaling pathway and the Ca(2+)-dependent exocytosis of dense-core granules in the PC12 cells. Frequenin exhibits these biological activities through its Ca(2+) myristoyl switch, yet the switch is functionally inactive. These structural and functional traits of frequenin have been derived through the use of recombinant frequenin. In the present study, frequenin (BovFrq) native to the bovine hippocampus has been purified, sequenced for its 9 internal fragments, cloned, and studied. The findings show that structure of the BovFrq is identical to its form present in chicken, rat, mouse and human, indicating its evolutionary conservation. Its Ca(2+) myristoyl switch is active in the hippocampus. And, BovFrq physically interacts and turns on yet undisclosed ONE-GC-like ROS-GC membrane guanylate cyclase transduction machinery in the hippocampal neurons. This makes BovFrq a new Ca(2+)-sensor modulator of a novel ROS-GC transduction machinery. The study demonstrates the presence and mechanistic features of this cyclic GMP signaling pathway in the hippocampal neurons, and also provides one more support for the evolving concept where the Ca(2+)-modulated membrane guanylate cyclase transduction machinery in its variant forms is a central operational component of all neurons.

Traffic of Kv4 K+ channels mediated by KChIP1 is via a novel post-ER vesicular pathway

The traffic of Kv4 K⁺ channels is regulated by the potassium channel interacting proteins (KChIPs). Kv4.2 expressed alone was not retained within the ER, but reached the Golgi complex. Coexpression of KChIP1 resulted in traffic of the channel to the plasma membrane, and traffic was abolished when mutations were introduced into the EF-hands with channel captured on vesicular structures that colocalized with KChIP1(2–4)-EYFP. The EF-hand mutant had no effect on general exocytic traffic. Traffic of Kv4.2 was coat protein complex I (COPI)–dependent, but KChIP1-containing vesicles were not COPII-coated, and expression of a GTP-loaded Sar1 mutant to block COPII function more effectively inhibited traffic of vesicular stomatitis virus glycoprotein (VSVG) than did KChIP1/Kv4.2 through the secretory pathway. Therefore, KChIP1seems to be targeted to post-ER transport vesicles, different from COPII-coated vesicles and those involved in traffic of VSVG. When expressed in hippocampal neurons, KChIP1 co-distributed with dendritic Golgi outposts; therefore, the KChIP1 pathway could play an important role in local vesicular traffic in neurons.

Hippocalcin functions as a calcium sensor in hippocampal LTD

It is not fully understood how NMDAR-dependent LTD causes Ca(2+)-dependent endocytosis of AMPARs. Here we show that the neuronal Ca(2+) sensor hippocalcin binds the beta2-adaptin subunit of the AP2 adaptor complex and that along with GluR2 these coimmunoprecipitate in a Ca(2+)-sensitive manner. Infusion of a truncated mutant of hippocalcin (HIP(2-72)) that lacks the Ca(2+) binding domains prevents synaptically evoked LTD but has no effect on LTP. These data indicate that the AP2-hippocalcin complex acts as a Ca(2+) sensor that couples NMDAR-dependent activation to regulated endocytosis of AMPARs during LTD.

High-affinity interaction of the N-terminal myristoylation motif of the neuronal calcium sensor protein hippocalcin with phosphatidylinositol 4,5-bisphosphate

Many proteins are associated with intracellular membranes due to their N-terminal myristoylation. Not all myristoylated proteins have the same localization within cells, indicating that other factors must determine their membrane targeting. The NCS (neuronal calcium sensor) proteins are a family of Ca2+-binding proteins with diverse functions. Most members of the family are N-terminally myristoylated and are either constitutively membrane-bound or have a Ca2+/myristoyl switch that allows their reversible membrane association in response to Ca2+ signals. In the case of hippocalcin and NCS-1, or alternatively KChIP1 (K+ channel-interacting protein 1), their N-terminal myristoylation motifs are sufficient for targeting to distinct organelles. We have shown that an N-terminal myristoylated hippocalcin peptide is able to specifically reproduce the membrane targeting of hippocalcin/NCS-1 when introduced into permeabilized cells. The peptide binds to liposomes containing phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2] with high affinity (K(d) 50 nM). Full-length hippocalcin also bound preferentially to liposomes supplemented with PtdIns(4,5)P2. Co-expression of hippocalcin-(1-14)-ECFP (enhanced cyan fluorescent protein) or NCS-1-ECFP partially displaced the expressed PH (pleckstrin homology) domain of phospholipase delta1 from the plasma membrane in live cells, indicating that they have a higher affinity for PtdIns(4,5)P2 than does this PH domain. The Golgi localization of the PH domain of FAPP1 (four-phosphate-adaptor protein 1), which binds to phosphatidylinositol 4-phosphate, was unaffected. The localization of NCS-1 and hippocalcin is likely to be determined, therefore, by their interaction with PtdIns(4,5)P2.

Hippocalcin protects against caspase-12-induced and age-dependent neuronal degeneration

Hippocalcin is a neuronal calcium binding protein, but its physiological function in brain is unknown. We show here that hippocampal neurons from hippocalcin-deficient mice are more vulnerable to degeneration, particularly using thapsigargin, elevating intracellular calcium. Caspase-12 was activated in neurons lacking hippocalcin, while calpain was unchanged. Neuronal viability was accompanied by endoplasmic reticulum (ER) stress and a change in the relative induction of the ER chaperone, BiP/GRP78. Neuronal apoptosis inhibitor protein (NAIP), known to interact with hippocalcin, was not altered, but hippocampal neurons from gene-deleted mice were more sensitive to excitotoxicity caused by kainic acid. In addition, an age-dependent increase in neurodegeneration occurred in the gene-deleted mice, showing that hippocalcin contributes to neuronal viability during aging.

Hippocalcin-deficient mice display a defect in cAMP response element-binding protein activation associated with impaired spatial and associative memory

Hippocalcin is a member of the neuronal calcium sensor (NCS) protein family that is highly expressed in hippocampal pyramidal cells and moderately expressed in the neurons of cerebral cortex, cerebellum and striatum. Here we examined the physiological roles of hippocalcin using targeted gene disruption. Hippocalcin-deficient (-/-) mice displayed no obvious structural abnormalities in the brain including hippocampal formation at the light microscopic level. Deletion of hippocalcin did not result in up-regulation of the hippocalcin-related proteins; neural visinin-like Ca(2+)-binding proteins (NVP) 1, 2, and 3. The synaptic excitability of hippocampal CA1 neurons appeared to be normal, as estimated by the shape of field excitatory postsynaptic potentials elicited by single- and paired-pulse stimuli, and by tetanic stimulation. However, N-methyl-d-aspartate stimulation- and depolarization-induced phosphorylation of cAMP-response element-binding protein (CREB) was significantly attenuated in -/- hippocampal neurons, suggesting an impairment in an activity-dependent gene expression cascade. In the Morris water maze test, the performance of -/- mice was comparable to that of wild-type littermates except in the probe test, where -/- mice crossed the previous location of the platform significantly less often than +/+ mice. Hippocalcin-deficient mice were also impaired on a discrimination learning task in which they needed to respond to a lamp illuminated on the left or right side to obtain food reinforcement. No abnormalities were observed in motor activity, anxiety behavior, or fear learning. These results suggest that hippocalcin plays a crucial role in the Ca(2+)-signaling pathway that underlies long-lasting neural plasticity and that leads to spatial and associative memory.

The neuronal calcium-sensor proteins

Changes in intracellular free Ca(2+) concentration ([Ca(2+)](i)) affect many different aspects of neuronal function ranging from millisecond regulation of ion channels to long term changes in gene expression. These effects of Ca(2+) are transduced by Ca(2+)-binding proteins that act as Ca(2+) sensors by binding Ca(2+), undergoing a conformational change and then modifying the function of additional target proteins. Mammalian species express 14 members of the neuronal calcium sensor (NCS) family of EF hand-containing Ca(2+)-binding proteins which are expressed mainly in photoreceptor cells or neurons. Many of the NCS proteins are membrane targeted through their N-terminal myristoylation either constitutively or following exposure of the myristoyl group after Ca(2+) binding (the Ca(2+)/myristoyl switch). The NCS proteins have been implicated in a wide range of functional roles in neuronal regulation, several of which have been confirmed though molecular genetic analyses.

Calcium binding sequences in calmyrin regulates interaction with presenilin-2

Calmyrin is a myristoylated calcium binding protein that contains four putative EF-hands. Calmyrin interacts with a number of proteins, including presenilin-2 (PS2). However, the biophysical properties of calmyrin, and the molecular mechanisms that regulate its binding to different partners, are not well understood. By site-directed mutagenesis and Ca2+ binding studies, we found that calmyrin binds two Ca2+ ions with a dissociation constant of approximately 53 microM, and that the two C-terminal EF-hands 3 and 4 bind calcium. Using ultraviolet spectroscopy, circular dichroism (CD), and NMR, we found that Ca(2+)-free and -bound calmyrin have substantially different protein conformations. By yeast two-hybrid assays, we found that both EF-hands 3 and 4 of calmyrin must be intact for calmyrin to interact with PS2-loop sequences. Pulse-chase studies of HeLa cells transfected with calmyrin expression constructs indicated that wild-type (Wt) calmyrin has a half-life of approximately 75 min, whereas a mutant defective in myristoylation turns over more rapidly (half-life of 35 min). By contrast, the half-lives of calmyrin mutants with a disrupted EF-hand 3 or EF-hand 4 were 52 and 170 min, respectively. Using immunofluorescence staining of HeLa cells transfected with Wt and mutant calmyrin cDNAs, we found that both calcium binding and myristoylation are important for dynamic intracellular targeting of calmyrin. Double immunofluorescence microscopy indicated that Wt and myristoylation-defective calmyrin proteins colocalize efficiently and to the same extent with PS2, whereas calmyrin mutants defective in calcium binding display less colocalization with PS2. Our results suggest that calmyrin functions as a calcium sensor and that calcium binding sequences in calmyrin are important for interaction with the PS2 loop.

Hippocalcin in the olfactory epithelium: a mediator of second messenger signaling

Intracellular Ca2+ plays an important role in a variety of second messenger cascades. The function of Ca2+ is mediated, in part, by Ca2+-binding proteins such as calmodulin, calretinin, calbindin, neurocalcin, recoverin, and visinin-like proteins (VILIPs). These proteins are highly expressed in rat olfactory receptor neurons (ORNs) and are localized to distinct intracellular regions. In the present study, we have identified another Ca2+-binding protein, hippocalcin, in the rat olfactory epithelium (OE). Olfactory/brain hippocalcin shows high sequence homology with hippocalcins expressed in mice and humans. Hippocalcin was predominantly localized to the olfactory cilia, the site of the initial events of olfactory signal transduction, and was found to regulate the activity of ciliary adenylate cyclases (ACs) and particulate guanylyl cyclases (GCs) in a Ca2+-dependent manner. These data indicate that hippocalcin is expressed in rat ORNs, and is likely to regulate second messenger cascades in a Ca2+-dependent manner.

Neuronal Ca2+-sensor proteins: multitalented regulators of neuronal function

Many aspects of neuronal activity are regulated by Ca2+ signals. The transduction of temporally and spatially distinct Ca2+ signals requires the action of Ca2+-sensor proteins including various EF-hand-containing Ca2+-binding proteins. The neuronal Ca2+ sensor (NCS) protein family and the related Ca2+-binding proteins (CaBPs) have begun to emerge as key players in neuronal function. Many of these proteins are expressed predominantly or only in neurons, sometimes with cell-specific patterns of expression. Their ability to associate with membranes either constitutively or in response to elevated Ca2+ concentration allows the NCS proteins to discriminate between different spatial and temporal patterns of Ca2+ signals. Recent work has established several physiological roles of these proteins, including diverse actions on gene expression, ion channel function, membrane traffic of ion channels and receptors, and the control of apoptosis.