Electrophysiological study to explore the functional consequences of ng expression on neuroblastoma cells (n2a)

... according to their concentration gradient, thereby causing changes in their cytosolic concentration These changes trigger other cellular reactions that result in the message being transmitted along... resolution of the selectivity and regulation of individual ion channels The outside-out version permits study of changes occurring in response to stimulations of ion channels that are no longer part of. .. clone3 Lane7: blank (loading buffer only) 28 B Ng Dox µg/ml Ng C: control 0.1 µg clone1 clone3 clone7 Fig 2.5 A: Western blot result showing the expression of Ng protein Only the clone N2A cells

Chapter 1: Literature Review Higher brain functions, including learning and memory formation, are thought to involve changes in synaptic such as changes in synaptic transmission efficiency. Therefore, it is useful to address cellular and molecular processes which regulate the ability of brain cells to modify themselves in response to stimulation, in order to gain a better understanding of the biochemical basis of synaptic strength (Ramaker et al., 1997). Phosphorylation and dephosphorylation of key proteins by the respective protein kinases and phosphatases play a vital role in this regulation (Nestler and Greengard, 1983; 1984; 1994). A protein phosphorylation system consists of a protein kinase, a substrate protein and a phosphatase protein. The substrate protein is phosphorylated by the protein kinase, which catalyzes the transfer of the phosphate group of ATP to the hydroxyl moiety of a specific amino acid, and it is converted back to the dephosphorylated form by protein phosphatases, which hydrolyze the phosphodiester bond (Walaas and Greengard, 1991; Girault et al., 1993). One of the most important and best-studied protein kinases in the mammalian brain is Ca2+ and phospholipid dependent protein kinase C (PKC) (Nishizuka, 1995; Ramaker et al; 1997). PKC is a serine/threonine kinase and is part of a large family of proteins with closely related structures but distinct enzymatic characteristics. Currently, two well-characterized neuron-specific PKC substrates are the presynaptic phosphoprotein neuromodulin (also known as Nm, GAP-43, B-50) and the postsynaptic phosphoprotein neurogranin (also known as Ng, RC3, BICKS). Rat Nm and Ng share a nearly identical region of 18 amino acids, containing a PKC phosphorylation site and an overlapping atypical calmodulin (CaM) binding domain (Baudier et al., 1991; Huang et 1 al., 1993; Gerendasy et al., 1994). This common amino acid sequence is composed of an Isoleucine followed by a Glutamine, and is represented as IQ in the single letter coding of amino acids. Therefore, this domain is called the IQ domain. Although Nm and Ng share several biochemical characteristics because of this almost identical amino acid sequence, they are different in their anatomical regions of expression, subcellular distribution, time of expression, and therefore synaptic functions. In our present thesis, our investigation was mainly focused on the postsynaptic PKC substrate Ng protein. In subsequent sections, we will give a literature review of Ng including its expression pattern, biochemical characteristics, phosphorylation by PKC and binding affinity with CaM, and its involvement in learning and memory (LTP and LTD) on a behavioral study level. 1.1 Characteristics of Neurogranin 1.1.1 Identification and Biochemistry of Neurogranin Neurogranin (Ng), also known as BICKS, p17, and RC3, is a second neuronspecific PKC substrate. It was first purified and characterized from bovine brain based on its binding CaM-Sepharose column, and called p17 from its apparent molecular mass in SDS gels (Baudier et al., 1989, 1991; Respresa et al., 1990). Ng was identified as a postnatal rodent cerebral cortex- enriched cDNA by subtractive hybridization (Watson et al., 1990). Ng is a 78 amino acid protein (Watson et al., 1990) and has a molecular mass of 7.5 kDa, determined by electrospray mass spectrometry (Huang et al., 1993). Similar to Nm, Ng is soluble in 2.5% perchloric acid (PCA) due to the fact that both proteins are acidic (Baudier et al., 1989). On SDS-PAGE, Ng displays anomalous behavior with an 2 apparent molecular mass of 1.5 to 1.9 kDa, depending on the percentage of polyacrylamide (Baudier et al., 1991; Huang et al., 1993). 1.1.2 Subcellular localization and expression pattern of Neurogranin In contrast to Nm, Ng is found infrequently in axons and is accumulated postsynaptically in dendritic spines of neostriatal neurons (Watson et al., 1992). By immunohistochemistry, Ng protein expression was found essentially in adult rat telencephalon, specifically located in neurons of the cerebral cortex, hippocampus, striatum, and a few other discrete areas, but was nearly absent in the thalamus, cerebellum, and brain stem (Represa et al., 1990). In 2004, Singec et al. showed that Ng expression could also be found in Golgi cells of mouse cerebellum and in unipolar brush cells, which localized in the granular layer of the monkey cerebellum. These results suggest that Ng has a cell type-specific and species-specific expression pattern. In 2000, an immunohistological analysis done by Houben et al. revealed the presence of Ngpositive cell bodies and dendrites in specific regions in the gray matter of the adult spinal cord. Axon-containing structures, such as the dorsal and ventral corticospinal tract, were also found to be Ng-positive. These data demonstrated that Ng is present in the adult rat spinal cord in both pre- and postsynaptic structures. Northern blot analysis and in situ hybridization results showed that Ng mRNA was highly enriched in adult rat telencephalon, especially in cell bodies and dendritic processes of neurons of the cerebral cortex, striatum and hippocampus, as well as certain nuclei within the thalamus, hypothalamus and olfactory bulb (Respresa et al., 1990). However, in Alzheimer disease (AD) neocortex tissue, there was little or no evidence for 3 Ng mRNA translocation to dendrites, while in fronto-temporal dementia (FTD) neocortex, targeting of Ng mRNA to apical dendrites was preserved (Chang et al., 1997). This observation suggested that Ng mRNA was selectively translocated to dendrites, where it might be translated locally in response to synaptic activity. The above comparative studies in AD and FTD also indicated the potential importance of synapse integrity and the dendritic cytoskeleton in Ng mRNA targeting in the human neocortex. Ng synthesis is developmentally regulated, with no expression in the embryonic and neonatal period, followed by an abrupt increase 2 to 3 weeks after birth. Most of the structures that express Ng during development conserve it in the adult stage (AlvarezBolado et al., 1996). Ageing produced significant decrease in Ng mRNA expression, and Ng protein expression was also downregulated in the ageing mouse brain (Mons et al. 2001). 1.1.3 IQ domain A comparison of the amino acid sequence of Ng with that of another brainspecific PKC substrate Nm, revealed a strikingly conserved amino acid sequence AA(X)KIQASFRGH -(X)(X)RKK(X)K. Besides this, the two proteins are not related over the rest of their amino acid sequences (Baudier et al.,1991). This common amino acid sequence contains an Isoleucine followed by a Glutamine and is represented as “IQ” in the single letter coding of amino acids. Therefore, this domain is called the IQ domain (Fig. 1.1). 4 Fig. 1.1. The “IQ” motif in Ng. The consensus sequence for the “IQ” motif is bracketed within the Ng amino acid sequence. Serine36, the site of PKC phosphorylation, is indicated by an arrow while the CaM-binding domain is denoted by a line below the sequence. CaM binds Ng at this IQ domain. Single amino acid mutagenesis of Ile-33 --> Gln completely inhibited and Arg-38 --> Gln and Ser-36 --> Asp reduced Ng/CaM interactions, indicating that residues within this specific domain are important for CaM binding in vivo (Prichard et al., 1999). This IQ domain is very conserved in other CaMbinding proteins, and IQ motifs are found in diverse families of CaM-binding proteins. Some of these, like PEP-19, Ng, Nm, Igloo and Sp17 are representatives of the small IQmotif-containing protein family (Neel and Young 1994; Richardson et al., 1994). Through some cellular mechanism, the IQ-peptide/CaM interaction could regulate biological processes including gene transcription (Slemmon et al 2000). This IQ domain also contains a unique PKC phosphorylation site at Ser36. Ng was shown to be phosphorylated in hippocampal slices incubated with 32Pi and phorbol ester, suggesting that Ng is an in vivo substrate of PKC. Tryptic digestion of the phosphorylated protein yielded a single phosphopeptide having the sequence IQASFR, where the serine residue is the phosphorylated amino acid, and the Ser36 is adjacent to the CaM-binding domain (Ala29-Ser48) (Huang et al., 1993). In 2003, Putkey et al. proposed a new role of the IQ motif. Their results showed that PEP-19 accelerated by 40 to 50-fold both the slow 5 association and dissociation of Ca2+ from the C-domain of free CaM. Importantly, they demonstrated that PEP-19 also increased the rate of dissociation of Ca2+ from the Cdomain of CaM when bound to intact CaM-dependent protein kinase II. Thus, PEP-19, and presumably similar members of the IQ family of proteins, such as Ng, has the potential to alter the Ca2+ -binding dynamics of free CaM and CaM that is bound to other target proteins. 1.2 Calmodulin and Neurogranin Like Nm, Ng binds CaM at low Ca2+ concentrations, and the affinity of CaM for Ng is greatly reduced when the Ca2+ concentration is elevated (Baudier et al., 1991; Coggins and Zwiers 1991). In rat brain homogenate, Ng forms a stable complex with CaM in the absence of Ca2+, as evidenced by an increase in the M(r) determined by immunoblot analysis following gel filtration chromatography. Also, the addition of a negative charge to residue 36 is sufficient to disrupt all detectable Ng-CaM interactions (Gerendasy et al., 1994). Huang et al. (1993), using a 20-amino-acid synthetic peptide (AS-20F-W) containing the PKC phosphorylation site and CaM-binding domain of Ng (Ala29-Ser48), proved that it interacts with CaM through an electrostatic interaction in the absence of Ca2+ but through a hydrophobic interaction in the presence of Ca2+. This provides structural confirmation for two-way binding modes and suggests that CaM regulates the biological activities of Ng through an allosteric, Ca2+- sensitive mechanism that can be uncoupled by protein kinase C-mediated phosphorylation (Gerendasy et al., 1995). In 1999, Prichard et al. illustrated that CaM not only binds to Ng in vivo, but is also the only Ng-interacting protein isolated from brain cDNA libraries. Cui et al. 6 (2003), using nuclear magnetic resonance (NMR), demonstrated that the interaction between Ng and CaM was a type of weak binding. Special CaM-binding properties, together with a high abundance of Ng in the hippocampus, suggests that Ng might play an important role in regulating local CaM concentrations in the postsynaptic cell. The interaction of Ng-CaM also indicated that Ng could regulate CaM-dependent targets (Martzen and Slemmon, 1995). For example, a study of Ng knock out mice carried out by Pak et al. in 2000 showed that deletion of the Ng gene in mice did not result in obvious developmental or neuroanatomical abnormalities, but caused an impairment of spatial learning and changes in hippocampal short- and long-term plasticity (paired-pulse depression, synaptic fatigue and LTP). A decreased basal level of activated CaMKII accompanied the deficits. In addition, hippocampal slices of Ng mutant mice displayed a reduced ability to generate activated CaMKII after stimulation of protein phosphorylation and oxidation. These results indicate a central role of Ng in the regulation of CaMKII activity, with decisive influences on synaptic plasticity and spatial learning (Pak et al., 2000). On the contrary, the interaction of CaM with its target proteins is known to affect the kinetics and affinity of Ca2+ binding to CaM. Gaertner et al. (2004) reported that Ng and CaMKII produce opposing effects on the affinity of CaM for calcium. Ng is shown to decreased the binding affinity of CaM for Ca2+ ; however, CaMKII increased the CaM binding affinity for Ca2+ . with the existence of CaMKII, by increasing the dissociation rate of CaM, Ng can weaker the effects of CaMKII on Ca2+ dissociation This effect is not found with phosphorylated CaMKII. In conclusion, there are competing pathways for the 7 dissociation of the Ca2+-CaM target complex, and that the Ca2+ -CaM binding properties are highly regulated (Gaertner et al., 2004). 1.3 Protein Kinase C and Neurogranin Protein phoshorylation is a final common pathway of fundamental importance in biological regulation. Virtually all types of extracellular signals, both inside and outside the nervous system, are known to produce many of their diverse physiological effects by regulating the state of phosphorylation of specific phosphoproteins within their target cells. Indeed, protein phosphorylation has been shown to regulate many neuronal functions (Walaas and Greengard, 1991; Girault 1993; Nestler and Greengard, 1984). 1.3.1 Neurogranin phosphorylation by PKC Protein kinases differ in their cellular and subcellular distribution and substrate specificity. They are generally classified as protein serine/threonine kinases, which phosphorylate substrate proteins on serine or threonine residues, or as protein tyrosine kinases, which phosphorylate substrate proteins on tyrosine residues (Girault et al., 1993). PKC is a serine/threonine kinase that is the major receptor for tumor-promoting phorbol esters, which activate the kinase in vitro in a way very similar to DAG (Bell and Burns, 1991). Ng is identified as a cellular postsynaptic PKC substrate: first, it can be phosphorylated in hippocampal slices incubated with 32Pi, and secondly phorbol ester, a PKC activator, can stimulate Ng phosphorylation. Both above evdiences suggested that Ng is likely to be an in vivo substrate for PKC. In vitro phosphorylation of Ng by PKC produced a shift of the isoelectric point of the protein (pI 5.6) to a more acidic value (pI 5.4) (Baudier et al., 1991). Tryptic digestion of the phosphorylated protein yielded a 8 single phosphopeptide having the sequence IQASFR, where the serine residue is the phosphorylated amino acid. The Ng peptide (28-43) was shown to be phosphorylated by PKC with a Km of 150 nM. Substituting Arg36 with Ile caused a significant reduction in the affinity for PKC (Chen et al., 1993). The sequence IQASFR was later proven to be part of the IQ motif, as discussed in section 1.1.3 concerning the IQ domain. PKC is present ubiquitously in a variety of tissues, and the PKC family consists of at least 12 isoforms that possess distinct differences in structure, substrate requirement, expression and localization (Huang et al., 1986; Way et al., 2000 for review; Ramakers, 1997). Three different PKC isozymes, designated as I, II and III, are products of α, β and γ genes. Differences in the distribution and developmental expression of PKC isozymes suggest that the different isozymes of PKC have distinct functions. Ng can be phosphorylated by PKC α, β, and γ, and these enzymes phosphorylated Ng at the single site, Ser36 (Huang et al., 1993). Among these 12 isozymes of PKC, the cellular and subcellular localizations and the pattern of expression of Ng during development, in many respects, resemble those of the protein PKC-γ isozyme, which is also a neuronspecific protein, although the Ng gene has a more restricted neuronal expression than PKC-γ. The Ng and PKC-γ genes have a similar expression pattern in the brain during development. These two genes share at least four conserved sequence segments 1.5 kilobase pairs upstream from their transcriptional start sites (Sato et al., 1995). Similarly to Ng, the expression of PKC-γ progressively increases from the fetal stage up to 2-3 weeks after birth. The characteristic brain-specific and developmental stage-regulated expression of the two proteins PKC-γ and Ng is distinctly different from 9 that of PKC-α and β. These latter two PKCs are ubiquitously expressed in a variety of tissues and cell types. In 1999, Remakers et al. demonstrated that when a brain slice was depolarized with potassium, activated by glutamate receptors binding glutamate, or directly stimulated by phorbol ester-mediated activation of PKC, an increased Ng phosphorylation could only be found in wild type animals, with no effect on Ng phosphorylation in mice lacking PKC-γ. These results obtained from PKC-γ knockout mice suggest that Ng could be phosphorylated by the γ isoform of PKC after stimulation of the metabotropic glutamate receptor. It is more likely that Ng is a PKC-γ substrate in vivo. An important question is: what are the functional consequences of PKC phosphorylation of Ng? Cohen et al. in 1993 first reported that the activation of endogenous PKC by phorbol esters generated inward chloride (Cl-) currents in Ng oocytes but not in control oocytes. These acetylcholine (ACh)-evoked inward Cl- currents were dependent on both IP3 release and intracellular Ca2+. Their results suggested that PKC-phosphorylated Ng can enhance the mobilization of intracellular Ca2+ in Xenopus oocytes and, by inference, may play a role in Ca2+ homeostasis in neurons. Functional studies in Xenopus oocytes of single-site variants in the CaM-binding domain of Ng by the same group further proved that Ng not only regulated the availability of free CaM but also, when phosphorylated, independently stimulated G-protein coupled second messenger pathways that generated inositol 1,4,5-trisphosphate (IP3), diacylglycerol (DAG) and intracellular Ca2+ (Watson et al., 1996). 10 1.3.2. Neurogranin dephosphorylation Phosphorylation of proteins is one part of the protein phosphorylation cascade; the other is their dephosphorylation. However, compared to protein kinases, much less is known about the functional role of protein phosphatases. This reflects the general inclination in biological research to concentrate on the “turn on” processes as opposed to the “turn off” processes. It also reflects the greater technical difficulties associated with the study of protein phosphatases. Very limited studies on the phosphorylation of Ng suggest that this phosphoprotein is dephosphorylated by three isoforms of rat brain calcineurin, also known as CaM-dependent protein phosphatase (CaMPP). Dephosphorylation of Ng by all three CaMPP isozymes, CaMPP-1, -2, and -3, was stimulated to a higher extent in the Due to the fact that CaMPP was highly concentrated in the brain and colocalized with major PKC substrates in various brain regions, Seki et al proposed that CaMPP had the potential to reverse the action of PKC (Seki et al., 1995). 1.4 Relationship between Ca2+/CaM and Neurogranin 1.4.1 Phosphorylation of Neurogranin is involved in Ca2+ regulation The hypothesis that Ng functions by modulating the IP3/DAG second messenger pathway after its phosphorylation by PKC was indirectly tested by heterologous expression in Xenopus oocytes, suggesting that PKC-phosphorylated Ng is capable of enhancing the mobilization of intracellular Ca2+ in Xenopus oocytes and may play a role in Ca2+ homeostasis in dendrites of forebrain neurons (Cohen et al., 1993). Nonetheless, 11 direct measurements of [Ca2+]i changes with phosphorylation of Ng was lacking. See more discussion of the relationship between Ng and intracellular calcium in 4.1.2. 1.4.2 Conflicting convergences of Ca2+/CaM and PKC on shared Neurogranin Ca2+/CaM and PKC are involved in neuronal activities, such as axon sprouting, dendritic spine formation, gene expression, growth cone navigation, LTP and neurotransmitter release (Abraham and Tate, 1997; Porumb et al., 1997). Until recently, it seemed that these two multipurpose signals did not impinge on each other’s targeting sites, but evidence is now accumulating rapidly that suggests that they compete for the same domains in a variety of substrates in neurons and other cells. Ng, as discussed in the above several sections, contains the KIQASF motif that combines the IQ motif of a subset of neuron specifically CaM binding domain with the SF motif, a specific targeting site of PKC phosphorylation. NMR results have shown that Ng is made of flexible, unstructured, randomly coiling rods, but CaM can stiffen it and lock its domain III into amphiphilic α helices in the absence of Ca2+ (Ran et al., 2003; Gerendasy et al., 1997). CaM stabilizes a basic, amphiphilic alpha helix within Ng under physiological salt concentrations only when Ca2+ is absent. This provides structural confirmation for two binding modes and suggests that CaM regulates the biological activities of Ng through an allosteric, Ca2+-sensitive mechanism that can be uncoupled by protein kinase C-mediated phosphorylation (Gerendasy et al., 1995). Phosphorylation of Ser36 in domain III of Ng by PKC prevents the protein from binding CaM (Baudier et al., 1991; Son et al., 1997), and Ser-36 --> Asp point mutants reduced Ng/CaM interactions (Prichard, 1999). Huang et al. (1993) reported the phosphorylation of Ng by PKC or 12 PKM, a protease-degraded PKC, was inhibited by CaM. In the presence of Ca2+, the affinity of Ng with CaM was greatly reduced, and Ca2+/CaM became less inhibitory of the PKM-catalyzed phosphorylation of Ng. In contrast, phosphorylation of Ng by PKM prevented its interaction with CaM even in with Ca2+ (Huang et al., 1993; Sheu et al.,1995). On the other hand, CaM binding to Ng and PKC phosphorylation are convergent. Wang and Kelly (1995) demonstrated that postsynaptic injection of Ca2+/CaM can induce synaptic potentiation, and pseudosubstrate inhibitors or highaffinity substrates of CaMKII or PKC blocked Ca2+/CaM-induced potentiation, indicating the requirement of CaMKII and PKC activities in synaptic potentiation induced by Ca2+/CaM. Besides the effect of phosphorylation of Ng by PKC on its binding with CaM, oxidation of Ng can also affect its CaM binding affinity. In 1995, Sheu et al. first reported that Ng’s four cysteine residues were readily oxidized by Nitric Oxide (NO) donors, and by several oxidants to form intramolecular disulfide bridges. Modification of Ng by DEANO, a NO donor, was blocked by CaM in the absence of Ca2+; while in the presence of Ca2+, CaM did not protect Ng from oxidation by DEANO. This oxidized form, unlike the reduced form, did not bind to a CaM affinity column, and the oxidized Ng was also a poorer substrate for PKC. In 1999, Li et al reported that the transient oxidation of Ng induced by NMDA, which can be blocked either by NMDA receptor antagonist AP5, or NO synthase inhibitor. These results suggested that redox of Ng is involved in the NMDA-mediated signaling pathway, and some enzymes may catalyze the oxidation and reduction of Ng in the brain. While the redox state of Ng, similar to the 13 state of phosphorylation of Ng, may regulate the level of CaM, which in turn of course modulates the activities of CaM-dependent enzymes. Computer-aided modeling of Ng-CaM interactions suggests the relationship between Ca2+ flux size and CaM availability. Simulations of the interactions between Ng/CaM and Ca2+ imply that Ng tunes and homeostatically constrains the Ca2+ signal transduction system. By doing so, it may link Ca2+ fluxes to downstream elements of a signaling cascade that generates LTP (Chakravarthy et al., 1999). In summary, both oxidation of Ng and phosphorylation of Ng are effective in modulating the intracellular level of CaM. These results indicate that modification of Ng to form intramolecular disulfides outside the IQ domain provides an alternative mechanism for regulation of its binding affinity to CaM (Huang et al., 2000). Thus it was proposed that Ng releases CaM rapidly in response to large influxes of Ca2+ and slowly in response to small increases. This nonlinear response is analogous to the behavior of a capacitor, hence the name calpacitin. The capacitance of the system is regulated by phosphorylation by PKC, which abrogates interactions between CaM and Ng. It was further proposed that the ratio of phosphorylated to unphosphorylated Ng determines the sliding LTP/LTD threshold, in concert with Ca2+/CaM-dependent kinase II (Gerendasy and Sutcliffe, 1997; Gerendasy, 1999; Chakravarthy, 1999). Pak et al. (2000) showed that hippocampal slices of Ng mutant mice displayed a reduced ability to generate activated CaMKII after phosphorylation and oxidation,indicating a central role of Ng in the regulation of CaMKII activity. By its Ca2+-sensitive CaM-binding feature, and through its phosphorylation and oxidation, Ng can regulate the Ca2+- and Ca2+/CaM-dependent 14 signaling pathways downstream of NMDA receptor stimulation (Pak et al., 2000; Wu et al., 2003). 1.5 Significance of studies As the post synaptic substrate of PKC, Ng was predicated and proved to involve many aspects of intracellular process through its binding propertiy with CaM, and the binding affinity of which could be modified by the several factors such as phosphorylation, oxidation or the level of intracellular free calcium. Although a small, soluable protein which moved around in the microenviroment, Ng was rated as top five most abundant proteins (~60uM, Huang et al 2004). This fact leads us a concern that Ng may have much more intracellular influence than just an capacitor or modulator of CaM. 1.5.1 Missing gaps Although the molecular and biochemical properties of Ng have been well studied, considerably less is known about its physiological roles in neuronal function at single cell level. . The well-defined tissue and cellular distribution suggests that Ng may play a key role in postsynaptic neuronal functions. Cohen et al. (1993) first provided indirect evidence that PKC-phosphorylated Ng can enhance the mobilization of intracellular Ca2+, suggesting that Ng may play a role in Ca2+ homeostasis in neurons. Functional studies in Xenopus oocytes of single-site variants in the CaM-binding domain of Ng, by the same group, further suggested that Ng not only regulates the availability of free CaM, but also, when phosphorylated, independently modulates the intracellular Ca2+ concentration and stimulates G-protein coupled second messenger pathways that generate IP3 and DAG (Watson et al., 1996). van Dalen et al. (2003), using fura-2 based Ca2+ imaging, provided 15 the first direct calcium detection for several aspects of Ca2+ dynamics in cultured cortical neurons obtained from Ng knockout (Ng-/-) and wild type mice (Ng+/+), and for the Ca2+ response produced by stimulation with phorbol ester. Their results also suggested a regulatory role of Ng in neuronal Ca2+ dynamics. However, more detailed mechanisms of the regulatory role of Ng in Ca2+ dynamics after its phosphorylation by PKC have not been reported yet, and how the Ng binding affinity with CaM after its phosphorylation affects its modulation of Ca2+ dynamics has not been examined yet. Studies of Ng involvement in second messenger pathways by using direct measurements of intracellular Ca2+ are also lacking. On the other hand, although it was well accepted that Ng is involved in the modulation of synaptic plasticity, as well as learning and memory, there is no report on how membrane and cell activities are affected by the activation of Ng at a single cell level. 1.5.2 Objectives 1.5.2.1 Cell model The most important evidences for the role of Ng in neural function came from the comparative analysis of wild type and Ng knockout mice (Pak et al., 2000; Krucker et al., 2002). Results from rat or mouse brains, which have endogenous Ng expression, may come from the downstream influence of Ng expression, and thus this experimental model canot provide a blank control without Ng. While in Ng knockout rats or mice, although we can obtain some evidence about the consequence of deleting Ng, suggesting it is involved in learning and memory, we also cannot exclude that this is to some degree due to physiological changes in knockout rat or mice (Pak et al., 2000). 16 To avoid the above-mentioned shortcomings of models, we stably transfected Ng into a neuronal cell line, neuroblastoma N2A. On the one hand, this experimental model provides wild type N2A as blank control, and at same time avoids changing cell function by deleting endogenous proteins. 1.5.2.2 Influence of Neurogranin on channel activities using patchclamp In this established cell model, electrophysiological experiments were carried out to determine how the voltage gated ionic channels’ activities were affected by the expression of neurogranin at both the macroscopic current level and single channel level. First, a whole-cell model of patch-clamp was used to determine whether channel current would be changed due to the expression of Ng by the comparative studies conducted on control and Ng transfected cells. Later, single channel experiments were performed to clarify which aspects of channel activities, on a single channel level, contributed to the whole-cell current changes. In short, this part of the work, by using patch-clamp techniques, is an attempt to determine whether Ng expression can regulate the functional properties of voltage gated channels, and what is the molecular mechanism accounting for this regulation. We hope that our present findings not only provide evidence for the involvement of Ng in the regulation of channel properties in neuronal cell lines, but also open up a far reaching impact on certain aspects of Ng related cellular functions regarding the cellular plasticity that leads to learning and memory formation mechanisms. 17 1.5.2.3 Influence of Neurogranin on cytosolic Ca2+ levels using fura-2 imaging The purposes of this part of the experiments were: first, to determine the functional consequence of Ng expression in a neuronal cell line’s cytosolic calcium level changes; and secondly, to test the hypothesis that Ng modulates intracellular calcium levels by the IP3 and DAG second messenger pathways or CaMKII pathway after its phosphorylation by PKC at the membrane (Randy et al., 1993). In the established cell model, microscopic, fura-2- based calcium imaging experiments were carried out to monitor real time cytosolic calcium level changes. We examined several aspects of Ca2+ signaling, including baseline Ca2+ levels measured in both the Ng transfected cell line and wild type control cell line, Ca2+ dynamic changes induced by Ng phosphorylation by PKC, and pharmacological studies by blocking CaMKII downstream signaling or PIP2 synthesis. Our present study in this thesis should provide direct electrophysiological data to demonstrate that the expression of Ng changes cytosolic calcium levels directly or indirectly after its phosphorylation by PKC. Our data also helps to clarify by which second messenger pathway Ng regulates cytosolic calcium levels after its phosphorylation, and how the Ca2+/CaM complex and PKC phosphorylation cooperate with each other to modify this process. Both aspects of investigation should be useful for our understanding of Ng’s regulatory role in Ca2+ dynamics, and helpful for obtaining new and direct evidence concerning which second messenger pathway regulates cytosolic Ca2+ levels. 18 Chapter 2: Materials and methods 2.1 Establishment of a cell model The first step involved in designing a system in which to perform electrophysiological recordings was the selection of a suitable cell type. Following on from this was the development of a protocol for transfecting the cells with Ng and making a stable cell line of Ng expression in order to investigate Ng modification on cell physiological properties. Below, we outline the reasons behind the choice of cell line, describe the problems encountered during the design of the protocol and explain the measures taken to resolve those issues. Finally, the resulting protocol is detailed. 2.1.1 Choosing an appropriate cell model As we stated in the literature review, Ng is a neuron specific protein enriched in the rat forebrain. Neuronal cell cultures are much more popular among researchers compared to animal models, owing to the ease in maintenance and longer culture lifespan. There are two types of neuronal cultures: primary cultures and immortalized neuronal cell lines. The former is generated by direct transfer of the neuronal cell into culture from the appropriate brain region and is often preferred, as it pertains to many more morphological and biochemical features of their in vivo counterparts, as compared to the metabolically deviated immortalized cells. However, primary neuronal cultures have a limited lifespan in vitro and can be maintained only up to a few passages before showing senescence, making the following investigations very difficult (Thonemann and Schmalz, 2000). Primary cultured neurons from brain tissue also could not provide a blank control without Ng expression except by using Ng knockout mice, which have shown several abnormal aspects (Pak et al., 2000). 19 Especially for our studies, immortalized cultures are more suitable for long-term experiments. The cell type used was selected with reference to the cell type used by Prof Sheu, who is the sponsor of this project. Clone neuro-2a was derived from a mouse neuroblastoma cell line (N2A), and established by Klebe and Ruddle from a spontaneous tumor of a strain of an albino mouse. This tumor line, designated C1300, was obtained from the Jackson Laboratory, Bar Harbor, Maine [22161]. The cell line displays some of the morphological and electrophysiological properties of cells derived from neurons. N2A cells produce large quantities of microtubular protein, which is believed to play a role in a contractile system responsible for axoplasmic flow in nerve cells. The cell line has been used for studies on the mechanism of vinblastine precipitation of microtubular protein, the kinetics of GTP binding to isolated proteins, the turnover of microtubules in vivo, and the synthesis and assembly of microtubular protein (Olmsted et al., 1970, 1971). N2A cells are also being used in a variety of different studies that investigate ion channel properties in neural cells, especially examining the properties of calcium (Fan et al., 1995; Kennedy and Henderson, 1992; Morton et al., 1992; Murphy et al., 1991; Nobile et al., 1998; Reeve et al., 1994, 1995a, 1995b; Yang et al., 2004) This immortalized cell line is also suitable for creation of special clones such as one harbouring a plasmid. In our studies, we transfected Ng plasmid into N2A cells and established several clones with stable Ng expression. In this way, we also obtained two control N2A cell lines: one being wild type control N2A cells without plasmid transfected and another being plasmid transfected control with plasmid inserted but 20 showing no Ng expression. These properties make transfected N2A cells a suitable model in which to investigate the functional characteristics of Ng. We are especially interested in the neuroblast-like cells, having a relatively big cell body with relatively short neuritic processes (Fig.2.1). These cells have displayed many of the biochemical and biophysiological properties found in neuronal cells and can grow on the surface of coated cover slips attached to the bottom. This property makes them fit for electrophysiological recordings such as patch-clamp and calcium imaging. 20 µm Fig. 2.1. Photograph of N2A cells. Cells were cultured for 2 days (taken with a light microscope (Mag x400). Cells grow as a monolayer. 21 2.1.2 Physiological properties of mouse neuroblastoma N2A cells Since most of our research would be focused on the investigation of the effect of Ng protein at the physiological level, including ion channels and intracellular calcium signaling pathways, the chosen cell model must have basic channel expression variety and excitability, which can mimic neuron characteristics. The mouse neuroblastoma N2A cell line is a good candidate for this consideration. The cell line has been the subject of many general studies involving calcium currents, sodium currents and potassium currents. N2A was characterized by three types of delayed rectifier K+ channels: one transient K+ channel and at least one type of Ca2+-activated K+ channel, as well as TTX sensitive Na+ channels (Nobile and Lagostena, 1998, 2000; Nobile and Vercellino, 1997; Backus et al 1991). Those properties make N2A a good model for the investigation of action potentials, which could be influenced by the expression of Ng on this cell line, considering its possible effects on channel properties. In addition, the presence of calcium channels in the mouse neuroblastoma cell line N2A or human neuroblastoma cell line SH-SY5Y has been widely accepted. Some studies have also been performed examining the type of calcium channel present in this cell type; both L and N type calcium channels have been identified in differentiated and undifferentiated versions of this cell type (Kennedy and Henderson, 1992; Murphy et al., 1991; Morton et al., 1992; Reeve et al., 1994, 1995; Reuveny and Narahashi, 1993). There is no record in the literature of the presence of T-type and P/Q type calcium channels in either differentiated or undifferentiated N2A cells. The ability to synthesize various neurotransmitters, and the presence of voltage gated ion channels and neurotransmitter receptors in the plasma membrane combine to make N2A cells a useful 22 system for studying the function of Ng in channel properties and intra-cellular signaling pathways. 2.1.3 Stable transfection of pTRE-Tet-Ng and Neurogranin expression Stably transfected N2A cells were established in our group by Yang et al. in 2004. Briefly, N2A cells were transfected using the CLONfectin Transfection kit (Clontech, Palo Alto, CA). Two days after transfection, culture medium was supplemented with 0.5 mg/ml G418 (Clonetech, Palo Alto, CA) to positively select stable integrants. Single transfected cell colonies were isolated using cloning rings and then expanded. To construct pTRE-Tet-Ng, modifications were made on the Tet-On system by combining two plasmids of the system, pTRE2 and pTet-On, into one to reduce the amount of time that was required for creation of the cell line, as shown in Fig.2.2. 23 Transfect N2A cell line with the construct Select in presence of G418 Isolate G418-resistant clones Screen by genomic DNA PCR, RT-PCR and Western blot Freeze stocks of stable Ng-expression cell lines Fig. 2.2. Schematic diagram of the steps involved in the transfection and screening for the creation of stable Ng-expression N2A cell lines. (Diagram was modified from Yang HM). 24 2.2 Identification of stable expression of Neurogranin After screening, four stable cell lines were proven to express Ng and were then frozen as stock for further use. One clone, which has plasmid transfected but without detectable expression of Ng, was also selected as a plasmid transfected control. In our study, we used two experimental controls to compare results derived from the stable expression of pTRE-Tet-Ng: a wild type control and a plasmid transfected control, in which the plasmid pTRE-Tet-Ng was inserted into the chromosome but no Ng expression was detected by western blot. The N2A clone was defined as the N2A cells that were transfected with pTRE-Tet-Ng and had detectable Ng expression by western blot. To maintain consistency between experiments and to ensure the same expression level of protein among the N2A clone cells, the same N2A clone was used for all experiments carried out in our study. 2.2.1 PCR PCR was carried out on the wild type control, the plasmid inserted control, and the N2A clone to confirm plasmid insertion. For detection of the rtTA gene, the following forward (F) and reverse primers (R) were used: rtTA(F) [5’-TAGAGCTGCTTAA TGAGGTCGG-3’]; rtTA(R): [5’-CCTCGATGGTAGACCCGTAATT-3’] respectively. For the detection of the TRE gene, the following primers were used: pTRE(F) [5’TRGTGAACCGTCAGATCGCC-3’] and pTRE(R) [5’-TGAAAACTTTGCCCCCTCC 3’] respectively. 25 M 1 2 3 4 5 6 7 Lane M: 100 bp DNA ladder from Fermentas (1500,1000,900,800,700,600,500) Lane1: blank (loading buffer only) Lane2: N2A wild type control Lane3: N2A plasmid transfected control Lane4: N2A clone3 Lane5: N2A clone4 Lane6: blank (loading buffer only) Lane7: rtTA control Fig. 2.3. PCR results showing the rtTA gene. rtTA gene could be detected in the plasmid transfected control and the clone N2A cells but was absent from the wild type control N2A cells. 26 M 1 2 3 4 5 6 7 Lane M: 100 bp DNA ladder from Fermentas (1500,1000,900,800,700,600,500) Lane 1: N2A clone3 Lane 2&3: blank (loading buffer only) Lane 4: N2A wild type control Lane 5&6: blank (loading buffer only) Lane 7: N2A plasmid transfected control Fig. 2.4. PCR results showing the pTRE gene. The pTRE gene could be detected in the plasmid transfected control and the clone N2A cells but was absent from the wild type control N2A cells. 27 2.2.2 Western blot Western blot was later performed to verify Ng protein expression. The cells (wild type control, plasmid transfected control and clone) were treated with cell lysis buffer which contained (in mM): 20 Tris, pH 7.5, 150 NaCl, 10 EDTA, 1 PMSF, 1% NP-40, 10% glycerol, and 1 µg/ml aprotinin. Aliquots containing 80 µg of total protein extract were separated by 15% SDS-PAGE gel, transferred to a nitrocellulose membrane, and probed for Ng using a primary antibody (against the c-terminal 66-78 amino acids of Ng, raised in rabbit) at 1:2500 dilution followed by goat anti-rabbit IgG conjugated with peroxidase as the secondary antibody at 1:5000 dilution. A Ng 1 2 3 4 5 6 7 Lane1: 0.1ug purified Ng as positive control Lane2: blank (loading buffer only) Lane3: N2A wild type control Lane4: N2A plasmid transfected control Lane5: N2A plasmid transfected control with 2 µg/ml Dox added in medium Lane6: N2A clone3 Lane7: blank (loading buffer only) 28 B Ng Dox 0 2 µg/ml Ng C: control 0.1 µg clone1 clone3 clone7 Fig. 2.5. A: Western blot result showing the expression of Ng protein. Only the clone N2A cells exhibited a detectable expression level of Ng whereas both the wild type control and the plasmid transfected control N2A cells did not show any detectable level of Ng expression. B: The clone N2A cells were incubated with the medium containing different concentration of Dox (0-2 µg/ml) at 37 ºC in the CO2 incubator for one day. The result shows that without addition of Dox, there was background expression of Ng. C: data shows that there is no significant difference on the Ng expression level for three clones screened, first lane is 0.1 µg purified Ng. The Dox-dose-dependent response of Ng expression level in the N2A cells was detected by western blot at different concentration of Dox (0-2 µg/ml) in the medium. Unfortunately, our results show that the Ng-expressing clones also display background expression without addition of Dox, 2 µg/ml Dox did not induce a significant high expression of Ng (Fig2.5B). To exclude the possible influorescence due to the inaccurates Dox concentration applied to cell, in our experiment we did not treat cells with extra Dox. The several N2A clones screened all show the detectable Ng expression level. And in this thesis, we will use clone 3 to keep the Ng expression level consistent. The level of Ng stable expression could be maintained for three months, indicating a stable expression of Ng in the clone N2A cells we chose throughout this study. 29 2.3 Cell culture N2A cells were routinely cultured in 75 cm2 tissue culture flasks, 35 mm culture dishes or glass coverslips coated with Poly-L-ornithine (15 µg/ml) in Dulbecco’s MEM (GIBCO™, Grand Island, NY) containing 1% v/v penicillin-streptomycin, 10% v/v Tetracycline-free fetal bovine serum (Clontech, Palo Alto, CA) and 25 mM HEPES (Sigma-Aldrich, St. Louis, MO), pH 7.4. The cells were plated at a density of 2.5 x 105 cells/cm2 in 35 mm Petri dishes containing circular 10 mm diameter poly-L-Lysine coated coverslips. The cultures were maintained in a humidified CO2 incubator (5% CO2, 37°C). For clone cells, Geneticin (G418) was added in the above complete media to provide the stably transfected cell line with selective pressure. For subculturing, cells were washed once with phosphate-buffered saline (PBS), ethylenediamineteraacetic acid (EDTA) and dissociated with trypsin-EDTA (Invitrogen) for 1-3 mins at 37 oC. For cryoconservation of the cell line, cells was detached by trypsin-EDTA from the bottom of culture dishes and washed twice with PBS-EDTA followed by washing once with DMEM. Cells were then pelleted down and resuspended into DMEM with 10% dimethysulfoxide (DMSO, Sigma) as a cryo-protective agent. Cells in cryogenic tubes were stored in a –150oC nitrogen freezer after two days of pre-cooling in a –80oC freezer. For recovery of frozen cell lines, cells were released from the nitrogen freezer by rapidly thawing the cryogenic tubes in a 37oC water bath and cultured in an excess volume of pre-warmed DMEM, with changing of the medium the next day. 30 2.4 Methods of monitoring electrical activities in cells 2.4.1 Intracellular electrode and patch-clamp 2.4.1.1 Introduction of the patch-clamp technique The advent of intracellular electrodes, and subsequently the patch-clamp technique, has revolutionized the studies of both the structure and function of ion channels in the plasma membrane (Hodgkin and Huxley, 1952a, b). Intracellular microelectrodes are usually made of metal, or from pieces of fine glass tubing which have been pulled such that the tip has a diameter of less than one micron. This tubing may then be filled with a conducting solution, for example a simple salt solution such as potassium chloride. The tip of this electrode is inserted into the cytoplasm of the cell, and the plasma membrane then adheres tightly to the pipette, forming a secure seal, although most of the cell interior remains relatively undisturbed. This electrode may then be used to measure intracellular concentrations of inorganic ions, such as sodium, potassium or calcium; or the pipette may be used to inject molecules into the cell, and the effects of those molecules studied. The patch-clamp technique was introduced to biology initially by Cole in the late 1930s and was exploited by the Hodgkin & Huxley team in the late 1940s. The Hodgkin & Huxley conceptual model of the axon presented in 1952 was based on data acquired using this technique. The development of techniques using conventional microelectrodes for the intracellular recording of cellular electrical phenomena has permitted the investigation of the electrical responses of the apical membrane of the intestinal epithelium to secretagogues (Stewart and Turnberg, 1989) and changes in extracellular osmolarity (Giraldez et al., 1988). However, the resolution of intracellular microelectrode 31 recordings is limited by the leak current resulting from membrane puncture of small, relatively inaccessible cells and the inability to characterise the individual channels underlying the macroscopic conductances. The discovery that blunt, polished glass micropipettes can form seals of high electrical resistance when brought in contact with the cell membrane (Neher & Sakmann, 1976; Hamill et al., 1981) has made possible the resolution of picoampere-sized currents flowing through individual ion channels. Any current passing into or out of the patch electrode must pass through ion channels in the ‘patch’ of the membrane covering the tip of the pipette, and may be recorded. The close proximity of the cell to the electrode yields excellent signal to noise ratios. It is possible to manipulate the patch of membrane to produce four different configurations: the ‘cell-attached’ patch, the ‘inside-out’ patch, the ‘whole-cell’ patch, and the ‘outside-out’ patch (Fig. 2.6). Each of these configurations allows different aspects of ionic activity to be observed. The cell-attached configuration allows measurements of the effects of extracellular stimulation on all ion channels in the patch of membrane clamped. Furthermore, the patch-pipette can be withdrawn ("excised") from a cell with the seal intact (Horn & Patlak, 1980), permitting exposure of both the intracellular ("inside-out") and extracellular ("outside-out") faces of a membrane patch to bathing solutions of defined composition, allowing the detailed resolution of the selectivity and regulation of individual ion channels. The outside-out version permits study of changes occurring in response to stimulations of ion channels that are no longer part of the cell. The inside-out method allows the effects of intracellular messengers (such as calcium or cAMP) to be monitored. 32 By back-filling the pipette with a nystatin-containing solution after filling the pipette tip with nystatin-free solution, stable low-access resistance, or simply applying negative pressure to ‘suck’ to break that patch of membrane, ‘whole-cell ’ recordings can be obtained within 1-3 minutes of giga-ohm seal formation. The whole-cell configuration permits measurements of the effect of hormones or other electrical stimulation on all the ion channels in the cell membrane. Four types of configuration of patch-clamp are shown as cartoons in Fig. 2.6. In our studies, both whole-cell and cellattached configurations were used to determine the functional consequence of the expression of Ng on channel activities. Another advantage of this technique is that it allows a drug or some similar substance to be introduced directly into the interior of the cell, through the patch electrode. We also take advantage of this technique and apply purified Ng into the cell through the electrode pipette. 33 Fig. 2.6. Four different configurations of the patch-clamp technique, and how they are achieved. 34 2.4.1.2 Patch-clamp electrophysiological recording N2A cells were plated on Petri dishes 2-3 days before the electrophysiological experiments. All experiments were carried out at room temperature (22-24ºC). Voltageclamp recordings were made from cells using standard patch-clamp methods in the whole-cell and cell-attached configurations. The cells were observed under an Axiovert 200 microscope (Zeiss, Germany) at 10X magnification. Signals were recorded using a HEKA EPC 9 triple patch-clamp amplifier. Patch pipettes were made from 1.5 mm borosilicate capillary glass (World Precision Instruments, Sarasota, FL) using a Sutter P97 puller (Sutter Instrument Co., Novato, CA). The pipettes were fire-polished and the tip resistance was 2-5 MΩ for whole-cell and 8-12 MΩ for single-channel recordings when filled with pipette solution. Currents were recorded using an EPC-9 amplifier (Heka Elektronik). Data acquisition was controlled by PULSE 8.63 (Heka Elektronik). Currents were filtered at 1-2 kHz and digitized directly to a Macintosh Power G4 PC. A ground electrode consisting of an Ag+/AgCl wire was placed in the chamber. The patch pipette holder was also fitted with an Ag+/AgCl wire electrode. Both electrodes were periodically re-chlorided by placing in a FeCl3/HCl solution for 30-60 min. 2.4.2 Optical recording techniques and fura-2 experiments 2.4.2.1 Types of fluorescent dye imaging At least three types of signals may be imaged during cellular activity: those intrinsic to the biological preparation, those from absorbance dyes, and those from fluorescent indicators. Fluorescent indicators are by far the most common tools for measuring cellular activity. These indicators, many of which are both sensitive and selective, fall into two categories commonly used in cell biology: potential-dependent 35 indicators and ion/analyte concentration-sensitive indicators. The popularity of dynamic fluorescence imaging is in large part due to the availability of selective and sensitive fluorescent indicators. Indicators are now available to measure concentrations of ions (Ca2+, H+, Mg2+, Cl-, Zn2+, K+, Ni2+, etc.), biochemical agents (cAMP and protein kinase C), and transmembrane potential. There is a wide range of dyes available commercially, but the underlying principle is the same: the dye undergoes a change in fluorescence as the cell responds to a stimulatory event, indicating the concentration changes of the ion. The use of fluorescent dyes was felt to have the potential to provide a simple method for optical recording of the electrical activity of a cell. A variety of techniques are available to measure fluorescent signals through a microscope, the simplest of which is photometry. In photometry, a single-element light detector is used, either a photomultiplier tube or a photodiode. All the light accepted by a usually adjustable rectangular or circular aperture is passed to the detector. The primary limitation of photometry is that the single detector constrains measurements to only one multicellular or subcellular Region of Interest (ROI), unless scanning is used. In contrast, camera-based fluorescence imaging offers the advantage of measuring fluorescence changes over an entire field of view or more than one ROI. Images can be acquired by video camera to 8-bit or 10-bit resolution at video rates, which are up to 30 frames/s in the US. Some digital cameras are capable of faster rates over limited subregions of the image, and offer higher spatial resolution as well. 36 Fig. 2.7. Side view of the equipment for image acquisition. Solid line highlights the excitation light path and dotted line highlights the emission light path. 2.4.2.2 Dual wavelength ratiometric imaging The advantage of dual wavelength ratiometric imaging over single wavelength methods is that the former allows the calibration of the image intensities in terms of absolute concentrations, or at least eliminates the confounding effects of cell depth changes and other factors such as changes in optical path length, cellular swelling, changes in indicator concentration between measurement, and photobleaching. This method can also be used to detect standing gradients of ion concentration. Ratiometric imaging requires that the indicator has different sensitivities to its substrate at two or more excitation or emission wavelengths. For example, the peak wavelength of fura-2 37 shifts from about 380 nm in calcium-free to about 340 nm in calcium-saturated solution. The changes in fluorescence produced by changes in concentration of ions or analytes are therefore in opposite directions at the two excitation wavelengths. If a ratio image is calculated between images collected at the two excitation (or emission) wavelengths, several hard-to-measure parameters are effectively canceled, since they contribute in the same way to the fluorescence intensity at all wavelengths. Ratiometric indicators are available for measurements of membrane potential, pH and concentration of calcium, magnesium, chloride and cAMP. 2.4.2.2.1 Choosing a fluorescent indicator Fura-2 is an improved fluorescent calcium probe to its precursor "Quin-2": the first cell permeant calcium dye. The development of fura-2 by Roger Tsien and collaborators in 1989 has been an important contribution to the better understanding of the role of calcium in cellular processes. Fura-2 (C44H47N3O24) is a Ca2+ indicator, a dye molecule that binds Ca2+ with four carboxyl groups. Its structure is derived from EGTA (Ethylen glycolbis (beta-amino-ethyl ether)) that also has 4 COOH-groups that bind calcium. Fura-2 is a calcium chelator, which means that it is able to bind Ca2+ such that the ions are not able to react with anything and can easily be transported. This highly selective substance for calcium is nearly insensitive to slight fluctuations in the physiological range of pH values. The calcium affinity is very useful for physiological measurements, as it has a dissociation constant of about 0.25 µM. This is in the range of cellular calcium signals (0.1-1µM). Fura 2-AM is an acetoxymethyl ester derivative of fura-2 that can be easily loaded into cells by incubation. The acetoxymethyl (AM) ester derivatives of fluorescent 38 indicators and chelators make up one of the most useful groups of compounds for the study of live cells. Modification of carboxylic acids with AM ester groups results in an uncharged molecule that can permeate cell membranes. Once inside the cell, the lipophilic blocking groups are cleaved by nonspecific esterases; finally, hydrolysis of the esterified groups is essential for binding of the target calcium ion. The AM ester of fura-2 can passively diffuse across the cell membrane, enabling researchers to avoid the use of invasive loading techniques. The fluorescence intensity is also dependent on many other poorly quantified or variable factors such as illumination intensity, emission collection efficiency, dye concentrations, autofluorescence of the cell and effective cell thickness in the optical beam. Fura-2 responds to Ca2+ by shifting its wavelengths while maintaining strong fluorescence. The ratio of the fluorescence at two suitable chosen wavelengths would then signal for Ca2+ while canceling out most or all of the possible variability due to instrument efficiency and content of effective dye. Thus it is a good way to determine Ca2+ concentration by using ratiometric measurements, which are essentially independent of uneven dye loading, cell thickness, photobleaching and dye leakage. Excitation and emission wavelength preferences depend on the type of instrumentation being used, as well as on sample autofluorescence and on the presence of other fluorescent or photoactivatable probes in the experiment. As a rule, AM esters are used at a final working concentration of between 1 and 10 µM. The AM ester concentration should be kept as low as possible to reduce potential artifacts from overloading, including incomplete hydrolysis, compartmentalization and toxic effects of hydrolysis by-products such as formaldehyde or acetic acid. Loading must be done at a temperature that is optimal for the cells. 39 fura-2 fura-2 AM -----------------------------------------------------------------------------------------------------------Fig. 2.8. Structure of fura-2 and fura-2 AM. The most interesting difference between fura-2 AM and other fluorescent dyes like Qin-2 is the ability of calcium to alter the wavelengths for quantitative meaning. Via fura-2, the intracellular calcium concentration can be displayed by using ratio values of 340/380. Ratioing minimizes a number of negative effects that occur and disturb measurements, such as uneven dye loading, leakage of fura-2 and bleaching. Fura-2 dye provides the possibility to perform measurements for about 1 hour without significant bleaching. Even different thicknesses do not corrupt the result when using the ratio of 340/380. With the help of the ratio values, the exact intracellular calcium concentration of cells (single cells or mean value for cell population) can be easily calculated. 40 2.4.2.2.2 Determination of the dissociation constant (Kd) of the fluorescent Ca2+ indicator fura-2 Fura-2 is an UV light excitable ratiometric Ca2+ indicator. Upon binding Ca2+, fura-2 exhibits a large enhancement in fluorescence signal, which can be monitored by the ratio of fluorescence at 340 nm to fluorescence at 380 nm (Fig. 2.6). Fig. 2.9. Spectral response for fura-2 in 0-10 mM CaEGTA buffers using a calcium calibration buffer kit (Molecular Probes, USA). In order to accurately measure the level of [Ca2+]free in a particular cytosolic environment using a fluorescent indicator, we used calcium calibration buffer kits to determine the dissociate constant (Kd) of fura-2 at a chosen temprature, ionic strength and 41 pH. Using the quantitative imaging system, the Kd of fura-2 can be calculated from a plot generated by scanning the excitation or emission of fura-2 in the presence of eleven different Ca2+ concentrations. In our calcium imaging experiments, calibrations were performed using the fura-2 Ca2+ imaging calibration kit (Molecular Probes, USA) to obtain values for Rmin, Fmax380, Fmin380, Rmax, and Kd. Using the formulae below, ratio values were converted to Ca2+ concentrations. [Ca2+]free = Kd*{(R-Rmin)/(RmaxR)}*[Fmax380/Fmin380] where R is the ratio of emission intensity excited at 340 nm to that excited at 380 nm; Rmin is the ratio at zero free Ca2+; Rmax is the ratio at saturating Ca2+ (e.g. 39 µM); Fmax380 is the fluorescence intensity excited at 380 nm, for zero Ca2+; and Fmin380 is the fluorescence intensity at saturating free Ca2+. The free Ca2+ for any experimental sample can then be calculated from the corresponding R value. The plot of the log of [Ca2+]free (x-axis) versus the log of {(R - Rmin)/(Rmax - R)} *[Fmax380/Fmin380] yields a straight line, the X-intercept of which is the log of Kd. Thus, an empirical value for Kd can be determined from the fura-2-containing Ca2+ standards. The [Ca2+]i was calculated after background subtraction. The background intensity can be determined by the blank buffer without fura-2, provided in the calibration kit. 42 Fig. 2.10. Calibration curve of fura-2 with the fura-2 Ca2+ imaging calibration kit. The plot of the log of [Ca2+]free (x-axis) versus the log of {(R - Rmin)/(Rmax - R)} * [Fmax380/Fmin380] (Y-axis) yields a straight line, the X-intercept of which is the log of Kd. (Kd is 582 nM from the data). 43 Chapter3: Neurogranin modulates the functional properties of Na+ channels as a neuron-specific, calmodulin-binding protein 3.1 Introduction The generation of electrical signals in neurons requires both selective membrane permeability and specific ion concentration gradients across the plasma membrane. The membrane proteins that give rise to these two essential conditions are called ion channels and pumps, respectively. Ion channels, as the phrase implies, have pores that allow particular ions to diffuse across the neuronal membrane. Some channels also have specialized domains that sense the electrical potential across the membrane. Such channels open or close in response to the level of membrane potential, allowing the membrane permeability to be voltage-sensitive. These channels are said to be voltagegated. Ions are then free to move either way across the cell membrane according to their concentration gradient, thereby causing changes in their cytosolic concentration. These changes trigger other cellular reactions that result in the message being transmitted along the cell. Many types of voltage-gated ion channels have been identified, and this diversity generates a wide spectrum of electrical characteristics among neuron types. In the human body, sodium, potassium and calcium play a major role in the relay of messages from one cell to another. The well-defined tissue and cellular distribution suggests that Ng may play a key role in postsynaptic neuronal function. Several lines of evidence have suggested that Ng is involved in various forms of synaptic plasticity. Briefly, Ng was found to be phosphorylated by PKC after the induction of LTP, an electrophysiological model of synaptic plasticity, and the intracellular application of antibodies recognizing the Ng 44 phosphorylation site domain prevented the induction of LTP (Klann et al., 1992; Ramakers et al., 1995; Fedorov et al., 1995). Synaptic transmission and synaptic plasticity were altered in hippocampal slices from Ng knockout mice (Pak et al., 2000; Krucker et al., 2002). At the level of gene expression, the Ng gene was extensively downregulated during the consolidation phase of fear learning (Ressler et al., 2002). It has been reported that Ng is involved in different learning behaviors in mice (Feldker et al., 2003). See further details in the literature review. Whereas the biochemical properties of Ng have been well characterized, considerably less is known about its physiological role in neuronal function. We report here the generation of stable mouse neuroblastoma cells (N2A) capable of expressing rat Ng. The electrophysiological experiments were conducted at the level of both macroscopic whole-cell currents and microscopic single channel currents in an attempt to clarify the role of Ng in functional alterations of voltage-dependent Na+ channels in the N2A cell line. Our data show that overexpression of Ng decreased Na+ current density in the N2A cell line by reducing the open channel probability, and such reduction is in turn due in large part to the prolonged close time of the Na+ channel, whereas the channel conductance and the mean channel open time remain unchanged. 45 3.2 Materials and Methods 3.2.1 Cell culture N2A cells were routinely cultured in the Dulbecco’s MEM (GIBCO™, Grand Island, NY) containing 1% v/v penicillin-streptomycin, 10% v/v Tetracycline-free fetal bovine serum (Clontech, Palo Alto, CA) and 25 mM of HEPES (Sigma-Aldrich, St. Louis, MO), at pH 7.4. The cells were plated at a density of 2.5×105 cells/cm2 in 35 mm Petri dishes containing circular 10 mm diameter poly-L-Lysine coated coverslips. Cultures were maintained in a humidified CO2 incubator (5% CO2/95% room air, at 37°C). For clone cells, G418 was added in the above complete medium to provide the stably transfected cell line with selective pressure. Only trelatively round single cells without obvious dendrite extension were selected, to avoid possible voltage clamping difficulty as well as interference due to cell-cell interactions. 3.2.2 Experiment setup and pipette preparation Patch-clamp recordings were performed on cells using standard patch-clamp methods in the whole-cell and cell-attached configurations ( refer to section 2.4.1.2 for the detail preparation of pipette) The pipette potential was zeroed before seal formation, and the voltage was not corrected for the liquid junction potential. Series resistances for all whole-cell recordings were controlled below 10 MΩ to minimize membrane voltage control errors, and serial resistance capacity transients were electronically compensated. For whole-cell recordings, currents were filtered at 1 kHz and sampled at 50 kHz. For single channel recordings, currents were filtered at 1 kHz and sampled at 2 kHz. 46 3.2.3 Solutions for patch-clamp recording The solutions used in whole-cell and cell-attached voltage-clamp configurations and current-clamp recordings are shown in Table 3.1. pH values of all bath solutions were adjusted to 7.4 with NaOH, and pH values of all pipette solutions were adjusted to 7.2 with CsOH, except for the single channel recording pipette solution, which was also adjusted with NaOH. The osmolarity was adjusted to 320 mOsm/L with glucose. During the recording, all chemicals dissolved in external bath solution were perfused into a small recording chamber (2 ml) by a gravity-driven perfusion system (PHASE II 900-302, Singapore). Table 3.1. Solutions used in patch-clamp recording. Bath solution (in mM) Pipette solution (in mM) Whole-cell recording 145 NaCl, 2.8 KCl, 2 MgCl2, without blockage of 2 CaCl2, 10 HEPES channels 145 KCl, 2.8 NaCl, 2 MgCl2, 10 HEPES, 2 ATP, 0.1 MgGTP Na+ current only 110 NaCl, 35 TEA, 2 MgCl2, 5 EGTA, 10 HEPES 145 CsCl, 2.8 NaCl, 2 MgCl2, 10 HEPES, 2 EGTA, 2 ATP, 0.1 MgGTP Single channel recording 145 NaCl, 2.8 KCl, 2 MgCl2, 2CaCl2, 10 HEPES 110 NaCl, 35 TEA, 2 EGTA, 10 HEPES 47 3.2.4 Data analysis Results were expressed as mean ± SEM of three or more independent experiments. Statistical significance was determined by an unpaired t-test or one-way ANOVA, with p < 0.05 representing significance. All whole-cell data obtained from N2A cells were analyzed by IGOR Pro software (WaveMetrics, Lake Oswego, OR). For measurements of whole-cell current density, membrane currents were normalized to membrane capacitance (pA/pF). Single channel data were analyzed with TAC and TACFIT 4.15 software (Bruxton Co., Seattle, WA). Data were filtered at 300 Hz by a Gaussian filter. Leak subtraction was achieved by subtracting each active sweep with the averaged sweeps with no channel activity. Event detection was performed automatically with the 50% threshold detection method, and each transition was visually inspected before being accepted. Only patches with one channel opening were selected for data analyses. The open probability, indicted by Po, was calculated from the mean open and closed times using the equation τopen P= τopen + τclosed Data were not corrected for missing events. The distributions of open and closed times were fitted by exponential function: f(t) = Σ i Weighti × exp (τi t τi ) Where Weighti and τi are the weight coefficient and time constant, respectively, of each exponential component. 48 The unitary mean currents at different voltages were obtained from all point histograms that were fitted by the sum of the Gaussian distributions. The unitary conductance was evaluated by linear regression of the mean unitary current recorded at voltages ranging from -60 mV to -20 mV. 3.3 Results 3.3.1 Channel properties of control N2A cells To avoid spatial membrane potential control difficulty due to interneuronal connectivity, only round, slightly elongated and healthy cells were selected for whole-cell recording. Fig. 3.1. Patch-clamp recording in cultured wild type control N2A cells, with glass pipette touching the cell. First we used standard bath solution and pipette solution for whole-cell recording, without blockage of any channel current (Table 3.1). The cells were held at a resting 49 potential of –70 mV. No evidence of run-down of current was observed during the recording period. 500pA 20ms 0mV -70mV Fig. 3.2. Whole-cell current induced by a single depolarization test at 0 mV in a wild type control N2A cell. As shown in Fig. 3.2, in the control N2A cell (no Ng expression), when delivering a 0 mV single depolarization pulse from the resting potential (-70mV), we observed a typical whole-cell current recording, with an inward transient current and outward stable current. Later, we used chemicals to confirm the type of channel currents, including outward and inward currents. As shown in Fig. 3.3, 300 nM of the sodium channel blocker, tetrodotoxin (TTX), could block most transient inward current, suggesting that this transient current is mainly carried by the TTX-sensitive Na+ channel. The substitution of K+ by Cs+ in the pipette solution combined with the Ca2+ chelator EGTA could also block most of the outward stable current, suggesting that the outward current mainly consists of K+ current. 50 Current (nA) Time (s) Fig. 3.3. The representative inward transient current can be totally blocked by 300 nM TTX. Current was induced by a single depolarizing test from a resting potential of –70 mV to –10 mV in a wild type control N2A cell. The K+ in the pipette solution was substituted by Cs+. control 35mMTEA 100pA 20ms 20mV 0mV -70mV Fig.3.4. 35 mM TEA can totally block the K+ current in a wild type control N2A cell. Current was induced by a single depolarizing test from a resting potential of –70 mV to 20 mV in a control neuroblastoma N2A cell. 51 When we blocked the outward K+ current with 35 mM tetraethylammonium chloride (TEA) in the bath solution, as shown in Fig. 3.4, we noticed that the potassium current was totally abolished. This potassium current is consistent with previous reported results on the neuroblastoma cell N2A (Nobile and Lagostena, 1998; Nobile and Vercellino, 1997). It is a type of delay-rectifier potassium channel, which can be blocked by adding TEA to the external solution. The resting inward stable current after blocking with TEA was regarded as Ca2+ current, which was below 20 pA (average value) in our cells. The amplitude of Ca2+ current in the neuroblastoma N2A cell was smaller compared with those recorded from human neuroblastoma SH-SY5Y cells (Reuveny and Narahashi, 1993). The small amplitude of Ca2+ current in N2A cells makes it quite difficult to compare the amplitude of Ca2+ current in the three cell lines (wild type control, plasmid transfected control and clone cells). Therefore, in the following experiments, we will focus on whether Na+ or K+ whole- cell current would be affected by the expression of Ng in this mouse neuroblastoma cell line. 3.3.2 Neurogranin expression in the stably transfected N2A cells leads to a decrease in whole-cell Na+ current density Next we examined the whole-cell potassium and sodium current in three cell types: wild type control N2A, which is a blank control without Ng transfected; plasmid transfected control N2A which was plasmid transfected but with no detectable Ng expression, and clone N2A, which has stable Ng expression. Both Na+ and K+ currents recorded from these three cell types were compared to determine if the expression of Ng affected channel properties at the whole-cell current level. But before we draw any conclusion based on possible current changes, we need to consider a factor that may 52 change the amplitude of channel current, that is, cell volume; since bigger cells will surely have large currents. To exclude the possible influence of cell volume changes on the amplitude of whole-cell channel currents, we compared the channel current density value for the wild type control, plasmid transfected control and clone N2A cells respectively by dividing current by cell membrane capacitance. As shown in Fig. 3.5, three types of cells were depolarized by stepping from testing potential -60 mV to +30 mV in 10 mV increments to determine their currentvoltage relationship (I-V curve), resting potential is always -70mV. Fig. 3.5. I-V curve of Na+ current in wild type control, plasmid transfected control and clone N2A cells. Whole-cell sodium current was induced by a series of voltage-clamp steps, testing potentia from -60mv to 30 mV with 10 mV steps (resting potential is always -70 mV) In order to further study the effect of Ng stable expression on Na+ current in N2A cells, we compared the peak Na+ current densities obtained at the -10 mV testing 53 potential from wild type control, plasmid transfected control and N2A clone cells (Fig. 3.6: lower panel histograms). Stable expression of Ng resulted in a significant decrease (p < 0.05) in peak Na+ current (~ 60% as illustrated by columns of the peak currents in Fig. 3.6). Peak Na+ current density was 9.6 ± 1.4 pA/pF (n=11) in N2A clone cells stably expressing Ng, as compared to 23.9 ± 3.6 pA/pF (n = 11), and 24.8 ± 2.5 pA/pF (n = 15) in plasmid transfected control and wild type control cells, respectively. Membrane capacitances for three representative cells (insets in Fig. 3.6) are 26.54 pF (wild type control N2A), 25.13 pF (plasmid transfected control) and 23.35 pF (N2A clone). Fig. 3.6. Summary data of Na+ peak current density when giving a test pulse at –10 mV in three cell lines. The three inset representative single Na+ currents were from wild type control (26.54 pF) , plasmid transfected control (25.13 pF) and N2Aclone cells (23.35 pF). From the above two figures (Fig. 3.5. and Fig. 3.6.), we can draw a conclusion that even apart from the possibility of cell volume, here represented by the membrane 54 capacitance, the whole- cell Na+ current density still reduced to around 50% of the sodium current density in wild type control N2A cells. Although a decrease in Na+ currents was detected in clone cells, which have stable Ng expression, no significant difference was detected in the current density change of K+ when comparing the data from wild type control N2A, plasmid transfected control and clone cells (Fig. 3.7). Fig.3.7. I-V curve of K+ current in wild type control (N2A control), plasmid transfected control (N2A trans) and clone (N2A clone) cells. Whole-cell potassium current was induced by a series of voltage-clamp steps from a holding potential of -10mV to 50 mV with 10mV steps. Cells were held at a resting potential of –70 mV. Cell numbers are 10 for all three types of cell. The above results, that Ng expression in the neuroblastoma cell line only selectively changes the whole-cell channel current, suggested that there must be some direct or indirect interaction of this PKC substrate with sodium channels. This issue was investigated in further detail in our next studies, step by step. 55 3.3.3 Diffusion of Neurogranin protein into cells decreases whole-cell Na+ current To further confirm the effect of Ng on whole-cell Na+ current, we directly introduced purified Ng into wild type control N2A cells. Briefly, 0.2 µg/µl of purified Ng (courtesy of Sherry) was dissolved into the pipette solution and left to diffuse into the cytosol freely after breaking the membrane to form the whole-cell configuration in separate experiments. After the membrane was broken, whole- cell recording was maintained over a period of time (several minutes) to allow the protein to diffuse slowly into the cytosol and to examine its effect. As an internal control, current was recorded at the onset of whole-cell configuration, before protein diffused into the cell, and after the current reached a stable level. After 5-10 minutes of in the whole-cell configuration, we observed a decrease of whole-cell Na+ current in wild type control N2A cells (less than 20% at peak current, The degree of decrease induced by direct Ng introduction was smaller than that recorded in Ng stably transfected clone N2A cells, suggesting that delivery of protein cannot fully mimic the physiological function of Ng. However, this experiment does provide supplementary information that Ng can modulate the sodium channel and causes the decrease of whole-cell Na+ channel current. 3.3.4 Neurogranin changes the action potential pattern in N2A cells Voltage-gated Na+ channels (VGSCs) mediate rapid and transient Na+ influx, to generate and propagate action potential in neurons and other excitable cells. Electrical activity mediated by Na+ channels plays an important role in neuronal plasticity, especially in cellular plasticity, a form of plasticity that modifies the input-output 56 relationships of the entire neuron (Hausser et al., 2000; Stuart et al., 1997; Baranauskas and Nistri, 1998). Abnormalities of Na+ channels are associated with various neurological diseases (Taylor et al., 1997). As our above data showed that Na+ current density was decreased in Ng clone cells which have stable Ng transfection, it is natural to expect that the action potential pattern would be changed due to the Na+ current decrease. As shown in Fig. 3.8, the frequency and pattern of action potentials show no difference among these three groups of cells, whereas the action potential induced in clone cells had a small amplitude of voltage, and the same current injection could not even induce a full action potential, compared with those in wild type control and plasmid transfected N2A cells. This may be related to the decrease of whole-cell Na+ current density in Ng transfected clone cells. 57 A B C Fig.3.8. Representative traces of action potential patterns recorded from wild type control (A, 26.54 pF), plasmid transfected control (B, 25.13 pF) and clone (C, 23.35 pF) N2A cells. Inset (C) shows the applied stimulating protocol, a series of current pulses from rest at 0 pA to 120 pA, with 20 pA steps, in current-clamp configuration, to induce action potentials. The action potential induced here in N2A cells, whatever wild type control, plasmid transfected control, or clone cells, was not typical all-or-none action potential pattern which we observed from real excitable cells such as brain neurons. This is understandable since N2A, as immortalized neuroblastoma cell line, may not keep all the morphological or physiological features which are only exclusive to neurons. 58 3.3.5 Effect of Neurogranin expression in stably transfected N2A cells on single Na+ channel activity The decrease in whole-cell peak Na+ current and current density in Ng expressing clone N2A cells could result from changes in either Na+ ion permeation, channel gating, or decreased expression of functional channels. To further address this issue, we conducted single Na+ channel recording and analysis by using the cell-attached configuration. Unitary Na+ currents were recorded from cell-attached patches. The cellattached configuration was chosen because it allowed for possible effects of intracellular mediators on the parameters studied. Excision of the patch was avoided because it had been reported to alter the mean open time (Kirsch and Brown, 1989). 3.3.5.1 Single channel conductance of Na+ remains unchanged Single Na+ channel activities were examined in recordings from patches that contained only one or two channels in the wild type control, the plasmid transfected control, and the clone N2A cells. Cell-attached membrane patches were held at -70 mV and given a series of test pulses that ranged from -60 mV to -20 mV in 20 mV increments. Unitary Na+ currents in these three groups of cells were comparable over the entire voltage range studied. Single channel conductance (slope conductance) was estimated by measuring currents at a range of voltages (-60 mV, -40 mV and –20 mV). 59 wild type control N2A Fig.3.9. Mean I-V curve from wild type control N2A cells. The data points have been fitted by a linear regression line to give unitary conductance of the Na+ channel: 21.37 ±1.81pS (n=6), as two channels included in this patch. Inset is a representative single channel recording trace on wild type control N2A cells at –40 mV. Single channel recordings were repeated on other two cell groups: plasmid transfected control and clone N2A cells to compare the possible effects of Ng on the channel conductance. Results are shown in Fig. 3.10 and Fig. 3.11. 60 plasmid transfected control N2A Fig. 3.10: Mean current-voltage relationship from plasmid transfected control N2A cells. The data points have been fitted by a linear regression line to give unitary conductance of the Na+ channel is 24 ± 0.59 pS (n=5), as two channels included in this patch. Inset is a representative single channel recording trace from a wild type control N2A cell at –40 mV. 61 Clone N2A Fig. 3.11. Mean I-V curve from clone N2A cells. The data points have been fitted by a linear regression line to give unitary conductance of the Na+ channel is 20.2 ± 0.72 pS (n=3), as two channels included in this patch. Inset is a representative single channel recording trace from clone N2A cells at –40 mV. There was no significant difference (p > 0.4) between the conductances of the three groups (wild type control, plasmid transfected control and clone cells), indicating that stable Ng expression in the N2A cells did not change Na+ permeation. 62 3.3.5.2 Closed time of single Na+ channels is increased by Neurogranin expression We further analyzed the properties of open and closed Na+ channel events within bursts. Fig. 3.12 shows representative open (A, B, and C) and closed (D, E, and F) time distributions and fitted curves for three groups of cells: wild type control, plasmid transfected control, and clone N2A cells. In these particular cases, the open time distributions are best fitted by a single exponential function, with a mean open time of 7.4 ± 0.3 ms for the wild type control, 7.8 ± 1.1 ms for the plasmid transfected control and 4.45 ± 0.6 ms for the clone N2A cell. Also, only a single exponential function best fits the closed time distributions as shown in Fig. 3.12 D, E, and F. Two or more exponents do not give a better fit. 63 open time closed time A: wild type control D Counts 30 30 20 τo=7.4±0.3ms 20 10 0 20 Counts Counts 30 10 10 0 0 -4 -3 -2 -1 0 log Duration (s) 1 1 2 log Duration (s) -3 -3 -2 -1 0 log Duration (s) -2 -1 1 0 Counts 30 30 30 30 20 20 τo=7.8±1.1ms 20 Counts Counts -4 E Counts 10 10 τc=13.5±1.5ms 20 10 10 0 0 -4 -4 -3 -2 -1 0 -3 -2 -1 log Duration (s) 0 0 1 -4 -4 log Duration (s) C -4 log Duration (s) B: plasmid transfected control 0 τc=10.2±1.3ms -3 -3 -2 -1 0 1 -2 -1 0 1 log Duration (s) 2 log Duration (s) F clone Counts Counts 30 20 20 Counts Counts 30 τo=4.5±0.6ms 10 10 20 20 τc =26.2±3.2ms 10 10 0 0 0 0 -4 -4 -3 -3 -2 -1 0 -2 -1 0 log Duration (s) log Duration (s) 1 1 2 -4 -4 -3-3 -2-2log Duration -1-1 (s) 00 log Duration (s) 11 2 Fig.3.12. Open time and closed time distribution of Na+ channel currents in cell-attached patches of three cell lines. 64 3.3.5.3 Open probability of single Na+ channels is decreased by Neurogranin expression Although the single channel conductance of Na+ channels in Ng expressing cells remained unchanged, the open probability (Po) of single Na+ channels in Ng expressing clone N2A cells was significantly reduced as compared to those in the wild type control and the plasmid transfected control N2A cells (Fig. 3.12.). The mean open times for the three groups of cells were very similar, without significant difference (p > 0.4) (although the mean open time was apparently shorter in the clone N2A cells). On the other hand, N2A clone cells demonstrated a significantly (p[...]... result showing the expression of Ng protein Only the clone N2A cells exhibited a detectable expression level of Ng whereas both the wild type control and the plasmid transfected control N2A cells did not show any detectable level of Ng expression B: The clone N2A cells were incubated with the medium containing different concentration of Dox (0-2 µg/ml) at 37 ºC in the CO2 incubator for one day The result... selection of a suitable cell type Following on from this was the development of a protocol for transfecting the cells with Ng and making a stable cell line of Ng expression in order to investigate Ng modification on cell physiological properties Below, we outline the reasons behind the choice of cell line, describe the problems encountered during the design of the protocol and explain the measures taken to. .. selectivity and regulation of individual ion channels The outside-out version permits study of changes occurring in response to stimulations of ion channels that are no longer part of the cell The inside-out method allows the effects of intracellular messengers (such as calcium or cAMP) to be monitored 32 By back-filling the pipette with a nystatin-containing solution after filling the pipette tip with... that redox of Ng is involved in the NMDA-mediated signaling pathway, and some enzymes may catalyze the oxidation and reduction of Ng in the brain While the redox state of Ng, similar to the 13 state of phosphorylation of Ng, may regulate the level of CaM, which in turn of course modulates the activities of CaM-dependent enzymes Computer-aided modeling of Ng- CaM interactions suggests the relationship between... addition of Dox, there was background expression of Ng C: data shows that there is no significant difference on the Ng expression level for three clones screened, first lane is 0.1 µg purified Ng The Dox-dose-dependent response of Ng expression level in the N2A cells was detected by western blot at different concentration of Dox (0-2 µg/ml) in the medium Unfortunately, our results show that the Ng- expressing... Palo Alto, CA) Two days after transfection, culture medium was supplemented with 0.5 mg/ml G418 (Clonetech, Palo Alto, CA) to positively select stable integrants Single transfected cell colonies were isolated using cloning rings and then expanded To construct pTRE-Tet -Ng, modifications were made on the Tet -On system by combining two plasmids of the system, pTRE2 and pTet -On, into one to reduce the amount... Huxley team in the late 1940s The Hodgkin & Huxley conceptual model of the axon presented in 1952 was based on data acquired using this technique The development of techniques using conventional microelectrodes for the intracellular recording of cellular electrical phenomena has permitted the investigation of the electrical responses of the apical membrane of the intestinal epithelium to secretagogues... dephosphorylation Phosphorylation of proteins is one part of the protein phosphorylation cascade; the other is their dephosphorylation However, compared to protein kinases, much less is known about the functional role of protein phosphatases This reflects the general inclination in biological research to concentrate on the “turn on processes as opposed to the “turn off” processes It also reflects the greater... adjustable rectangular or circular aperture is passed to the detector The primary limitation of photometry is that the single detector constrains measurements to only one multicellular or subcellular Region of Interest (ROI), unless scanning is used In contrast, camera-based fluorescence imaging offers the advantage of measuring fluorescence changes over an entire field of view or more than one ROI Images... stable expression of pTRE-Tet -Ng: a wild type control and a plasmid transfected control, in which the plasmid pTRE-Tet -Ng was inserted into the chromosome but no Ng expression was detected by western blot The N2A clone was defined as the N2A cells that were transfected with pTRE-Tet -Ng and had detectable Ng expression by western blot To maintain consistency between experiments and to ensure the same expression