Changes in Temperature Have Opposing Effects on Current Amplitude in α7 and α4β2 Nicotinic Acetylcholine Receptors

We have examined the effect of temperature on the electrophysiological properties of three neuronal nicotinic acetylcholine receptor (nAChR) subtypes: the rapidly desensitizing homomeric α7 nAChR, the more slowly desensitizing heteromeric α4β2 nAChR and on α7 nAChRs containing a transmembrane mutation (L247T) that results in dramatically reduced desensitization. In all cases, the functional properties of receptors expressed in Xenopus oocytes at room temperature (RT; 21°C) were compared to those recorded at either physiological temperature (37°C) or at lower temperature (4°C). Alterations in temperature had dramatically differing effects on the amplitude of whole-cell responses detected with these three nAChR subtypes. Compared to responses at RT, the amplitude of agonist-evoked responses with α4β2 nAChRs was increased at high temperature (125±9%, n = 6, P<0.01) and reduced at low temperature (47±5%, n = 6, P<0.01), whereas the amplitude of α7 responses was reduced at high temperature (27±7%, n = 11, P<0.001) and increased at low temperatures (224±16%, n = 10, P<0.001). In contrast to the effects of temperature on α4β2 and wild type α7 nAChRs, the amplitude of α7 nAChRs containing the L247T mutation was unaffected by changes in temperature. In addition, changes in temperature had little or no effect on current amplitude when α7 nAChRs were activated by the largely non-desensitizing allosteric agonist 4BP-TQS. Despite these differing effects of temperature on the amplitude of agonist-evoked responses in different nAChRs, changes in temperature had a consistent effect on the rate of receptor desensitization on all subtypes examined. In all cases, higher temperature resulted in increased rates of desensitization. Thus, it appears that the differing effects of temperature on the amplitudes of whole-cell responses cannot be explained by temperature-induced changes in receptor desensitization rates.


Introduction
Nicotinic acetylcholine receptors (nAChRs) are members of the Cys-loop family of ligand-gated ion channels which also includes receptors for 5-hydroxytryptamine (5-HT), c-aminobutyric acid (GABA) and glycine [1]. In common with other Cys-loop receptors, nAChRs are oligomeric transmembrane proteins in which five subunits co-assemble to form a central ion-channel pore [2]. In addition, subunits of Cys-loop receptors share a common transmembrane topology, containing an extracellular N-terminal region and four a-helical transmembrane domains [1].
Seventeen nAChRs subunits (a1-a10, b1-b4, c, d and e) have been identified in vertebrate species and can co-assemble into a large number of nAChR subtypes with considerable diversity in subunit composition [3]. In most cases, nAChRs are heteromeric complexes (containing two, three or four different subunit subtypes) but some subunits, such as a7, are capable of forming homomeric receptors (containing five copies of a single subunit). Within the mammalian brain, two nAChR subtypes (heteromeric a4b2 nAChRs and homomeric a7 nAChRs) have attracted particular attention as targets for pharmaceutical drug discovery. Homomeric a7 nAChRs have high calcium permeability and very rapid desensitization. They have been identified as potential drug targets in treatment of disorders such as Alzheimer's disease and schizophrenia [4,5,6]. Heteromeric a4b2 nAChRs have lower calcium permeability and display less agonist-induced desensitization. Receptors containing a4 and b2 subunits mediate the effects of nicotine associated with tobacco smoking and are the site of action of drugs used to assist with smoking cessation [7]. In addition, a4b2 nAChRs are targets for drug discovery in areas such as cognition, attention and pain [4,5,6].
A variety of experimental approaches have confirmed that conventional orthosteric agonists such as acetylcholine bind at an extracellular site located at the interface of two subunits [8,9]. More recently, studies with a7 nAChRs have demonstrated that nAChRs can also be activated by agonists binding to an allosteric site located in the transmembrane region [10], a site that has previously been proposed as the binding site for a range of allosteric modulators of a7 nAChRs [11,12]. Whereas activation of a7 nAChRs by acetylcholine results in rapid desensitization [13], activation by allosteric agonists such as 4BP-TQS results in very low levels of desensitization [10], consistent with these two agonists having different mechanisms of action. Previous studies have also demonstrated that the rapid rate of desensitization observed when a7 nAChRs is activated by orthosteric agonists such as acetylcholine can be reduced dramatically by the introduction of a single point mutation (L247T) located within the second transmembrane domain [14].
A particular advantage of studies conducted with recombinant nAChRs in artificial expression systems is the ability to examine the properties of receptors with defined subunit composition, as well as the ability to examine the effects of alterations in amino acid composition by means of techniques such as site-directed mutagenesis [15]. Typically, electrophysiological studies of recombinant nAChRs are conducted at room temperature, as are studies of native receptors from isolated cell and tissue preparations. In the present study we have examined the influence of conducting electrophysiological recordings at temperatures above and below room temperature (37uC and 4uC). By means of expression studies in Xenopus oocytes, we have examined heteromeric a4b2 and homomeric a7 nAChRs. In addition, we have examined the effect of changes in temperature on responses evoked by both orthosteric and allosteric agonists, as well as on a7 nAChRs containing a mutation that slows the rate of desensitization caused by orthosteric agonists such as acetylcholine. Changes in temperature resulted in changes in the magnitude of agonist-evoked responses. However, opposing effects were observed on different nAChR subtypes. Changes in temperature were also associated with changes in rates of receptor desensitization, but this does not appear to explain the differences observed in current amplitudes at different temperatures.

Results
Effect of temperature on current amplitude in a7 nAChRs Expression of the human a7 nAChR in Xenopus oocytes was examined by two-electrode voltage-clamp recording. When acetylcholine-evoked responses were recorded at room temperature (RT; 21uC), rapidly desensitizing currents were observed ( Fig. 1), typical of a7 nAChRs [13]. When responses were recorded from the same oocytes at physiological temperature (37uC), agonist-evoked current amplitudes were significantly smaller (P,0.001; Fig. 1, Table 1). This was the case, irrespective of whether a maximal concentration (3 mM; Fig. 1A) or an EC 50 concentration (100 mM; Fig. 1B) of acetylcholine or was used. Conversely, when responses were recorded at lower temperature (4uC), currents were significantly larger (P,0.001; Fig. 1, Table 1). Again, this was the case, irrespective of whether a maximal concentration or an EC 50 concentration of acetylcholine or was used ( Fig. 1, Table 1).

Effect of temperature on current amplitude in a4b2 nAChRs
A similar series of experiments was performed with the human a4b2 nAChR subtype, which displays lower levels of desensitization than homomeric a7 nAChRs (Fig. 2). With a4b2 nAChRs, the effect of temperature on current amplitude was the opposite of that observed with a7 nAChRs. When responses were recorded from oocytes at 37uC, agonist-evoked current amplitudes were significantly larger than at RT (P = 0.002; Fig. 2, Table 1). Conversely, when responses were recorded at 4uC, currents were significantly smaller than at RT (P = 0.002; Fig. 2, Table 1).
Whereas oocyte recordings with a4b2 nAChRs were conducted in calcium-containing saline, recordings with a7 nAChRs were conducted in barium-containing (calcium-free) saline. A calciumfree saline is typically used with a7 nAChRs to minimize the possibility of calcium-activated chloride channels (due to the higher calcium permeability of a7 nAChRs). In order to examine whether differences in saline composition might explain differences in the effect of temperature on current amplitudes, experiments with a4b2 nAChRs were repeated in calcium-free saline. The effect of temperature on current amplitude with a4b2 nAChRs was consistent between the two saline solutions (Fig. 2, Table 1). In addition, further experiments were performed with a7 nAChRs to examine the effects of temperature on responses recorded in calcium-containing saline, rather than in barium-containing saline. The effect of both high (37uC) and low (4uC) temperature on current amplitude of a7 nAChRs was not significantly different to that observed in barium-containing saline (n = 5).
Effect of temperature on current amplitude in a7 L247T nAChRs A further series of experiments was performed with the human a7 nAChR containing a single point mutation (L247T) in the second transmembrane domain. As has been described previously [14], one consequence of this mutation is a reduction in the extent of receptor desensitization (Fig. 3). In contrast to the situation with the wild-type a7 nAChR ( Fig. 1), no significant difference was observed in the effect of temperature on the amplitude of responses evoked by acetylcholine on a7 nAChRs containing the L247T mutation (Fig. 3). This was the case, irrespective of whether a maximal concentration (30 mM) or an EC 50 concentration (0.4 mM) of acetylcholine or was used (Fig. 3, Table 1).

Effect of temperature on a7 nAChRs activated by 4BP-TQS
Whereas activation of wild-type a7 nAChRs by acetylcholine results in rapidly desensitizing responses, activation of a7 nAChRs by the allosteric agonist 4BP-TQS has a much slower onset and causes much less desensitization [10]. The effect of changes in temperature was examined on a7 nAChRs after activation by 4BP-TQS (Fig. 4). As reported previously [10], slowly activating responses were observed with minimal levels of desensitization. In contrast to the marked effect of temperature changes on the amplitude of currents evoked by acetylcholine on a7 nAChRs ( Fig. 1), temperature changes had little or no effect on responses to 4BP-TQS (Fig. 4). Lower temperature had no significant effect and higher temperature caused a relatively small but statistically significant (P = 0.002), reduction in current amplitude.
One possible explanation for these results is that the relatively small effect of high temperature on current amplitude in a7 nAChRs activated by 4BP-TQS might be due to instability of the compound at 37uC. To examine this possibility, control experiments were performed with solutions of 4BP-TQS that had been stored at 37uC for three hours and then cooled to room temperature. No differences in agonist responses were observed, when compared with freshly prepared solutions (data not shown).

Effect of temperature on receptor desensitization and deactivation
In addition to examining changes in current amplitude, the effect of altering temperature was examined on the rate of receptor desensitization (Fig. 5). Despite both high and low temperatures having opposing effects on current amplitudes on a4b2 and a7 nAChRs, changes in temperature had a consistent effect on the rate of receptor desensitization for a4b2 and a7 nAChRs. For both receptor subtypes (a4b2 and a7), an increase in the rate of desensitization was observed at 37uC and a decrease in the rate of desensitization at 4uC (Fig. 5). For both a7 L247T nAChRs and for wild-type a7 nAChRs activated by 4BP-TQS, despite changes in temperature having little or no effect on current amplitude (Figs. 3 and 4), an increase in the rate of receptor desensitization was observed at 37uC, consistent with the effects seen with a4b2 and wild-type a7 nAChRs (Fig. 5). There was also evidence of a reduction in the rate of desensitization at 4uC for both a7 L247T nAChRs and for wild-type a7 nAChRs activated by 4BP-TQS, consistent with the effects seen with a4b2 and wild-type a7 nAChRs, but this was not statistically significant (Fig. 5, Table 2). The influence of changes in temperature was also examined on the rate of receptor deactivation after removal of agonist (Fig. 5E). It was not possible to measure this parameter for of a7 nAChRs activated by acetylcholine, due to the very rapid desensitization but, in all other cases, changes in temperature had a consistent effect (Table 2), with an increased rate of deactivation observed at 37uC and a reduced rate at 4uC.

Effects of temperature are reversible
For all experiments, the effects of raising or lowering temperature on agonist responses were normalized to responses    recorded at RT on the same oocyte. In addition, after responses had been recorded at a temperature above or below RT, further responses were measured on the same oocyte at RT. In all cases, changes in current amplitude or in the rate of receptor desensitization were found to be reversible (data not shown).

Discussion
It has been known for several decades that changes in temperature can influence the properties of nAChRs. For example, a series of  Representative traces are shown illustrating responses obtained at RT (black), 4uC (blue) and 37uC (red). Current traces obtained at each temperature have been normalized to the same peak response. In each case, the response showing the fastest rate of desensitization was observed at 37uC and the slowest rate of desensitization was observed at 4uC. Responses are from a7 nAChRs with 3 mM acetylcholine (A), a4b2 nAChRs with 1 mM acetylcholine in calciumcontaining saline (B), a7 L247T nAChRs with 30 mM acetylcholine (C) and a7 nAChRs with 10 mM 4BP-TQS (D). Rates of receptor deactivation after removal of agonist were also influenced in a consistent manner by changes in temperature (faster at 37uC and slower at 4uC). Representative traces from a7 L247T nAChRs with 30 mM acetylcholine are illustrated (E) and are typical of results from all receptor/agonist combinations studied (see Tables 1 and 2  studies conducted in the early 1970s demonstrated that end-plate currents recorded at the frog neuromuscular junction are influenced by temperature [16,17,18,19,20]. Typically, the effects observed include increased rates of decay of end-plate currents with increased temperature. This is consistent with the increase in the rate of receptor desensitization that we have observed with both a4b2 and a7 nAChRs. Similarly, previous studies conducted with rat diaphragm preparations reported similar effects of temperature on rates of decay of end-plate currents [21]. Studies of native neuronal nAChRs expressed in cultured PC12 cells also demonstrated that rates of recovery from desensitization were faster at higher temperatures [22]. Thus, it seems that changes in temperature appear to have a consistent effect on the rate of agonist-induced desensitization and the decay of end-plate potentials.
An increase in current amplitude with increased temperature, similar to the effect we have observed with a4b2 nAChRs, has been reported recently for recombinant P2X 3 receptors [23]. However, in contrast to the findings reported here, the development of desensitization in P2X 3 receptors has been reported to be independent of temperature, at least between 25uC and 40uC [23]. A similar increase in current amplitude with increased temperature has also been reported for NMDA-type glutamate receptors [24]. Conversely, an increase in current amplitude in response to decreased temperature, similar to the effect that we observe with a7 nAChRs, has been reported with an acetylcholine-gated chloride channel from the parietal ganglion of the pond snail Lymnaea [25]. Thus, there are precedents for changes in temperature having opposing effects on the amplitude of ion channel currents, albeit from different classes of ligandgated ion channels. In addition, we have recently become aware of a paper, published after the submission of this manuscript, examining the influence of higher temperature on a7 nAChRs [26]. Higher temperature resulted in reduced responses to acetylcholine when co-applied with the positive allosteric modulator PNU-120597, consistent with our results obtained with either acetylcholine or 4BP-TQS applied alone.
We have found that introduction of a single point mutation (L247T) in the a7 nAChR completely abolishes the effect of temperature on current amplitude, whilst retaining an effect of temperature on the rate of desensitization of the macroscopic response. This observation adds to the many effects that have been reported for the L247T mutation in a7 nAChRs. These include an increase in agonist potency, reduction in receptor desensitization, increased spontaneous openings, the conversion of competitive antagonists into agonists and the conversion of allosteric potentiators into allosteric agonists [10,14,27,28]. As has been discussed previously [10], the non-desensitizing responses observed when a7 is activated by 4BP-TQS resembles that of acetylcholine responses on a7 L247T nAChRs. Changes in temperature had a greatly reduced effect on the amplitude of a7 responses in response to 4BP-TQS compared to responses to acetylcholine. Indeed, no significant effect was observed when responses at 4uC were compared to those at RT (Fig. 4). This is consistent with previous evidence [10] indicating that acetylcholine and 4BP-TQS cause activation of a7 nAChRs through different mechanisms of action.
It has been reported previously that exposure of some nAChRs to low temperature for several hours can facilitate more efficient protein folding and assembly. This was first demonstrated for the muscle-type nAChR from the marine ray Torpedo, expressed in cultured mammalian cell lines [29] and can be explained by the fact that proteins in cold water fish have not evolved to fold efficiently at 37uC. Similar effects have been reported for insect nAChRs expressed in cultured cell lines [30,31,32]. In fact, even mammalian nAChRs have been reported to fold and assemble more efficiently at lower temperature [33,34]. However, in contrast to these long-term effects occurring over several hours, it is unlikely that increased efficiency of subunit folding and assembly explains the effects reported here. The effects of temperature on current amplitude examine in this study are both very rapid (occurring within seconds) and are reversible, suggesting that the effects observed are a consequence of a change in the thermodynamic properties of already assembled cell-surface receptors, rather than a change in the efficiency of receptor folding and assembly. In contrast to the experiments examining agonist responses at physiological temperature, the rationale for examining responses of human nAChRs at 4uC is probably less obvious. Although responses measured at 4uC do not have direct physiological relevance for human receptors, the fact that lowering temperature to 4uC has opposing effects on a4b2 and a7 nAChRs provides evidence that these two nAChR subtypes have differing biophysical properties.
It is unclear why changes in temperature should have opposing effects on current amplitude in two closely related ion channels (a4b2 and a7 nAChRs) or why these effects should be largely or completely abolished by either a single point mutation or by activation by an allosteric agonist, rather than by the conventional orthosteric agonist, acetylcholine. It appears, however, that these effects are not a consequence of changes in the rate of receptor  desensitization observed during macroscopic oocyte responses, since changes in temperature were found to have a broadly consistent effect on this parameter. In addition, changes in temperature had a consistent effect on the rate of receptor deactivation after removal of agonist (faster at 37uC and slower at 4uC), presumably reflecting changes in the off-rate of agonists from their binding sites. It is possible that receptors can adopt multiple open or desensitized states, and that entry into these various states occurs at different rates and can be affected differently by phenomena such as changes in temperature, mutagenesis or by allosteric modulators. This idea is consistent with models that have been advanced to explain mechanisms of allosteric modulation of nAChRs [35]. It is possible that the changes in current amplitude that we have observed with a4b2 and a7 nAChRs may be a consequence of temperature-induced changes in single-channel conductance. Indeed, it has been reported that the conductance of muscle-type nAChRs increases with increasing temperature [36,37,38,39]. As has been discussed previously in connection with the effects of temperature on functional properties of acetylcholine-gated chloride channels from the pond snail Lymnaea [25], alternative possibilities for the effects that we have observed include temperature-dependent changes in single-channel kinetics and/ or changes in the affinity of agonist binding. Recently, evidence has been obtained indicating that the gating rate constants of muscle nAChRs are altered by temperature [40]. Further work will be required to establish the precise mechanism of action of the temperature dependent effects that we have observed on the amplitudes of whole-cell responses in different subtypes of neuronal nAChRs. We can conclude, however, that changes in temperature can have opposing effects on the amplitude of nAChR whole-cell responses, whilst having a similar effect on the rate of desensitization of the macroscopic agonist-evoked response.

Materials and Methods
Plasmids, site-directed mutagenesis and cRNA synthesis Plasmid pSP64GL constructs containing human a4, a7 and b2 cDNA have been described previously [41] as has a plasmid containing the a7 cDNA with the L247T mutation [10]. All pSP64GL plasmids were linearized with BamHI and purified with QIAquick PCR purification kit (Qiagen). In vitro synthesis of cRNA was performed using the SP6 mMessage mMachine SP6 kit (Ambion).

Two-electrode voltage-clamp recording in Xenopus oocytes
Adult female Xenopus laevis frogs were obtained from the European Xenopus Resource Centre (University of Portsmouth). Oocytes were isolated and defolliculated as described previously [42] following procedures that have been approved by both UCL's Biological Services Management Group and the UK Home Office (under licences PIL70/23585 and PPL70/06819). Heterologous expression was achieved by injection of cRNA (8-12 ng) into the oocyte cytoplasm. Oocytes were injected in a volume of 32.2 nl using a variable volume Nanoject II microinjector (Drummond Scientific). cRNAs encoding a4 and b2 subunits were injected in a 1:1 ratio in a total injection volume of 32.2 nl/oocyte. After injection, oocytes were incubated at 18uC in a modified Barth's solution containing: 88 mM NaCl, 1 mM KCl, 0.82 mM MgCl 2 , 0.77 mM CaCl 2 , 2.4 mM NaHCO 3 , and 15 mM Tris, 50 mg/l tetracycline, 50 mg/l penicillin and 50 mg/l streptomycin (pH 7.5 with HCl). Experiments were performed on oocytes 3 to 5 days of injection. Oocytes were placed in a recording chamber (internal diameter, 3 mm), which was continuously perfused with a modified Ringer solution (115 mM NaCl, 2.5 mM KCl, 10 mM HEPES, 1.8 mM BaCl 2 or CaCl 2 ). Oocytes were impaled by two microelectrodes filled with 3 M KCl (0.5-1.5 MV) and voltage-clamped at 260 mV using a GeneClamp 500 amplifier (Axon Instruments). Drugs were applied by switching between control and drugcontaining solution using a BPS-8 valve control system (ALA Scientific). Agonists were applied for either 5 seconds (acetylcholine) or up to 40 seconds (4BP-TQS) with 2 minute (acetylcholine) or 8 minute (4BP-TQS) intervals between applications. The experiments were performed at room temperature (RT; approx. 21uC). To detect effects of changed temperature on kinetics of nAChRs the modified Ringer solution was warmed/cooled to 37uC or 4uC in water bath or on ice. Agonist solutions were prepared fresh and warmed or cooled for 10-15 min immediately prior to use. In control experiment, it was confirmed that warming agonist solutions, even for 3 hours did not alter agonist potency. To examine effects of temperature, oocytes were continuously perfused with warm/cold modified Ringer solution for 40 sec prior to agonist application. Peak current amplitudes evoked by warm/cold acetylcholine were normalized to the peak amplitudes at RT from the same oocyte. To examine reversibility of any temperature induced changes, oocytes were then perfused with RT Ringer for 5 mins, followed by application of acetylcholine at RT. Peak amplitudes, the rise time and time to decay to half of the maximum amplitude were measured using Clampfit 9.2 (Molecular Devices). Effects of temperature on current amplitude were determined by measuring peak responses, however, results were not significantly different when responses were measure as area above the curve/total charge transfer. Statistical significance was determined using with SigmaStat (Aspire Software International) using non-paired Student's t-test. The effects of raising or lowering temperature on agonist-evoked responses were normalized to responses recorded at RT on the same oocyte. All data are reported as mean 6 S.E.M.