Neuroprotective Effect Of [Gly¹⁴]-humanin Against Amyloid β Protein-Induced Neurotoxicity And Its Possible Mechanism: Electrophysiological And Intracellular Calcium Imaging Study

Alzheimer's disease (AD) is the most prevalent neurodegenerative disease in the elderly leading to progressive loss of memory and cognitive deficits. One of the predominant neuropathological features of AD is the presence of high density of senile plaques in the brain. The main constituent of senile plaques is amyloidβ-protein (Aβ), which consisting of 39-43 amino acids and coming from the proteolysis of amyloid precursor protein (APP). The neurotoxicity of Aβhas been widely reported in vivo and in vitro, including the impairment of synaptic plasticity, such us long-term potentiation (LTP). The mechanisms underlying Aβ-neurotoxicity are complex and unclear so far but may involve the vulnerability of glutamatergic and GABAergic system. Considerable evidences suggest that Aβand its peptide fragments influence cellular homeostasis and neuronal signaling through modulation of ion channel function.Hippocampus in the central nervous system has been widely considered to be a crucial center for learning and memory. Hippocampal LTP is an activity-dependent increase in the synaptic response and has been accepted as a popular electrophysiological model for the cellular basis of learning and memory, owing to the close relationship between the hippocampal LTP impairment and the animal cognitive deficient. Further, N-methyl-D-aspartate (NMDA) receptor plays a pivotal role in the LTP induction of CA1 pyramidal neurons because of its high permeability to Ca2+ ions; changes in the efficacy of synaptic inhibition mediated by GABAA receptors are believed to play central roles in certain forms synaptic plasticity, and some reports indicate the potential role of Aβin the deficits of the GABAergic system in the development and pathology progresses of AD. Hence the modulation of LTP-related receptors such as NMDA and GABA receptors is a potential target for the neurotoxicity of Aβon LTP. At the same time, calcium ion is one of the most important intracellular second messengers in the brain, being essential for a variety of neuronal functions such as neuronal development, synaptic transmission and plasticity, and the regulation of various metabolic pathways. It is well known that Aβinduced Ca2+ influx is proposed as the initial event of Aβ-induced multiple events. Although the mechanisms by which Aβincreased [Ca2+]i are still not well known, the disruption of Ca2+ homeostasis is the primary event in Aβneurotoxicity. Thus, it will be very important to search for effective medicines or measures that can protect NMDA receptors, GABA receptors, and intracellular Ca2+ homeostasis against Aβ-induced neurotoxicity in the prevention and treatment of AD.An important clue in the development of AD therapy is the finding of humanin (HN), a novel short polypeptide, and its potent derivative [Gly14]-humanin (HNG). HN can prevent neuronal cell death caused by various AD-relevant insults such as neurotoxic Aβprotein, and is the only factor that is effective in suppressing various types of AD-related neuronal death so far compared with many neurotrophic factors. The fact that HNG reverses the impairment of memory induced by Aβpeptide suggests that HNG is a promising candidate in the treatment of AD. However, it is still short of in vivo and in vitro electrophysiological evidence for the protection of HN on synaptic plasticity such as LTP and LTP-related postsynaptic receptors/channels, and the molecular mechanisms that underling the neuroprotective function of HN also remain largely unknown.Therefore, the purposes of the present study were as follows: (1) Examining the effects of intracerebroventricular (i.c.v.) injection of HNG on the different Aβfragments-induced suppression of LTP in the rat hippocampal CA1 region in vivo with field potential recording technique; (2) Investigating whether HNG affects the effects of different Aβfragments on LTP induction-related postsynaptic ligand-gated excitatory receptor channels (NMDAR) and inhibitory receptor channels (GABAR) by using whole-cell patch clamp technique in cultured primary rat cortical neurons; (3) Observing the effects of different Aβfragments on [Ca2+]i, and investigating the possible neuroprotective effects of HNG against Aβ-induced disruption of Ca2+ homeostasis in cultured primary rat cortical neurons by using calcium imaging via laser-scanning confocal imaging system; (4) Clarifying whether tyrosine kinases are involved in the neuroprotective function of HNG by using a specific tyrosine kinase inhibitor, Genistein; and (5) Identifying whether Aβ31-35 is a shorter active center than Aβ25-35 in full length of Aβmolecule by comparing the effects of Aβfragments in the experiments of LTP, whole-cell currents and [Ca2+]i imaging.PartⅠ:[Gly14]-Humanin Rescues Long-Term Potentiation from AmyloidβProtein-Induced Impairment in the Rat Hippocampal CA1 Region In VivoThe present study observed the effects of intracerebroventricular (i.c.v.) injection of HNG on the different Aβfragments-induced suppression of hippocampal LTP in area CA1 of urethane anesthetized rats, and explored the probable tyrosine kinase mechanism by which HNG protect Aβ-induced LTP impairment. The results showed that: (1) The amplitudes of the field excitatory postsynaptic potentials (fEPSPs) in control group (i.c.v. injection of saline) increased to 215.1±8.6% immediately after applying high-frequency stimulus (HFS), as compared to that measured before HFS and set arbitrarily as 100%, and remains at 163.3±8.5% 60 min after HFS (n=8). (2) i.c.v. injection of 20 nmol Aβ25-35 and Aβ31-35 had no effect on baseline synaptic transmission, but both of them similarly and significantly depressed the induction of LTP, with fEPSPs of 111.4±8.9% (n=7) and 103.3±7.4% (n=6) at 60 min after HFS, respectively. (3) Application of 5 nmol HNG alone caused no significant effect on baseline synaptic transmission and HFS-induced LTP compared with control, and the LTP at 60 min post HFS was 159.2±6.4%(n=5), which is close to the value of 163.3±8.5% (n=8) found in control group (P>0.05). However, co-injection of HNG and Aβ25-35 reversed the depression of LTP by Aβ25-35 alone in a dose-dependent manner. After i.c.v. co-injection of 0.2 nmol, 1 nmol and 5 nmol HNG with 20 nmol Aβ25-35, the LTP at 60 min post HFS increased from 111.4±8.9% (n=7) in Aβ25-35 alone group to 114.9±6.4% (n=9, P>0.05), 130.9±6.5% (n=8, P<0.01) and 163.7±8.6% (n=7, P<0.01) in three HNG groups, respectively. (4) Similar to Aβ25-35, co-injection of 5 nmol HNG with 20 nmol Aβ31-35 did not affect baseline synaptic transmission, but effectively reversed Aβ31-35-induced LTP suppression, and the LTP value was 158.6±6.1% (n=5) 60 min after HFS, significantly larger than the effect when using 20 nmol Aβ31-35 alone (103.3±7.4%, n=6, P<0.01). (5) Separate application or co-application of HNG and Aβ25-35 showed no significant effect on paired pulse-evoked facilitation. (6) Pretreatment with 200 nmol Genistein obviously attenuated the protection of HNG against Aβ25-35-induced LTP impairment, and the LTP value decreased from 163.7±8.6% (n=7) to 123.0±7.7% (n=4, P<0.01) 60 min after HFS. The former was very close to the value in normal control group (164.3±5.3%, n=7, P>0.05), while the later has been close to the value in Aβ25-35 alone group (111.4±8.9%, n=7, P>0.05).These results demonstrate that: (1) Aβ31-35 may be a shorter active fragment of Aβbecause of the similar effects of Aβ31-35 and Aβ25-35; (2) HNG could dose-dependently protect against the neurotoxic Aβ-induced hippocampal LTP impairment; and (3) the tyrosine kinase pathway was involved in the neuroprotective action of HNG. All of these results suggest that HNG might be one of the promising candidates for the treatment of AD in the future.Part II:[Gly14]-Humanin Abolishes the Effects of AmyloidβProtein on NMDA and GABAA Receptor-Mediated Currents in Cultured Primary Rat Cortical NeuronsThe proper modulation of excitatory and inhibitory synaptic transmission is critical for brain function including synaptic plasticity such as LTP. The present study extended our investigation of in vivo LTP by observing whether HNG can modulate the effects of different Aβ fragments on LTP induction-related postsynaptic ligand-gated excitatory receptor channels (NMDAR) and inhibitory receptor channels (GABAR) by using whole-cell patch clamp in cultured primary rat cortical neurons. The results showed that: (1) Acute application of Aβ25-35, Aβ31-35 or Aβ35-31 alone did not induce any membrane current, even at a higher concentration (10μM). (2) The NMDA-induced whole cell currents (INMDA) were reversibly inhibited by acute application of 0.1μM, 1μM and 10μM Aβ25-35 and Aβ31-35 in a dose-dependent manner. INMDA decreased to 86.0±6.7% (n=14, P<0.01), 74.9±8.2% (n=15, P<0.01) and 61.0±10.6% (n=13, P<0.01) in Aβ25-35 groups, 82.6±8.0% (n=11, P<0.01), 72.1±10.8% (n=14, P<0.01) and 60.6±7.8% (n=14, P<0.01) in Aβ31-35 groups compared with 100% in control, respectively. (3) HNG dose-dependently reversed the inhibitory effect of Aβ25-35 and Aβ31-35 (10μM) on INMDA. INMDA changed from 61.0±10.6% (n=13) in Aβ25-35 alone group to 58.1±5.5% (n=15, P>0.05), 80.7±8.8% (n=12, P<0.01) and 96.5±6.0% (n=12, P<0.01) at 0.1 nM, 1 nM and 10 nM HNG, respectively; and from 60.6±7.8% (n=14) in Aβ31-35 alone group to 56.5±6.2% (n=15, P>0.05), 74.3±4.5% (n=14, P<0.01) and 93.9±11.0% (n=13, P<0.01), respectively. (4) Meanwhile, pretreatment with 0.1μM, 1μM and 10μM Aβ25-35 and Aβ31-35 reversibly suppressed the GABA-induced whole-cell currents (IGABA) in a concentration-dependent manner. IGABA decreased to 96.2±7.3% (n=12, P>0.05), 67.7±6.8% (n=14, P<0.01) and 55.5±12.2% (n=13, P<0.01) in Aβ25-35 groups, 91.0±8.0% (n=11, P<0.05), 71.3±9.1% (n=21, P<0.01) and 58.2±9.7% (n=19, P<0.01) in Aβ31-35 groups compared with 100% in control, respectively. (5) HNG concentration-dependently reversed Aβ-induced suppression. IGABA changed from 67.7±6.8% in Aβ25-35 alone group to 67.8±9.6% (n=14, P>0.05), 80.8±10.8% (n=12, P<0.01), 90.2±10.9 (n=12, P<0.01) and 99.6±9.4% (n=16, P<0.01) at 0.01 nM, 0.1 nM,? 1 nM and 10 nM HNG, respectively. IGABA changed from 71.3±9.1% in Aβ31-35 alone group to 66.9±8.8% (n=14, P>0.05), 83.8±8.4% (n=11, P<0.01), 92.2±8.8% (n=12, P<0.01) and 102.5±9.7% (n=15, P<0.01) at 0.01 nM, 0.1 nM,? 1 nM and 10 nM HNG pretreatment, respectively. (6) Aβ25-35 and Aβ31-35 have no significant effect on the higher concentration (100μM and 1000μM) of GABA-induced whole cell currents. 100μM GABA induced IGABA were still 99.9±5.9% (n=12, P>0.05), 96.1±7.2% (n=11, P>0.05) and 98.3±5.6% (n=12, P>0.05) after application of 10μM, 20μM or 40μM Aβ25-35. Similarly, after pretreatment with 10μM, 20μM or 40μM Aβ31-35, 100μM GABA induced IGABA were still 97.8±5.3% (n=18, P>0.05), 99.3±9.5% (n=12, P>0.05) and 95.1±9.6% (n=13, P>0.05), respecitively. (7) Pretreatment with Genistein (100μM), a tyrosine kinase inhibitor, nearly completely abolished the protective action of HNG (10 nM) on Aβ25-35 and Aβ31-35 induced inhibition of INMDA and IGABA, the values being 63.0±8.0% (n=14, P<0.01) and 58.9±4.8% (n=14) for INMDA, 70.5±11.4% (n=10) and 70.5±11.4% (n=10) for IGABA. (8) Aβ35-31, the reversed sequence of Aβ31-35, showed no effect on the currents induced by NMDA or GABA.In conclusion, the present study showed an acute modulation of Aβ25-35 and Aβ31-35 on NMDA receptors and GABAA receptors in cultured primary rat cortical neurons. These results may partly explain the reason of Aβ-induced suppression of hippocampal LTP, in which synaptic NMDA receptor might play important roles. In addition, our results provide an electrophysiological evidence for the Aβ-neurotoxicity on the glutamatergic/GABAergic system and the neuroprotective function of HNG on Aβ-neurotoxicity. The neuroprotection of HNG strongly suggests the potential clinical applications of HNG in novel AD therapies.PartⅢ:Effects of [Gly14]-Humanin on AmyloidβProtein-Induced Calcium Influx in Cultured Primary Rat Cortical NeuronsThe present study, by utilizing calcium imaging via laser-scanning confocal imaging system, observed the effects of different Aβfragments on [Ca2+]i, and investigated the neuroprotective effects of HNG against Aβ-induced disruption of Ca2+ homeostasis and its possible tyrosine kinase mechanism in cultured primary rat cortical neurons. The results showed that: (1) Both Aβ25-35 and Aβ31-35 significantly increased the intracellular calcium levels of the cultured primary rat cortical neurons. After application of 25μM Aβ25-35 or Aβ31-35 to cultured primary rat cortical neurons, the relative fluorescent intensity of neurons obviously increased from 100% in control to 174.3±10.7% (n=25, P<0.01) and 172.7±6.6% (n=40, P<0.01), respectively. However, 25μM Aβ35-31, the reverse peptide of Aβ31-35, had no effect on [Ca2+]i elevation. In addition, the effect of Aβ25-35 and Aβ31-35 on [Ca2+]i elevation, at the same concentration, does not show any statistical difference. (2) Neuroprotective polypeptide HNG blocked the [Ca2+]i elevation induced by Aβ25-35 or Aβ31-35 in a concentration-dependent manner. With the increasing of HNG concentration, Aβ25-35- or Aβ31-35-induced elevation in [Ca2+]i decreased. After pretreatment with 0.1 nM, 1 nM, 10 nM and 100 nM HNG, Aβ25-35-evoked relative fluorescent intensity decreased from 174.3±10.7% (n=25) to 173.6±9.0% (n=21, P>0.05), 144.6±8.2% (n=39, P<0.01), 120.0±6.3% (n=38, P<0.01), and 105.4±6.0% (n=33, P<0.01), respectively. Aβ31-35-evoked relative fluorescent intensity decreased from 172.7±6.6% (n=40) in Aβ31-35 alone to 164.4±8.8% (n=16, P<0.01), 137.1±6.7% (n=19, P<0.01), 114.4±9.9% (n=23, P<0.01) and 103.1±6.2% (n=29, P<0.01), respectively. (3) Pretreatment with Genistein (100μM), a specific tyrosine kinase inhibitor, essentially abolished the protective effect of HNG (10 nM) on Aβ25-35 or Aβ31-35 induced [Ca2+]i elevation. Compared with co-application of HNG and Aβ25-35 or Aβ31-35, the relative fluorescent intensity in co-application of Genistein, HNG and Aβ25-35 or Aβ31-35 increased to 178.3±9.9% (n=29, P<0.01) and 168.7±8.2% (n=27, P<0.01) from 120.0±6.3% (n=38) and 114.4±9.9% (n=23), respectively, similar to the values in Aβ25-35 or Aβ31-35 alone group.These results demonstrate that: (1) Both Aβ25-35 and Aβ31-35 can similarly elevate [Ca2+]i in cultured primary rat cortical neurons, suggesting that Aβ-neurotoxicity is related with Aβ-induced intracellular calcium overloading. (2) HNG can inhibit Aβ25-35 and Aβ31-35-induced elevation of [Ca2+]i in a concentration-dependent manner, suggesting that the protective action of HNG on the cultured primary rat cortical neurons was, at least in part, mediated by attenuating the intracellular calcium overloading induced by Aβ. (3) Genistein, a specific tyrosine kinase inhibitor, abolished the suppression of HNG on Aβ25-35 or Aβ31-35 induced [Ca2+]i elevation, indicating that certain tyrosine kinase(s) is (are) involved in the protective effects of HNG on Aβinduced [Ca2+]i elevation. (4) The similar [Ca2+]i elevation effects of Aβ31-35 and Aβ25-35 in various groups, including Aβfragments alone and Aβplus HNG and Genistein, supports the hypothesis we proposed previously that Aβ31-35 may be a more shorter fragment of the full length of Aβmolecule.In conclusion, the present study, by using field potential recording and whole-cell patch clamp technique, observed the modulation of HNG on the effects of Aβfragments on hippocampal LTP in vivo, INMDA and IGABA in cultured primary rat cortical neurons. The possible mechanisms underlying the neuroprotection of HNG were also explored by using Genistein and Ca2+ image technique. The results demonstrated for the first time that HNG could protect against the neurotoxic Aβ-induced hippocampal LTP deficit, reverse the impairment of LTP-related NMDA and GABA receptor currents by Aβand inhibit Aβ-induced [Ca2+]i elevation. Moreover, our results indicated that the neuroprotective actions of HNG may be involved in the tyrosine kinase pathway. Taken all together, all of our experiments in vivo and in vitro provide strong electrophysiological evidences for the neuroprotection of HN and suggest that HNG might be one of the promising candidates for the treatment of AD in the future.