From Department of Neurology Medical School of Hannover

(Direktor: Prof. Dr. R. Dengler)

(Advisor: PD Dr. K. Krampfl)

Antagonists of AMPA-type channels: A Patch Clamp Study

DISSERTATION

To achieve the doctor degree of medicine

at the Medical School of Hannover

Submitted by

Lingjing Jin

From Shanghai, P.R. China

Hannover 2007 Angenommen vom Senat der Medizinischen Hochschule Hannover am 25.09.2007 Gedruckt mit Genehmigung der Medizinischen Hochschule Hannover Präsident: Prof. Dr. Dieter Bitter-Suermann Betreuer : PD Dr. med. Klaus Krampfl Referent : Prof. Dr. med. Christoph Fahlke Korreferent: Prof. Dr. med. Günter Reuter

Tag der mündlichen Prüfung: 25.09.2007 Promotionsausschussmitglieder: Prof. Dr. Hermann Müller-Vahl Prof. Dr. Hans-Werner Künsebeck Prof. Dr. F.Th. Schuppert

Contents Abbreviations...... 1 Introduction ...... 2 1.1 ALS and the role of chronical excitotoxity...... 2 1.2 Glutamate receptors ...... 3 1.3 AMPA receptor...... 4 1.3.1 Structure and function...... 4 1.3.2 The electrophysiological characteristics of AMPA receptors...... 6

1.3.3 GluR2flipGQ ...... 7

1.3.4 Non-desensitizing GluR2L504Y ...... 7 1.4 AMPA receptor antagonists as candidate compounds for neuroprotection...... 8 1.4.1 ...... 8 1.4.2 Riluzole...... 9 1.4.3 ...... 9 1.4.4 ...... 10 1.4.5 Phenobarbital ...... 10 1.4.6 Valproic acid ...... 11 1.4.7 CNQX...... 11 1.4.8 RPR119990...... 12 1.5 Purpose ...... 12 Materials and methods...... 13 2.1 Solutions ...... 13 2.2 Cell culture and transfection ...... 14 2.3 Patch-clamp technique...... 14 2.3.1 Principle...... 14 2.3.2 Equipment...... 15 2.3.3 Pipette Fabrication and Mechanical Setup...... 17 2.3.4 Patch Technique...... 18 2.3.5 Fast application technique...... 19 2.3.6 Application protocols...... 20 2.3.7 Data acquisition and analysis...... 22 2.4 Statistic method...... 22 Results ...... 23 3.1 Effect of minocycline on AMPA-type receptor channels...... 23

3.1.1 Minocycline: GluR2L504Y receptor channels ...... 23

3.1.2 Minocycline: GluR2flipGQ receptor channels...... 25 3.2. Effect of riluzole on AMPA-type receptor channels ...... 29

3.1.1 Riluzole: GluR2L504Y receptor channels...... 29

3.1.2 Riluzole: GluR2flipGQ receptor channels...... 32 3.3. Effect of memantine on AMPA-type receptor channels...... 35

3.3.1 Memantine: GluR2L504Y receptor channels ...... 35

3.3.2 Memantine: GluR2flipGQ receptor channels...... 37 3.4. Effect of ketamine on AMPA-type receptor channels...... 39

3.4.1 Ketamine: GluR2L504Y receptor channels...... 39

3.1.2 Ketamine: GluR2flipGQ receptor channels...... 39

Contents

3.5. Effect of Phenobarbital on GluR2L504Y receptor channels ...... 41

3.6. Effect of valproic acid on GluR2L504Y receptor channels...... 44

3.7 Effect of CNQX on GluR2L504Y receptor channels...... 46 3.8 Effect of RPR119990 on AMPA-type receptor channels...... 48

3.8.1 RPR119990: GluR2L504Y receptor channels...... 48

3.8.2 RPR119990: GluR2flipGQ receptor channels...... 50 3.9. Comparison of the block effects of the different compounds tested...... 53 Discussion...... 54 1. Minocycline ...... 54 2 Riluzole...... 56 3. Memantine ...... 57 4. Ketamine...... 58 5. Phenobarbital ...... 58 6. Valproic acid ...... 60 7. CNQX...... 60 8. RPR119990...... 61 Summarize ...... 63 Reference...... 65 Curriculum vitae ...... 73 Declaration (Erklärung nach §2 Abs. 2 Nr. 5 und 6)...... 75 Acknowledgements...... 76

2

Abbreviations Abbreviations

ALS Amyotrophic lateral sclerosis

AMPA α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

NMDA N-methyl- D –aspartate

GluR glutamate receptors

Pmax maximal amplitude

Cdes steady-state-current

TTP time-to-peak (20-80% time to peak)

τdes time constant for desensitization

τdea time constant for deactivation

τB time constant current decay of block effect

AUC Area-Under-Curve rAmp relative maximal amplitude (explain on page 20) rTTP relative time to peak (explain on page 20) rAUC relative area of under curve (explain on page 20) rCdes relative steady-state-current (explain on page 20)

HEK293 cells human embryonic kidney 293 cells

IC50 the half maximal inhibitory concentration

1

Introduction

Introduction

1.1 ALS and the role of chronical excitotoxity

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by progressive loss of upper motor neurons in the motor cortex and lower motor neurons in the brainstem and spinal cord, which results in progressive paralysis of bulbar, respiratory and limb muscles, while sensory and cognitive functions are preserved in most cases. ALS typically affects adults in mid-life, with an incidence of 1–2/100,000/year1. Neurological examination reveals lower motor neuron signs (paralysis, muscle atrophy, fasciculation) and upper motor neuron signs (paralysis, hyperreflexia, Hoffmann and Babinski signs, increased muscle tone). The progressive muscle weakness results in death after 2–5 years following the onset of the disease, mostly due to respiratory failure. Treatment remains essentially supportive, since no substantial therapy is available. Riluzole, the only drug proven to slow down disease progression, prolongs survival of ALS patients by a few months 2, 3.

The pathogenic mechanism of the selective motor neuron death in ALS is not yet fully elucidated. In 5–10% of cases familial ALS occurs, predominantly showing an autosomal dominant inheritance. In 20% of familial cases (which represents only 1–2% of all cases), mutations in the superoxide dismutase-1 (SOD1) on chromosome 21q were identified4. Mutations in alsin5, in senataxin6, or in the gene encoding the vesicle-trafficking protein (VAPB)7 were also reported. But the majority of ALS patients has no affected family members and is considered to have sporadic ALS. A number of possible mechanisms have been proposed8-11. These include oxidative stress, axonal strangulation from neurofilamentous accumulations, toxicity from intracellular protein aggregates, mitochondrial dysfunction, inflamation, decreased availability of growth factors and excitotoxic death of motor neurons.

Excitotoxicity is undoubtedly one of the most robust pathogenic mechanisms, which is neuronal degeneration induced by overstimulation of glutamate receptors and can occur when the extracellular glutamate concentration increases or when the postsynaptic neuron becomes vulnerable to normal glutamate levels. The glutamate is the major excitatory in the mammalian with low extracellular glutamate concentration (approximately 0.6 µM12), in spite of a high intracellular concentration of

2

Introduction approximately 10 mM13 and frequent release of glutamate at glutamatergic . Elevation of the extracellular glutamate concentration to 2–5 µM is considered sufficient to cause degeneration of neurons through excessive stimulation of glutamate receptors14, 15. This can occur when the release from presynaptic terminals is augmented or when the re-uptake from the synaptic cleft is insufficient. In addition, lethal injury to neurons, or microglia can lead to the release of the intracellular glutamate content. At the postreceptor level, energy depletion renders neurons more vulnerable to , resulting in neuronal damage in the presence of normal glutamate levels16, 17. Motor neurons are extremely sensitive to excitotoxicity, as they have not only a limited capacity to buffer Ca2+ rises due to the low amount of Ca2+-buffering proteins in their cytosol, but also have a high proportion of Ca2+-permeable AMPA-type receptor channels-type glutamate receptors and stimulation of these receptors leads to selective motor neuron death18-20.

1.2 Glutamate receptors

Glutamate receptors can be divided into two functionally distinct categories: those that mediate their effects via coupling to G-protein second messenger systems, the metabotropic glutamate receptors (mGluRs) and ionotropic ligand-gated channels21. The ionotropic receptors can further be subdivided into three classes according to their preferred synthetic agonist22: AMPA (α-amino-3- -hydroxy-5-methyl-4-isoxazolepropionic acid), NMDA (N-methyl-D-aspartate) and kainate receptors (but AMPA receptors are also responsive to kainate). AMPA receptors are permeable to Na+, K+ and in a variable degree to Ca2+. NMDA receptors mediate the late component of excitatory transmission23 and play a key role in the induction of synaptic plasticity24. The role of kainate receptors in physiological and pathological conditions is less clear. Metabotropic glutamate receptors (mGluRs) appear to modulate excitotoxicity in motor neurons, but their exact role remains to be elucidated.

Among those glutamate receptors AMPA receptors are the most important glutamate receptors to mediate fast excitatory transmission. It was demonstrated by different groups that motor neurons in culture are particularly susceptible to , especially to AMPA receptor agonists18, 25. Ca2+ influx through Ca2+-permeable AMPA receptors was crucial for triggering motor neuron death19, 20. Intrathecal or intraspinal administration of AMPA receptor agonists has been shown to induce motor neuron degeneration in animals, whereas NMDA failed to damage spinal motor neurons26-28. In organotypic rat spinal cord cultures (spinal cord slices from postnatal

3

Introduction rats, which can be kept in culture for several weeks), motor neurons were also proved to be vulnerable to AMPA receptor-mediated excitotoxicity29, 30. Direct application of AMPA receptor agonists resulted in selective motor neuron loss, which could be prevented by antagonists of AMPA receptors. For that reason, we focus on AMPA-type receptors when investigating the pharmacological interactions of several candidate compounds for clinical neuroprotection.

1.3 AMPA receptor

1.3.1 Structure and function

Each mature AMPA-type receptor channel is a tetramer composed of a variable association of four individual subunits (GluR1–4) 31-35. The schematic topology36 of an AMPA-type receptor subunit is illustrated in Fig.1.1. The molecular architecture of each AMPA-type receptor subunit is very similar; each comprises ~900 amino acids and has a molecular weight of ~105 kDa37. There is approximately

70% sequence homology between encoding each subunit, although genes may undergo in two distinct regions, resulting in subunits that have either long or short C termini, and flip or flop variants in an extracellular domain36.

Fig.1.1 Schematic showing of topology of an AMPA receptor subunit. Each subunit consists of an extracellular N-terminal domain, four hydrophobic regions (TM1–4), and an intracellular C-terminal domain. The ligand-binding site is a conserved amino acid pocket formed from a conformational association between the N terminus and the loop linking TM3 and TM4. A flip/flop alternative splice region and R/G RNA editing site are also present within the TM3/TM4 loop. TM2 forms an intracellular re-entrant hairpin loop which contributes to the cation pore channel and is also the site for Q/R RNA editing in the GluR2 subunit.

4

Introduction

Each AMPA subunit includes a channel-forming domain consisting of three transmembrane domains, TM1, TM3, and TM4 and a re-entrant loop TM2; an intracellular C terminus; and two extracellular domains composed of the N terminus and the segment between TM3 and TM4 (S2) (Fig.1.1). The N terminus contains the N-terminal domain of ~400 amino acids and a ~150 amino acid ligand-binding segment (S1), which with S2 forms an extracellular -binding domain.

Suggestions for N-terminal domain function include receptor assembly, allosteric modulation of the , and binding of a second ligand. The ligand-binding core of AMPA-type receptor channel is the glutamate binding site38, 39, and it confers pharmacological specificity to the receptors. The structures of the ligand binding cores of GluR2 and GluR4 have been studied intensively40, 41. For GluR2, the ligand-binding core has been crystallized with various pharmacological agents42. The intracellular C terminus of eukaryotic AMPA-type receptor channels has been shown to be the interaction site for a range of different proteins, many of which are involved in the receptor trafficking43and synaptic plasticity44. Each of the four AMPA receptor subunits exists as two splice variants, named flip and flop45 (see Fig.1.1), generated by alternative splicing of two 115-base pair that encode 38 amino acids located just proximal to a putative fourth transmembrane domain. This modifies the channel’s kinetic and pharmacological properties. The receptors with flip splice variants desensitize four times slower than that with flop46, 47 and the flip module confers different sensitivity to allosteric modulators cyclothiazide48, 49, 4-[2-(phenylsulfonylamino) ethylthio]-2,6 -difluoro- phenoxyacetamide50, zinc51, and lithium52, although affinity to AMPA is unchanged53. The genomic DNA of the GluR2 subunit of AMPA-type receptor contains a (Q) residue at amino acid position 607 (Q/R site). However, the vast majority of neuronal cDNA contains an positive charged (R) at this position and occurs via a process of nuclear RNA editing54. Receptors containing edited GluR2 have a very low Ca2+ permeability55, low single-channel conductance (because of the size and charge of the amino acid side chain in the edited form56, 57).

Nonetheless, GluR2 (R)-containing AMPA-type receptor channels can still participate in intracellular Ca2+ signaling58 and can be trafficked in a Ca2+-dependent way59. The R/G site may be also edited in a region that immediately precedes the flip/flop splice module 60(Fig.1.1). This modification changes the desensitization and resensitization of the resulting AMPA-type receptor channels61. To show all of the information, glutamate receptors were noted in the form of:

5

Introduction Glu (tamate)R(eceptor)1-4 i/o(flip/flop) R(Glutamin)/G(Glycin) Q(Arginin/R(Glutamin)

Fig.1.2 The Ca2+ permeability of AMPA receptors is determined by the presence or absence of edited GluR2 in the receptor complex. AMPA receptors composed of a combination of GluR1-3-4 are permeable to Ca2+ (left). A positively charged arginine at position 607 in GluR2 prevents the permeation of Ca2+ ions, when present in the receptor complex (right).

1.3.2 The electrophysiological characteristics of AMPA receptors

Figure 1.3 shows an example for the current response of GluR2 flip-receptors upon application of glutamate at a saturating concentration (10 mM) recorded with the Patch-Clamp-Technique. Upon fast application of glutamate, a fast current transient was elicited that raised promptly up to the maximum peak current amplitude and then, in prolonged presence of the agonist, decreased to a „steady-state“ current amplitude (Abb. 3) due to desensitization. After cessation of agonist, the current reduced continuously to zero level due to deactivation. The activation time TTP (time-to-peak) is analyzed by measuring the 20-80% rise time. The maximum peak current amplitude (Pmax in pA) means the maximal current intensity, and steady-state-current (Cdes in pA) means the current intensity of equilibration in presence of the agonist between the open and desensitized states of the receptor activated. The decay phase of the response was fit to mono- or bi- exponential functions. The decay time constant for deactivation (τdea) or desensitization (τdes) was calculated from the fitted curve, and τ was the response decay time required to attain e-1 (approximately 36.8 %) of the maximal amplitude, as determined from the fitted curve. The extent of desensitization was expressed as current amplitude of the steady-state /current amplitude of the initial peak- like response (Cdes / Pmax). The area under the current curve (AUC) represents the total current response to agonist.

6

Introduction

Fig.1.3 The electrophysiological characteristics of AMPA receptor currents

1.3.3 GluR2flipGQ

The most widespread expression of glutamate receptor subunits in human brain was shown for the

GluR2flopGR subunit, but its permeability is rather low. So we chose GluR2flipGQ for the patch clamp experiments. The respective homooligomeric receptor is an AMPA-type receptor channel

2+ with unedited GluR2 at the Q/R-editing site, which means it is Ca permeable. The time course of 62 desensitization follows an exponential with a time constant τDes = 9.86 + 0.79 ms . Desensitization is incomplete to a steady state current amplitude around 10 % of the respective maximum peak current amplitude of the current transient elicited by a saturating concentration of glutamate.

1.3.4 Non-desensitizing GluR2L504Y

The rat AMPA receptor subunit GluR2L504YGQ is a mutant of the GluR2flipGQ subunit, which shows no desensitization at all. The point mutation is located within the S1 region (N-terminal), (L504Y in GluR2, L507Y in GluR3, L497Y in GluR138), which forms part of the ligand binding site. The exchange at position 504 from lysine to tyrosine, which has an aromatic ring, leads to a receptor with normal activity but completely without desensitization. This make it a very sensitive tool to test candidate medicaments63-65.

7

Introduction 1.4 AMPA receptor antagonists as candidate compounds for neuroprotection

AMPA receptor antagonists are of potential interest for the treatment of chronic neurodegenerative diseases, especially ALS. Up to now, direct AMPA-type receptor channel antagonists have not been proved effective in clinical test or could not be introduced into clinical use due to severe side effects. Up to present, only riluzole has showed to be significantly effective, but its effect on AMPA-type receptor channels is not clarified. Here, we selected 8 substances which are known to act neuroprotective in in-vitro tests of excitotoxicity, to investigate their interaction with AMPA-type receptor channels.

1.4.1 Minocycline

Minocycline

Molecular Formula: C23H27N3O7 · HCl Molecular Weight: 493.94

Fig.1.4.1 Chemical structure of minocycline hydrochloride

Minocycline is a semi-synthetic tetracycline derivative that has a spectrum of antibacterial activity similar to tetracycline. It inhibits endothelial cell proliferation and angiogenesis, is clinically well tolerated and effectively crosses the blood–brain barrier66, 67. It was shown that CSF from ALS patients activates microglia, and toxins released by those activated cells stimulate ionotropic glutamate receptors. This leads to the death of spinal motor neurons68. Minocycline inhibits the microglial activation and provides protection against this CSF toxicity68, 69. Moreover, others have shown that this semi-synthetic tetracycline analogue influences disease progression and survival of mutant SOD1 mice70-72, which indicats that it could be a candidate drug to treat ALS.

8

Introduction 1.4.2 Riluzole

Riluzole 2-Amino-6-(trifluoromethoxy)-benzothiazole

Molecular Formula: C8H5F3N2OS Molecular Weight: 234.20

Fig.1.4.2 Chemical structure of riluzole

Riluzole is the only drug which has been proved effectively against disease progression in ALS patients until now2, 73, 74, which has got anti-glutamate properties (amongst others inhibition of glutamate release and inhibition of glutamate receptors)75-77. On the other hand, there is an interaction of riluzole with inhibitory ligand gated ion channels like GABAA receptors that could add to the therapeutic effects78, 79.

1.4.3 Memantine

Memantine 3,5-Dimethyl-1-adamantanamine hydrochloride 3,5-Dimethylamantadine hydrochloride

Molecular Formula: C12H21N · HCl Molecular Weight: 215.76

Fig.1.4.3 Chemical structure of memantine.

Memantine is a non-competitive NMDA receptor antagonist80 , which has been widely used for the treatment of dementia with excellent safety and efficacy profiles81. It was reported that memantine protected cultured cortical neurons from the toxic effects of glutamate82 and prolonged survival in an ALS mouse model83. Its protection has also been shown in various models of acute and chronic neurodegeneration84.

9

Introduction 1.4.4 Ketamine

Ketamine

Molecular Formula: C13H16ClNO · HCl Molecular Weight: 274.19 S-(+)-2-(2-Chlorophenyl)-2-(methylamino) cyclohexanone hydrochloride

Fig.1.4.4 Chemical structure of S-(+)-Ketamine hydrochloride.

Ketamine acts as an open- at NMDA receptor channels85. Neuroprotection by ketamine has been described in a variety of different experimental settings, including transient focal and global, as well as permanent ischemia, traumatic brain injury, and in-vitro hypoxia/ischemia86.

1.4.5 Phenobarbital

Phenobarbital 5-Ethyl-5-phenyl-2,4,6-trioxohexahydropyrimidine salt 5-Ethyl-5-phenylbarbituric acid sodium salt Sodium 5-ethyl-5-phenylbarbiturate

Molecular Formula: C12H11N2NaO3 Molecular Weight: 254.22

Fig.1.4.5 Chemical structure of phenobarbital sodium salt

A traditional type medicament which has been widely used for its sedative, anesthetic, and anticonvulsant properties. It was reported that (thiopental) attenuate NMDA- and AMPA-mediated glutamate excitotoxicity87. The neuroprotective effects of barbiturates (thiopental and ) are thought to correlate at least in part to the inhibition of AMPA receptors88. In an experiment to test the effect of phenobarbital on -induced neurotoxicity, pre-treatment with phenobarbital blocked the kainic acid-induced deficits in acquisition learning89.

10

Introduction 1.4.6 Valproic acid

Valproic acid 2-Propylpentanoic acid

Molecular Formula: (CH3CH2CH2)2CHCO2H Molecular Weight: 144.21 Liquid, 0.9 g/ml at 25 °C, slightly soluble in water, soluble in and DMSO Fig.1.4.6 Chemical structure of valproic acid

Valproic acid is one of the most widely prescribed antiepileptic drugs worldwide90. Recently, it was reported to reduce neurodegeneration in a murine model of human immunodeficiency virus-1 (HIV-1) encephalitis91. There are also hints that valproic acid has neuroprotective effects both in vitro and in vivo models of ALS92.

1.4.7 CNQX

CNQX 6-Cyano-7-nitroquinoxaline-2,3-dione

Molecular Formula: C9H4N4O4 Molecular Weight: 232.15 Light yellow solid, soluble in DMSO, insoluble in Water

Fig.1.4.7 Chemical structure of CNQX

CNQX is a well known potent, competitive AMPA/ antagonist93. In vitro experiments could demonstrate a significant reduction of excitotoxic cell death in the presence of CNQX or analogue compounds like NBQX 94, 95. Due to major side effects clinical trials could not be performed to test for clinical neuroprotection in patients suffering from neurodegenerative diseases 96, 97.

11

Introduction 1.4.8 RPR119990

RPR119990 9-Carboxymethyl-4-oxo-5H,10H-imidazo[1,2-α]- indeno[1,2-e] pyrazin-2-phosphonic acid light yellow solid, good solubility in water (400mg/ml)

Fig.1.4.8 Chemical structure of RPR119990

RPR 119990 is a pyrazine derivative, which was effective in different disease models like ischemic stroke, traumatic brain injury and epileptic seizures. This compound was found active in a transgenic mouse model of familial ALS (SOD1-G93A) where it was able to improve grip muscle strength and glutamate uptake from spinal synaptosomal preparations, and prolong survival98. Binding assays revealed a marked selectivity of AMPA/kainate-type receptor channels; recombinant AMPA-type receptor channels showed a competitive inhibition of kainate induced membrane currents64, 98, 99.

1.5 Purpose

AMPA-type receptor channels are promising targets for pharmacological neuroprotection. However, none of the AMPA-type receptor channel blockers so far has passed clinical testing, mainly due to severe side effects97. On the other hand, several compounds that proved to be neuroprotective in in-vitro models of excitotoxicity are thought to interact with AMPA-type glutamate receptor channels. So we need new AMPA-type receptor antagonists. In this paper, we will use GluR2 L504Y and GluR2flipGQ receptors to screen antagonist of the AMPA receptors. This information could help to develop new and better therapeutic strategies that could protect motor neurons from excitotoxicity.

12

Materials and methods Materials and methods

2.1 Solutions

1. Culture medium The HEK cells were cultured before and after transfection in Dulbeccos modified Eagles medium (DMEM), supplemented with 10% fetal calf serum (FCS), 100 U/ml penicillin and 100 µg/ml streptomycin. 2. Electroporation buffer

Solution for electric transfection. Containing: 50mM K2HPO4, 20 mM K-acetate, pH 7.35. 3. Extracellular solution Extracellular solution for HEK293 cells, with which were the cells continuously perfused during each measurement. Containing: 162 mM NaCl, 5.3 mM KCl, 2 mM CaCl2, 0.67 mM NaHPO4,

0.22 mM KH2PO4, 15 mM Hepes, 5.6 mM glucose. The pH value was adjusted with NaOH to 7.4. The solution is filtered each time before use. 4. Intracellular solution Patch pipettes were filled with intracellular solution: 140 mM KCL, 11 mM EGTA, 10 mM Hepes,

10 mM glucose, 2 mM MgCl2. The osmolarity was adjusted to 340 mosM/l with mannitol. The pH value was adjusted with KOH to 7.3. 5. Solutions of different tested substance. Chemicals were purchased from Sigma (USA) except sodium-L-glutamate (monohydrate) which was obtained from Merck (USA), riluzole (from Rhône-Poulenc, France) and RPR119990 which was kindly provided by Sanofi-Aventis, Paris, France. Glutamate, RPR119990, memantine, minocycline, ketamine, and phenobarbital have good solubility in water. Before each experiment the tested substances were freshly solved in extracellular solution at different concentration yielding a clear solution. Riluzole, valproic acid and CNQX have less solubility in water and must be solved in DMSO and then diluted in Extracellular solution. The concentrations of different substances in DMSO were: riluzole 50mM; valproic acid 1M; CNQX 1mM. 5 Brillant green (E142, food dye) With brillant green the tested solutions will be better observed in the measuring pool.

13

Materials and methods 2.2 Cell culture and transfection

Transformed human embryonic kidney (HEK)293 cells were cultured in culture medium at 37℃ in a 5% CO2/95% air incubator. For transfection, cells were resuspended in electroporation buffer 5 to a density of approximately 8×10 cells/ml. MgSO4 was added to a final concentration of 25 mM. 0.4 ml of this cell suspension was added to purified plasmid DNA (10 µg for each GluR subunit), green fluorescent protein (GFP) 5 µl and mixed. Electroporation at room temperature was performed instantly after the addition of cells to DNA in 0.4 cm cuvette at 250V, 750µF and 329 , resulting in a pulse time of 18-21 ms. Transfected cells were plated on 12 mm glass coverslips (part of them were coated with poly-L-lysine, if we want to make outside-out patch) and incubated for 15-24 h prior to the experiments. The successful transfected cells will express glutamate receptors and GFP, which can be visualized using fluorescence microscopy.

2.3 Patch-clamp technique

2.3.1 Principle

The patch-clamp technique is an electrophysiological method that allows the recording of macroscopic whole-cell or microscopic single-channel currents flowing across biological membranes through ion channels. The technique allows one to experimentally control and manipulate the of membrane patches or the whole-cell (voltage clamp). Alternatively, one may monitor the changes in membrane potential in response to currents flowing across ion channels (current clamp), which constitute the physiological response of a cell. Thus, it is widely used to investigate ion channels in excitable membrane preparations.

. The voltage clamp technique was used in this experimental study. Upon the delivery of a voltage step to an excitable membrane, the elicited current is dependent on the resistance of the pipette – cell network. Stimulation of membrane resident ion channels will increase their open probability and thus increase membrane conductance. The resulting current can be monitored at a constant holding (= clamp) potential. Fig.2.1 shows the final current-to-voltage convert circuit. The way to measure small currents is to monitor the voltage drop across a large resistor. The operational amplifier can be used to automate the adjustment of VB , which is continuously adjusted to bring

Vpip to the correct value (Voltage clamp). The operational amplifier varies its output to keep the

14

Materials and methods pipette potential at Vref. This action can be made very rapid and precise, so that for practical purpose Vpip can be assumed to be precisely Vref. This allows us to measure VB-Vref to obtain the voltage drop across the resistor, the time course of Vout for an arbitrary Ipip can be calculated, and then the Ipip can be calculated.

Fig.2.1 Principle of Patch-Clamp technique. Vpip: Pipette Voltage = Membrane Voltage in mV;

Vref: Reference Voltage = Command Voltage in mV; Vout: Output Voltage in mV; Rf = feedback resistance in Ohm.

2.3.2 Equipment

Patch-clamp experiments were performed on vibration isolation table (low-end air-suspension table), to produce mechanical stability and avoid the unintended relative movements of the pipette/cell. As shows in Fig.2.2, in the center of the table is an upright microscope (Zeiss, amplification up to 40x, water-immersion-objective, fluorescence equipped), which can be moved vertically to allow focusing. The stage of the microscope is fixed on a small air-suspension table and then bound with structure that also holds the hydraulic micromanipulator. On the end of micromanipulator are preamplifier and pipette-holder mounted. To shield the sensitive patch-clamp preamplifier from electrical noise is the vibration isolation table surrounded with a Faraday cage.

15

Materials and methods

Fig.2.2 The patch-clamp setup is equipped with the following instruments: (1) Faraday cage, (2) vibration isolation table, (3) microscope, (4) micromanipulator with preamplifier and pipette-holder, (5) stage with bath-perfusion-system and fast-application-system. (6), control of the micromanipulator, (7) first outlet of fast perfusion system, (8) second outlet of perfusion system, (9) hydraulic pump for the drainage and overflow of bath solution, (10) Analog-Digital-converter, (11) video camera, (12) power supply for lamp of microscope (13) oscilloscope, (14) patch-clamp amplifier, (15) computer for signal recording and analysis.

In the center of the stage there is the recording chamber with perfusion- and drainage- systems (Fig.2.3), in which coverslips with cells can be perfused with extracellular solution. On the right side the fast-application-system is mounted (see later). On the left side there is the patch pipette and the pipette holder which are directly connected to the pre-amplifier of the patch clamp amplifier and mounted on a micromanipulator.

16

Materials and methods Fig.2.3 Stage with bath-perfusion-system and fast-application-system: (1) bath chamber, (2) silver chloride coated reference electrode, (3) connected to test-solutions , (4) Piezo-driven instrument, (5) holder for capillary or theta-glass, (6) perfusion system for water bad, (7) drainage system, (8) tube for application of suction or pressure to the inside of pipette, (9) pipette holder, (10) water-immersion-objective (working distance about 1.5mm), (11) patch pipette, (12) connected to the second hydraulic pump for prevention overflow, (13) capillary or theta-glass for fast-application, (14) stage.

2.3.3 Pipette Fabrication and Mechanical Setup

Patch pipettes were made from borosilicate glass, which is in the form of standard microelectrode capillaries (HARVARD apparatus LTD, GC150TF-10, 1.5 mm O.D.X 1.17 mm I.D.). The pulling happened in two stages using a horizontal microelectrode puller (DMZ – Pipettenpuller, Zeitz-Instrumente, Augsburg) and then heat polishing is done at the pipette tip. Pipettes are freshly made before each measurement, filled with intracellular solution, and then mounted on a pipette holder shown schematically in Fig.2.4. The resistance values of pipettes are in the range of 5 – 12 M. The holder serves two basic functions, firstly to provide electrical connection between the patch pipette solution and a BNC connector of the amplifier head stage which is mounted on a Narishige MO-103 hydraulic micromanipulator, and secondly to allow suction or pressure to be applied to the pipette interior.

The holder has a Teflon body with a central bore for tight fitting of a patch pipette and a chlorided silver wire which is soldered to the pin of a BNC connector. Outlet S connects to Silicone rubber tubing for application of suction or pressure to the inner compartment

Fig.2.4 Cross section through pipette holder.

17

Materials and methods 2.3.4 Patch Technique

The basic approach to measure small ionic currents in the picoampere range requires a low-noise recording technique. This is achieved by sealing patch-pipette tightly onto the plasma membrane of an intact cell, thereby isolating a small patch. When the tip of the pipette was pressed against the cell surface the seal resistance was less than 25 M (The resistance was measured by applying a 0.1 mV voltage pulse in the pipette and monitoring the resulting current flow). The resistance increased when a slight negative pressure of 20~30 cm H20 was applied. The development of giga-seals usually occurs within several seconds. The seal resistance should be in excess of 109 . This “cell attached” configuration is the precursor to all other variants of the patch-clamp technique. The contact between and pipette after formation of a giga-seal is not only electrically tight, but also mechanically very stable. The high resistance of this seal ensures that most of the currents originating in a small patch flow into the pipette, and from there into current-measurement circuitry. Giga-seal is also mechanically stable. After giga-seal formation, the cell surface membrane is separated into two parts: the area covered by the pipette (the patch area) and the rest of the cell. The membrane patch can be disrupted by applying brief voltage (of up to 200 mV) or negative pressure pulses (matching the pressure of a 100 cm H20 column) to the pipette interior, but keeping the pipette cell-attached. This provides a direct low resistance access to the cell interior which allows potential recording and voltage clamping of small cells (whole-cell). When the pipette tip was slowly withdrawn from the cell surface often a fine cytoplasmic bridge could then be observed between the cell surface and the pipette. Upon further removal the cytoplasmic bridge tied off but left the giga-seal intact (outside-out). Fig.2.5 illustrates schematically how this can be done100. The whole-cell models or outside-out patches can then be moved into fast-application-system to study the effect of drugs on extracellular surface of the membrane.

18

Materials and methods

Fig.2.5. Schematic representation of the procedures which lead to recording configurations. The two recording configurations, described in this paper are: " whole-cell model" and "outside-out patch" (modified from Hamill et al100).

Fig.2.6 Patch pipette on the cell surface

2.3.5 Fast application technique

The patch pipette, with an outside-out or a whole-cell patch, was moved in a tube so that the cell was continuously perfused by the control solution (1~2 ml/min). The test solution flowed out of a single-barreled application pipette with a stable pressure (0.3~0.5 bar) into the perfused tube, formed an extremely sharp interface between two laminarly flowing solutions, which can be rapidly moved, under computer control, across the patch by a piezoelectric element (Fig.2.3, Fig.2.7) (piezo-driven ultra-fast perfusion system)101. Using this system can solutions be not only extremely fast applied, but also removed, like the pulse, and the application and wash-out of glutamate were 80 µs and 0.5 ms61, 102, 103.

19

Materials and methods

A B

Fig.2.7 Ultra-fast perfusion: A: The patch-pipette is at the edge of laminarly tested solution, B: The application pipette was moved by piezoelectric element, and the patch was deep into tested solution.

The application pipette can be changed into dual-channel glass tubing, namely “Theta Glass” (Fig.2.8). Two solutions can be perfused at the same time with the same speed (at same pressure). The patch was located at one side of the interface, which acted as background perfusion. When the pulse comes, the interface moved toward the patch pipette, and then another side solution (foreground solution) contact with the patch. Both of the two channels are separately connected with a rotary dial selector, consisting of two barrels with the same high pressure, in which can different solutions be changed. The time necessary for complete exchange between two perfusing solutions is about 10~20 s.

Fig.2.8 Theta-Glass

2.3.6 Application protocols

Experiments were performed applying single 200 ms pulses (except in coapplication experiments with RPR119990 on GluR2flipGQ, with 50 ms pulses) of glutamate or glutamate + different substances. The pause between two pulses is 30 s to avoid the run-down effect. After change of

20

Materials and methods solutions around 30 s waiting time in coapplication experiments before a new group measurement begin for complete solution exchange, or around 120 s in preincubation experiments. (Fig.2.9). Pulse protocols were controlled by a PC using the pClamp6 software (Axon instruments, Foster City, USA).

Puls 1

Fig.2.9 Single pulse program (200 ms).

For quantitative evaluation 3-10 current traces for each group of an experiment were averaged. Generally, controls with pulses of saturating agonist concentration (10 mM glutamate) were made before (control 1) and after (control 2) trains of pulses with agonist and blocker to normalize the current responses. A relative value can be calculated to compare the block effect among different tests. For example: The measurement sequences of one patch for coapplication of minocycline and glutamate on GluR2L504Y : 1. 10mM glutamate 3~5 traces 2. 10mM glutamate + 0.1mM minocycline 5~10 traces 3. 10mM glutamate 3~5 traces 4. 10mM glutamate + 1mM minocycline 5~10 traces 5. 10mM glutamate 3~5 traces 6. 10mM glutamate + 3mM minocycline 5~10 traces 7. 10mM glutamate 3~5 traces ……

Amplitude of Pmax2 Relative Amplitude of 0.1mM minocycline = rAmp =

Ampl. Pmax1 + Ampl. Pmax3

Cdes. Relative steady-state-current = rCdes = P max Relative areas under the current curve (rAUC), relative time to peak (rTTP) are calculated in the same way with rAmp.

21

Materials and methods 2.3.7 Data acquisition and analysis

Patch-clamp measurements were performed using excised outside-out patches or whole-cell patches for fast application experiments using standard methods100. The holding potential was kept at -40mV. Data were recorded with an Axopatch 200B patch-clamp amplifier. Ensemble currents were sampled with 20 kHz onto the hard disk of a PC using a Digidata 1200 Interface and the pClAMP6 software suit (Axon Instruments). For further analysis, data were filtered at 5 kHz.

2.4 Statistic method

All data were giver as mean + S.E.M. When statistical analysis was performed, one-way ANOVA and Student’s t-tests were used, with p < 0.05 considered significant.

22

Results Results

3.1 Effect of minocycline on AMPA-type receptor channels

3.1.1 Minocycline: GluR2L504Y receptor channels

1. Coapplication of minocycline and agonist Coapplication experiments were performed using a modified fast application patch clamp technique applying 10 mM glutamate and different concentrations of minocycline ( 0.1, 1, 3 mM;

Fig.3.1.1) to whole-cell patches expressing GluR2L504Y receptor channels. Fig.3.1.1 shows averaged original current curves which were obtained from one small HEK293-cell lifted from the bottom expressing non-desensitizing mutant homomeric glutamate receptor channels (GluR2L504Y). The relative amplitude (Fig.3.1.1 A , B) was 0.97 (0.98 + 0.02, n = 9), 0.95 (0.96 + 0.01, n = 8), and 0.92 (0.91 + 0.03, n = 4), respectively, after coapplication of 10 mM glutamate and 0.1 mM, 1 mM or 3 mM minocycline. The differences did not reach significance (p = 0.16). The rAUC (Fig.3.1.1 A, B) was 0.98 (0.98 + 0.02), 0.86 (0.93 + 0.03), 0.29 (0.36 + 0.06) with significant difference (p < 0.001). The relative time to peak (Fig.3.1.1 A, B) was 0.96 (0.94 + 0.04), 0.93 (0.91 + 0.02), 0.68 (0.64 + 0.06), (p < 0.001).

The relative steady state (Fig.3.1.1 A, C; Cdes /Pmax) was 0.97 (0.95 + 0.01), 0.86 (0.87 + 0.02), 0.20 (0.21 + 0.07) with significant difference (p < 0.001).

The relative amplitude of reopening current (rP2) (Amplitude of reopening current /Pmax) increased from 0.08 (0.09 + 0.01) to 0.56 (0.57 + 0.01) respectively (p < 0.001) when 1 mM or 3 mM minocycline were applied.

A concentration dependent decrease of time constant of current decay (Fig.3.1.1 A, D), τB1,2, as bi-exponential occurred at coapplication of 10 mM glutamate and 1 mM or 3 mM minocycline. τB1 decreased from 8.33 (8.08 + 0.36, n = 4) ms to 5.18 (5.14 + 0.57, n = 4) ms (p = 0.005), while τB2 changed from 64.4 (65.9 + 2.78, n = 4) ms to 67.2 (64.9 + 3.41 n = 4) ms (p = 0.845), at the same time.

The deactivation time constant τdec (Fig.3.1.1 A, D) of 10 mM glutamate without or with 0.1, 1, and 3 mM minocycline was 10.1 (9.85 + 0.37, n = 25), 10.5 (10.32 + 0.84, n = 9), 10.1 (9.63 + 0.48, n = 8), 10.2 (9.56 + 1.57, n = 4) ms, respectively, with no significant difference (p = 0.87).

23

Results

Minocyline blocks the GluR2L504Y receptor channels dose-dependently. At higher concentration level of minocycline a decrease of the rAUC occurred as well as a current decay during application and a decrease of the steady state current amplitude. The reopening current after removing glutamate and blocker might be explained by assuming an open-channel-block.

1

0.5 rAmp rTTP rAUC B

0 0.1 mM minocycline 1 mM minocycline 3 mM minocycline

1

0.5 rCdes rP2

0 C 0 mM 0.1 mM 1 mM 3 mM minocycline minocycline minocycline minocycline

80

60 0 mM minocycline s 0.1 mM minocycline m 1 mM minocycline 40 3 mM minocycline

A 20 D

0 τdea τB1 τB2

Fig.3.1.1 (A) shows averaged original current curves after 200 ms pulses of 10 mM glutamate or

10 mM glutamate + increasing concentration of minocycline (as indicated) applied to GluR2 L504Y channels. (B) shows the relative peak current amplitude (■), time to peak (×), rAUC (▲) and (C) relative current amplitude of steady state (○), amplitude of channel reopening (△), (D) deactivation time (τdec) and time constant of current decay (τB1,2) in the presence of glutamate (as indicated) without or with minocycline. The holding potential was -40 mV in all experiments.

24

Results 2. Preincubation of patches with minocycline. We addressed the question if there is an additional block effect to the described open-channel block. To answer the question if there is a competitive block effect we preincubated the patched cell with increasing concentrations of minocycline before applying the test solution containing 10 mM glutamate and the corresponding concentration of minocycline. In Fig.3.1.2 averaged original current curves are shown before and after preincubation with a high concentration of 1 mM minocycline. The rAUC and rAMP were 1.07 + 0.08 and 1.04 + 0.07 (n = 5) respectively. No significant difference between control current before preincubation and test current after preincubation was observed. A competitive block could be excluded by our results.

Fig.3.1.2 Before and after 30 s. preincubation of small cells transfected with GluR2L504Y channels with 1 mM minocycline.

3.1.2 Minocycline: GluR2flipGQ receptor channels

1. Coapplication with or without background preincubation of patches with minocycline

To answer the question if there is an effect of minocycline in GluR2flipGQ receptor we performed coapplication and preincubation experiments as showed in Fig.3.1.3. First we measured the currents induced by 10 mM glutamate + increasing concentrations of minocycline (0.1, 0,3, 1 or 3 mM ) (Fig.3.1.3 A,C). In a second step we preincubated with concentrations of minocycline as indicated in Fig.3.1.3 B for 30 s. before applying the test solution containing 10 mM glutamate and the corresponding concentration of minocycline.

25

Results

A B

4 4

3.5 3.5 rAmp 3 rAmp 3 rTTP rTTP 2.5 2.5 rAUC rAUC 2 2

1.5 1.5 1 1 C D 0.5 0.5 0 0 0.1 mM 0.3 mM 1 mM 3 mM 0.1 mM 0.3 mM 1 mM 3 mM minocycline minocycline minocycline minocycline minocycline minocycline minocycline minocycline

0.2 8 τdes of coopplication τdes of preincubation 0.15 coapplication 6

preincubation ms 0.1 4

0.05 2 E F 0 0 0.1 mM 0.3 mM 1 mM 3 mM 0 mM 0.1 mM 0.3 mM 1 mM 3 mM minocycline minocycline minocycline minocycline minocycline minocycline minocycline minocycline minocycline

Fig.3.1.3 (A, C) 200 ms pulses of glutamate or glutamate + the respective minocycline concentration as indicated were applied to GluR2flipGQ channels. (B, D) The currents were recorded from patches that were incubated at least 30 s. with the minocycline as indicated before application of the test solution. (C, D) The relative peak current amplitude (■), time to peak (×), rAUC (▲), (E) relative current amplitude of steady state (○), (F) deactivation time (τdec) and time constant of current decay (τB1,2) in the presence of glutamate without or with minocycline in the

26

Results coapplication or preincubation experiments. The holding potential was -40 mV in all experiments.

The currents of Fig.3.1.3 A and B were obtained from one cell which expressed GluR2flipGQ channels. Fig.3.1.3 C, D, E and F were the evaluation of the results from independent experiments. After coapplication of 10 mM glutamate and 0.1, 0.3, 1 and 3 mM minocycline, the relative amplitude (Fig.3.1.3 A, C) was 0.86 (0.88 + 0.06, n = 8), 1.07 (1.06 + 0.08, n = 6), 1.19 (1.14 + 0.06, n = 6) and 1.49 (1.81 + 0.12, n = 5) respectively, with significant difference (p < 0.001). The rAUC was 0.82 (0.82 + 0.02), 1.10 (1.07 + 0.04), 1.26 (1.21 + 0.03)，4.07 (3.69 + 0.54) with significant difference ( p < 0.001). Both of them increased with increasing concentrations of minocycline, especially with 3 mM minocycline, The same change tendency also happened in the percentage of steady state current (rCdes = Cdes /Peak amplitude, Fig.3.1.3 A, E), which was 0.005 (0.006 + 0.002), 0.009 (0.009 + 0.001), 0.012 (0.011 + 0.002) , 0.15 (0.129 + 0.04) respectively with significant difference ( p = 0.01). The relative time to peak (Fig.3.1.3 A, C) was 0.99 (0.98 + 0.10), 1.03 (1.04 + 0.04), 1.07 (1.05 + 0.03)，1.03 (1.00 + 0.06), with no significant difference (p = 0.86). And the time constant of current decay τdec (Fig.3.1.3 A, F) of 10 mM glutamate without or with 0.1, 0.3, 1 and 3 mM minocycline was 6.17 (6.08 + 0.33), 6.14 (6.09 + 0.30), 6.12 (6.05 + 0.12)，6.15 (6.34 + 0.33), 6.37 (6.30+ 0.21) respectively, with no significant changed (p = 0.68). In the preincubation experiments, The rAmp (Fig.3.1.3 B, D) of 10 mM glutamate with 0.1 , 0.3, 1and 3 mM minocycline were 0.73 (0.71 + 0.07, n = 8), 1.05 (1.07 + 0.10, n = 6), 1.15 (1.18 + 0.07, n = 6) and 1.60 (1.68 + 0.23, n = 5) respectively, with significant difference (p < 0.001), the rAUC (Fig.3.1.3 B, D) were 0.73 (0.70 + 0.03), 1.07 (1.08 + 0.05), 1.25 (1.27 + 0.02), 2.02 (2.32 + 0.32), with significant difference ( p < 0.001), the rCdes (Fig.3.1.3 B, E) were 0.010 (0.010 + 0.002), 0.009 (0.011 + 0.002), 0.014 (0.012 + 0.001), 0.021 (0.033 + 0.013) but with no significant difference ( p = 0.16). The relative time to peak (Fig.3.1.3 B, C) and τdes (Fig.3.1.3 B, F) were also no significant changed (p = 0.12, P = 0.99). The comparison between coapplication and preincubation experiments at the same concentration of minocycline (3 mM) revealed that the current enhancing effect of preincubation is smaller than that of coapplication (rAUC (p = 0.03), rCdes (p = 0.04) ).

2. Potentiating effect of minocycline on GluR2flipGQ To better clarify the increase of current by minocycline, we performed the following experiment, in which different concentrations of minocycline were coapplicated with 3 µM glutamate. Fig.3.1.4 showes the currents from a whole-cell patch, which has been transfected with

GluR2flipGQ receptor channels.

27

Results The rAmp of 3 µM glutamate (Fig.3.1.4 A, B) without and with 0.3, 1, and 3 mM minocycline were 0.070 (0.073 + 0.06, n = 4), 0.068 (0.068 + 0.008), 0.088 (0.082 + 0.013) and 0.339 (0.312 + 0.100) respectively, obviously increased by 3 mM minocycline with significant difference (p = 0.03), the rAUC (Fig.3.1.4 A, B) were 0.174 (0.157 + 0.030), 0.185 (0.17 + 0.036), 0.20 (0.204 + 0.062), 1.30 (1.19 + 0.24) respectively, also at 3 mM with significant difference ( p < 0.001). It were different with what showed in Fig.3.1.3, the τdes (Fig.3.1.4 A, C) also increased with the increase of minocycline (6.64 + 0.96, 6.65 + 0.93, 8.13 + 1.85, 22.56 + 3.14, p = 0.01), but the τdea (Fig.3.1.4 A, C) were no significant different (p = 0.39). This proved that minocycline has a potentiating effect; it attenuates the extent of receptor desensitization and lightly slowes down desensitization of GluR2flipGQ receptor channels in a high concentration range.

1 rAmp rAUC

0.5

B

0 0 mM 0.3 mM 1 mM 3 mM minocycline minocycline minocycline minocycline

30 0 mM minocycline 0.3 mM minocycline 25 1 mM minocycline 3 mM minocycline 20

s m 15

10

5 C 0 A τdes τdec

Fig.3.1.4 (A) 200 ms pulses of glutamate or glutamate + the respective minocycline concentration as indicated were applied to GluR2flipGQ channels. (B) The relative peak current amplitude (■) and rAUC (▲) (C) the τdes and τdea in the presence of 3 µM glutamate without or with minocycline in the coapplication experiments. The holding potential was -40 mV in all experiments.

28

Results

3.2. Effect of riluzole on AMPA-type receptor channels

3.1.1 Riluzole: GluR2L504Y receptor channels

1. Coapplication of riluzole and agonist

1

rAmp 0.5 rTTP rAUC rCdes B 0 0.1 mM riluzole 1 mM riluzole 3 mM riluzole

150

0 mM riluzole 0.1 mM riluzole 100 1 mM riluzole s m 3 mM riluzole

50 C A 0 τdec τB1 τB2

Fig.3.2.1 (A) 200 ms pulses of glutamate or glutamate + the respective riluzole concentration (as indicated) were applied to GluR2L504Y channels. (B) shows the relative peak current amplitude (■), time to peak (×), rAUC (▲), relative current amplitude of steady state (○), (C) deactivation time (τdec) and time constant of current decay (τB1,2) in the presence of glutamate (as indicated) with or without riluzole. The holding potential was -40 mV in all experiments. Data points for results are presented as means ± S.E.M. Coapplication experiments were performed using a modified fast application patch-clamp technique applying 10 mM glutamate solution without or with riluzole 0.1 mM, 1 mM and 3 mM, respectively. Fig.3.2.1 A showed averaged original current curves which were obtained from a whole-cell model transfected with GluR2L504Y. The relative Amplitude after coapplication of 10 mM glutamate and 0.1 mM, 1 mM and 3 mM riluzole were 0.98, 1.01 and 1.00 respectively, the data from different experiments were showed in Fig.3.2.1 B and Table 3.2.1, with no significant difference. The rAUC was 0.99, 0.87, 0.79 respectively, and the percentage of desensitization was 0.93, 0.75, 0.52 respectively, both (Fig.3.2.1B and Table 3.2.1) were obviously reduced. The effect was

29

Results reversible upon transition back to the stream of control solution.

The rTTP (Fig.3.2.1 B and Table 3.2.1) was 1.00, 0.95, and 1.00 respectively. The τdea (Fig.3.2.1 C and Table 3.2.1) was 10.67, 10.38, 10.71 ms respectively, both with no significant difference.

The concentration dependent decrease of time constant of current decay (Fig.3.2.2 C), τB1,2, as bi-exponential occurred at coapplication of 10 mM glutamate and 1 mM or 3 mM riluzole. τB1, decreased from 11.46 (11.06 + 0.87, n = 7) ms to 7.65 (7.04 + 0.30, n = 4) ms (p = 0.02), τB2 decreased from 101 (103.32 + 4.84) ms to 69 (66.43 + 7.76) ms (p = 0.003). When 3 mM riluzole was coapplied, 8% (10 + 2%) reopening current was also observed.

Riluzole blocks the GluR2L504Y receptor channels dose-dependently. In higher concentration level of riluzole a decrease of the rAUC, a decrease of time constant of current decay during application and decrease of steady state current were obviously changed parameters. These and the reopening current phenomenon might explain an open-channel block effect.

Table3.2.1 Coapplication of riluzole and glutamate on GluR2L504Y receptor.

+ riluzole rAmp rTTP rAUC rCdes τdea n 0.1mM 0.98 + 0.01 0.93 + 0.04 0.98 + 0.01 0.94 + 0.01 10.67 + 1.54 9 1mM 1.02 + 0.02 0.96 + 0.03 0.84 + 0.04 0.77 + 0.02 10.39 + 1.44 9 3mM 1.01 + 0.02 1.02 + 0.04 0.76 + 0.04 0.51 + 0.02 10.71 + 0.81 4 P 0.07 0.44 <0.001 <0.001 0.99

2. Preincubation of patches with riluzole

1

rAmp 0.5 rAUC

0 0.001 mM 0.01 mM 0.1 mM 1 mM riluzole riluzole riluzole riluzole B

A

30

Results Fig.3.2.2 (A) shows averaged original current traces before and after 30 s. preincubation of outside-out patches transfected with GluR2L504Y channels with 0.001, 0.01, 0.1 and 1 mM riluzole (as indicated). (B) shows the relative peak current amplitude (■), rAUC (▲) in the presence of glutamate before and after 30 s preincubation with riluzole. The holding potential was -40 mV in all experiments.

To answer the question if there is an additional block effect to the described open-channel block, for example, competitive block effect, we preincubated the patched cell with increasing concentrations of riluzole before applying 10 mM glutamate and the corresponding concentration of riluzole. In Fig.3.2.2 an averaged original current traces were shown, which were obtained before and after preincubation outside-out patch containing the GluR2L504Y subunit with 0.001, 0.01, 0.1 and 1 mM riluzole respectively. The relative Amplitude was 0.98 (0.99 + 0.01), 0.98 (1.01 + 0.02), 0.93 (0.92 + 0.02), 0.19 (0.19 + 0.01) respectively, with significant different (p < 0.001). The rAUC was 0.98 (0.99 + 0.01), 0.97 (1.00 + 0.02), 0.91 (0.92 + 0.02), and 0.10 (0.14 + 0.03) respectively, with significant difference (p

< 0.001). The different between 0.01 and 0.1mM riluzole is also significant (PrAmp = 0.049, PrAUC = 0.046). Compared to the coapplication test, the block effect upon preincubation is stronger at the same concentration. For example, the rAUC at coapplication of 1mM riluzole was 0.84 + 0.04, at preincubation it was 0.14 + 0.03. The block from preincubation with riluzole should indicate an additional competitive block effect.

3. The effect of DMSO on glutamate receptors Because riluzole must be solved in DMSO and then diluted in buffer, an effect from DMSO must be excluded. In Fig.3.2.3 averaged original current traces were shown, which were elicited from a whole-cell patch with 200 ms pulses of 10 mM glutamate, 10 mM glutamate + 10% DMSO, or after 30 s preincubation with 6% DMSO. The rAUC and rAMP were 1.01 + 0.02, 0.99 + 0.03 from coapplication experiment, and 0.99 + 0.03, 0.98 + 0.04 from preincubation experiment, no significant changes were observed.

31

Results

Fig.3.2.3 shows averaged original current curves with 200 ms pulses of 10 mM glutamate, 10 mM glutamate + 10% DMSO, or after 30 s. preincubation with 6% DMSO.

3.1.2 Riluzole: GluR2flipGQ receptor channels

1. coapplication with or without background preincubation of patches with riluzole

A B

32

Results

1 1.8 rTTP pre.

rTTP co. 0.5 rAmp pre. 1.3 rAmp co. rAUC pre. C D rAUC co.

0 0.8 0.1mM 0.3 mM 1 mM 3 mM 0.1mM 0.3 mM 1 mM 3 mM riluzole riluzole riluzole riluzole riluzole riluzole riluzole riluzole

8

7

6

5 s τB co m 4 τB pre 3

2

1

0 0mM 0.1mM 0.3 mM 1 mM 3 mM E riluzole riluzole riluzole riluzole

Fig.3.2.4 (A) 200 ms pulses of glutamate or glutamate + the respective riluzole concentration as indicated were applied to GluR2flipGQ channels. (B) The currents were elicited from patches that were incubated 30 s with riluzole as indicated before application of test solutions. (C) The relative peak current amplitude (■), rAUC (▲), time to peak (×) in the preincubation experiments and the relative peak current amplitude ( ), rAUC (△), (D) time to peak (+), and (E) time constant of current decay (τB) in the presence of glutamate without or with riluzole in the coapplication or preincubation experiments. The holding potential was -40 mV in all experiments.

To answer the question if there is an effect of riluzole in GluR2flipGQ receptor we performed coapplication and preincubation experiments as showed in Fig.3.2.4. First we measure the currents induced by 10 mM glutamate + increasing concentrations of riluzole (0.1, 0.3, 1 or 3 mM ). In a second step we preincubated with riluzole as indicated in Fig.3.2.4 B. for 30 s. before applying the test solutions containing 10 mM glutamate and the corresponding concentration of riluzole.

The currents of Fig.3.2.4 A and B were obtained from one cell which expressed GluR2flipGQ channels. Fig3.2.4. C, D, E and Table 3.2.2 were the evaluation of the results from independent experiments. After coapplication of 10 mM glutamate and 0.1, 0.3, 1, or 3 mM riluzole, the relative amplitude (Fig.3.2.4 A, C, Table 3.2.2) was 0.99, 1.01, 0.98 and 1.02 respectively. The rAUC was 1.01, 1.00,

33

Results 1.03 and 1.10 respectively, both with no significant difference. The same change tendency also happened in the rTTP (Fig.3.1.3 A, D), which was 0.99, 0.95, 0.97 and 0.96 respectively. The time constant of current decay τB (Fig.3.1.3 A, E) of 10 mM glutamate without or with 0.1, 0.3, 1, or 3mM riluzole was 6.86, 6.77, 6.95, 7.10 and 7.20 ms respectively, with no significant changed (p = 0.92). In the preincubation experiments, the rAmp (Fig.3.2.4 B,C) of 10 mM glutamate with 0.1 , 0.3, 1, or 3 mM riluzole were 0.64, 0.66, 0.48 and 0.22 respectively, the rAUC (Fig.3.2.4 B,C) were 0.64, 0.66, 0.40 and 0.38, both with significant difference. The rTTP (Fig.3.2.4 B, D) prolonged from

1.08, 1.10, and 1.33 to 1.50. τB (Fig.3.2.4 B, E) were 6.48, 7.10, 7.13 and 7.8 respectively (p = 0.78). It was apparent that the block effects of riluzole on GluR2flipGQ were mainly on preincubation experiments. This could be explained by mechanism of competitive block. Table 3.2.2 coapplication with or without background preincubation of patches with riluzole Riluzole rAmp co. rAmp pre. rTTP co. rTTP pre. rAUC co. rAUC pre. 0.1 mM 0.97 + 0.02 0.68 + 0.03 0.99 + 0.02 1.08 + 0.05 0.97 + 0.05 0.72 + 0.04 0.3 mM 0.98 + 0.03 0.65 + 0.03 0.98 + 0.05 1.16 + 0.05 1.05 + 0.05 0.65 + 0.04 1 mM 1.02 + 0.03 0.44 + 0.07 1.04 + 0.05 1.37 + 0.01 0.98 + 0.05 0.38 + 0.06 3 mM 1.04 + 0.07 0.20 + 0.02 0.98 + 0.02 1.75 + 0.25 1.16 + 0.10 0.34 + 0.05 p 0.68 <0.001 0.66 0.001 0.18 <0.001 * Co. = Coapplication experiments; Pre. = Preincubation experiments

34

Results

3.3. Effect of memantine on AMPA-type receptor channels

3.3.1 Memantine: GluR2L504Y receptor channels

1. coapplication with or without background preincubation of patches with memantine

A B

1 1

rAMP rAMP 0.5 rTTP 0.5 rTTP rAUC rAUC rCdes rCdes D C 0 0 0.01 mM memantine 0.1 mM memantine 1 mM memantine 0.01 mM memantine 0.1 mM memantine 1 mM memantine

Fig.3.3.1 (A, C) 200 ms pulses of glutamate or glutamate + the respective memantine concentration as indicated were applied to GluR2L504Y receptor channels. (B, D) The currents were from patches that were incubated 30 s with memantine before application of the test solutions. (C, D) The relative peak current amplitude (■), time to peak (×), rAUC (▲) and relative current amplitude of steady state (○) in the presence of glutamate without or with memantine in the coapplication or preincubation experiments. The holding potential was -40 mV in all experiments. Coapplication experiments were performed using modified fast application patch-clamp technique

35

Results applying 10 mM glutamate and different concentrations of memantine (0.01, 0.1, 1 mM; Fig.3.3.1 A, C). And followed tests were performed after background preincubation 30 s. of patches with relative concentration memantine as showed in Fig.3.3.1 B, D. The currents of Fig.3.3.1 A and B were obtained from an outside-out patch which expressed

GluR2L504Y receptor channels. Fig.3.3.1 C, D and Table 3.3.1 were the evaluation of the results from some independent experiments. After coapplication of 10 mM glutamate and 0.01, 0.1, or 1 mM memantine, the relative amplitude (Fig.3.3.1 A, C) was 0.96, 0.95 and 0.68 respectively. The rAUC was 1.02, 0.93 and 0.49 respectively. Both of them decreased with increasing concentrations of memantine, especially with 1 mM memantine (as showed in Table 3.3.1). The same change tendency also happened in the percentage of steady state current (rCdes, Fig.3.1.3 A, C), which was 0.96 (0.95 + 0.01), 0.88 (0.86 + 0.03) and 0.45 (0.47 + 0.02) respectively with significant difference (p < 0.001). The relative time to peak was 0.96, 0.84, and 0.70, but with no significant difference (p = 0.12). The time constant of current decay (Fig.3.3.1 C), τB, as mono-exponential occurred at coapplication of

10 mM glutamate and 0.1 mM or 1 mM memantine. τB decreased form 4.51 (4.58 + 0.24) ms to 2.89 (3.15 + 0.44) ms (p = 0.047). When 1 mM memantine was coapplied, a reopening current with relative peak current amplitude of 60 % (53 + 10 %) relative to the peak current amplitude during application was observed. In the preincubation experiments, the rAmp (Fig.3.3.1 B, D) of 10 mM glutamate with 0.01, 0.1, or 1 mM memantine were 0.94, 0.91 and 0.62 respectively, the rAUC was 0.95, 0.90 and 0.44 respectively. Both of them decreased with increasing concentrations of memantine, as show in Table 3.3.1. The rCdes were 0.94 (0.92 + 0.02), 0.89 (0.90 + 0.01) and 0.47 (0.48 + 0.01), with significant difference (p < 0.001). The relative time to peak was 0.95, 0.98 and 0.69 respectively.

In preincubation experiments with 0.1 mM or 1 mM memantine, τB decreased from 4.82 (4.66 + 0.18) ms to 2.16 (2.65 + 0.50) ms (p = 0.041). When 1 mM memantine was coapplied, 58 % (57 + 14 %) reopening current was also observed.

Memantine blocks the GluR2L504Y receptor channels dose-dependently both in coapplication and in preincubation experiments. There were no significant difference between the two types of experiments when 1 mM memantine was used as coapplication or preincubation (PrAMP = 0.08,

PrAUC = 0.31). At a higher concentration level of memantine, a decrease of the rAMP, rAUC, time constant of current decay during application and decrease of steady state current were obviously changed parameters. These and the reopening current phenomenon could explain an open-channel block effect.

36

Results Table 3.3.1 coapplication with or without background preincubation with memantine memantine rAmp co. rAmp pre. rTTP co. rTTP pre. rAUC co. rAUC pre. n 0.01 mM 0.97 + 0.03 0.97 + 0.04 0.97 + 0.08 0.97 + 0.07 0.99 + 0.05 0.94 + 0.08 4 0.1 mM 0.95 + 0.01 0.90 + 0.07 0.84+ 0.05 0.99 + 0.01 0.94 + 0.03 0.91 + 0.06 4 1 mM 0.70 + 0.02 0.63+ 0.02 0.75 + 0.04 0.66 + 0.03 0.52 + 0.03 0.47 + 0.03 4 p <0.001 <0.001 0.12 0.01 0.001 0.001

3.3.2 Memantine: GluR2flipGQ receptor channels

1. coapplication of patches with memantine

A

1

rAmp 0.5 rTTP rAUC B 0 0.001 mM 0.01 mM 0.1 mM 1 mM memantine memantine memantine memantine

Fig.3.3.2 (A) 200 ms pulses of glutamate or glutamate + the respective riluzole concentration (as indicated) were applied to GluR2flipGQ receptor channels. (B) shows the relative peak current amplitude (■), time to peak (×), rAUC (▲) in the presence of glutamate (as indicated) without or with memantine. The holding potential was -40 mV in all experiments. Data points for results are presented as means ± S.E.M.

To answer the question if there is an effect of memantine on GluR2flipGQ receptor we performed coapplication experiments as showed in Fig.3.3.2.

The currents of Fig.3.3.2 A were obtained from one same cell which expressed GluR2flipGQ channels. Fig.3.3.2 B and Table 3.3.2 were the evaluation of the results from some independent experiments. After coapplication of 10 mM glutamate and 0.1 and 1 mM memantine, the relative amplitude

37

Results (Fig.3.3.2 A, B) was 0.96 and 0.79 respectively. The rAUC was 0.96 and 0.69 respectively, both with significant difference (as showed in Table 3.3.2). The rTTP was 1.01 and 0.97 respectively.

The time constant of current decay τB of 10 mM glutamate without or with 0.1 and 1 mM memantine was 7.17 (7.06 + 0.13) , 7.12 (7.06 + 0.18) and 4.91 (5.07 + 0.19) ms respectively, with significant changed (p < 0.001).

Table 3.3.2 coapplication with memantine on GluR2flipGQ receptors memantine rAmp rTTP rAUC n 0.001 mM 1.05 + 0.01 1.07 + 0.06 1.07 + 0.16 5 0.01 mM 1.04 + 0.02 1.06 + 0.03 1.03 + 0.01 4 0.1 mM 0.98 + 0.02 1.03 + 0.02 0.99 + 0.06 5 1 mM 0.81 + 0.02 0.97 + 0.02 0.72 + 0.03 5 p <0.001 0.05 0.002

2. Preincubation of patches with memantine

To exclude the competitive block effect of memantine in GluR2flipGQ receptor channels, we preincubated the patched cell with increasing concentrations of memantine before applying the test solution containing 10 mM glutamate and the corresponding concentration of memantine. In Fig.3.3.3 an averaged original current curve is shown before and after preincubation with 0.1 and 1 mM memantine. The rAmp was 0.96 (0.96 + 0.02) and 0.89 (0.88 + 0.02) respectively (p = 0.03), the rAUC was 0.99 (0.99 + 0.03) and 0.84 (0.83 + 0.02) respectively (p = 0.004), both with significant difference. The relative time to peak was 1.01 (1.00 + 0.03) and 1.03 (1.04 + 0.03) respectively (p = 0.49). The block effect of memantine in preincubation experiments was not stronger than that in coapplication experiments. Memantine acted as a non-competitive blocker in

GluR2flipGQ receptor.

Fig.3.3.3 Averaged original current curves recorded from a whole-cell patch with GluR2flipGQ receptor channels, which was preincubated 30 s with 0.1 or 1 mM memantine as indicated.

38

Results

3.4. Effect of ketamine on AMPA-type receptor channels

3.4.1 Ketamine: GluR2L504Y receptor channels

Fig.3.4.1 shows averaged original current curves after 200 ms pulses of 10 mM glutamate or 10 mM glutamate + 1 mM ketamine (as indicated) applied to a small cell transfected with

GluR2L504Y receptor channels, and the current curves after 30 s. preincubation with 1 mM ketamine before application of the test solution. The holding potential was -40 mV in all experiments.

To test the effect of ketamine in GluR2L504Y receptor we performed coapplication with or without background preincubation of patches with ketamine as showed in Fig.3.4.1. After coapplication of 10 mM glutamate and 1 mM ketamine, the rAMP, rTTP and rAUC were 1.01 (1.01 + 0.02), 1.03 (1.03 + 0.06) and 0.96 (0.98 + 0.02) (n = 4) respectively. After 30 s preincubation with 1 mM ketamine before the pulse, the rAMP, rTTP and rAUC were 0.99 (1.00 + 0.01), 1.02 (1.05 + 0.04) and 0.99 (0.99 + 0.01) respectively. No significant change was happened when 1 mM ketamine was preincubated or coapplied.

3.1.2 Ketamine: GluR2flipGQ receptor channels

Fig.3.4.2 shows averaged original current curves after 200 ms pulses of 10 mM glutamate or 10 mM glutamate + 1 mM ketamine (as indicated) applied to a small cell transfected with

GluR2flipGQ receptor channels. The holding potential was -40 mV in all experiments.

39

Results

Further to exclude the effect of ketamine in GluR2flipGQ receptor we used 1 mM ketamine coapplication with 10 mM glutamate as show in Fig.3.4.2. The rAMP, rTTP and rAUC were 1.04 (1.03 + 0.03), 0.96 (0.98 + 0.03) and 1.04 (1.04 + 0.05) (n = 6) respectively. No significant change was happened when 1 mM ketamine was coapplied with glutamate on GluR2flipGQ receptor.

40

Results

3.5. Effect of Phenobarbital on GluR2L504Y receptor channels

1. Coapplication of phenobarbital and agonist

1

0.5 rAmp rTTP rAUC B

0 1 mM phenobarbital 10 mM phenobarbital 30 mM phenobarbital

1

rP2 0.5 rCdes C 0 1 mM phenobarbital 10 mM phenobarbital 30 mM phenobarbital

120

100 0 mM phenobarbital 80 1 mM phenobarbital s 10 mM phenobarbital m60 30 mM phenobarbital

40

20 D

A 0 τdec τB1 τB2

Fig.3.5.1 (A) shows averaged original current curves after 200 ms pulses of 10 mM glutamate or 10 mM glutamate + increasing concentration of phenobarbital (as indicated) applied to

GluR2L504Y channels. (B) shows the relative peak current amplitude (■), time to peak (×), rAUC (▲) and (C) relative current amplitude of steady state (○), amplitude of channel reopening (△ ),

(D) deactivation time (τdec) and time constant of current decay (τB1,2) in the presence of glutamatewithout or with phenobarbital (as indicated). The holding potential was -40 mV in all experiments. Coapplication experiments were performed using a modified fast application patch-clamp technique applying 10 mM glutamate and different concentrations of phenobarbital (1, 10, 30 mM; Fig.3.5.1 A). The data from different experiments were shown in Fig.3.5.1 B, C, D and Table 3.5.1 Fig.3.5.1 A shows averaged original current curves which were obtained from whole-cell patch expressing homomeric AMPA-type channels containing the GluR2L504Y subunit. The relative Amplitude (Fig.3.5.1 A , B) was 0.99, 0.91 and 0.77 respectively after coaplication of 10 mM glutamate and 1 mM, 10 mM or 30 mM phenobarbital. The rAUC (Fig.3.5.1 A, B) was

41

Results 0.99, 0.77 and 0.50 respectively. As shown in table 3.5.1, both of them decreased significantly with the increasing concentration of phenobarbital. The relative steady state (Fig.3.5.1 A, C) decreased also at the same time, was 0.91, 0.76, and 0.56 respectively.

A concentration dependent decrease of time constant of current decay (Fig.3.5.1 A, D), τB1,2, as bi-exponential occurred at coapplication of 10 mM glutamate and 10 mM or 30 mM phenobarbital.

τB1 increased from 5.42 (5.68 + 0.42, n = 6) ms to 6.70 (6.90 + 0.50, n = 5) ms (p = 0.09), τB2 increased from 75.43 (81.97 + 11.53) ms to 97.12 (100.14 + 3.52) ms (p = 0.13), but both with no significant difference. At the same time the relative amplitude of reopening current (rP2) increased from 15% to 27% respectively, when 10 mM or 30 mM phenobarbital were applied. The relative time to peak (Fig.3.5.1 A, B) was 1.01, 0.99 or 0.98 respectively. The deactivation time τdec (Fig.3.5.1 A, D) of 10 mM glutamate without or with 1, 10, 30 mM phenobarbital were 8.38 (8.64 + 0.37, n = 22), 8.56 (8.77 + 0.28, n = 8), 8.79 (8.48 + 0.41, n = 6), 8.21 (8.44 + 0.59, n = 5) ms respectively, with no significant difference (p = 0.98).

Phenobarbital blocks the GluR2L504Y receptor channels dose-dependently. At a higher concentration level of phenobarbital a decrease of the rAMP, rAUC and time constant of current decay during application were obviously changed parameters. The reopening current after removing glutamate and blocker might be explained with as an open-channel block.

Table3.5.1 Coapplication of phenobarbital and glutamate on GluR2L504Y receptor. phenobarbital rAmp rTTP rAUC rCdes rP2 n 1 mM 0.98 + 0.02 1.03 + 0.06 0.97 + 0.03 0.93 + 0.02 0.05 + 0.01 8 10 mM 0.91 + 0.01 0.98 + 0.03 0.79 + 0.02 0.79 + 0.03 0.14 + 0.01 6 30 mM 0.76 + 0.05 0.98 + 0.01 0.55 + 0.05 0.59 + 0.04 0.26 + 0.04 5 p <0.001 0.76 <0.001 <0.001 0.002

2. Preincubation of patches with phenobarbital. To answer the question, if there is an additional block effect to the described open-channel block, we preincubated the patches with increasing concentrations of phenobarbital before applying 10 mM glutamate and the corresponding concentration of phenobarbital. In Fig.3.5.2 an averaged original current curves were shown, which were obtained before and after preincubation outside-out patch containing the GluR2L504Y subunit with increasing concentration phenobarbital. The data from different experiments were shown in Fig.3.5.2 B, C and Table 3.5.2. The relative Amplitude of preincubation tests with 0.1 mM, 1 mM, 3 mM and 10 mM phenobarbital were 0.98, 0.96, 0.88 and 0.79 respectively. The rAUC were 0.97, 0.98, 0.88 and 0.73 respectively. The rTTP were 1.01, 1.18, 2.4 and 4.03 respectively. The current amplitude and

42

Results area decreased, but the time to peak increased with the increasing concentration phenobarbital. Compare to coapplication test, the block effect is stronger at the same concentration. For example, the rAmp at coapplication of 10 mM phenobarbital was 0.91 + 0.01, at preincubation was 0.77+ 0.05 (p = 0.02). The rAUC was 0.79 + 0.02 and 0.71 + 0.04 respectively (p = 0.13). This further block and the prolonged time to peak from preincubation with Phenobarbital should indicate an additional competitive block effect.

1

0.5 rAmp rAUC B 0 0.1 mM 1 mM 3 mM 10 mM phenobarbital phenobarbital phenobarbital phenobarbital

5

4 rTTP

3 s m 2

1 A C 0 0.1mM 1mM 3mM 10mM

Fig.3.5.2 (A) shows averaged original current curves before and after 30 s. preincubation of outside-out patches transfected with GluR2L504Y channels with 1 and 10 mM phenobarbital (as indicated). (B) shows the relative peak current amplitude (■), rAUC (▲), (C) rTTP in the presence of glutamate before and after 30 s. preincubation with phenobarbital. The holding potential was -40 mV in all experiments.

Table3.5.2 Preincubation of patches with phenobarbital on GluR2L504Y receptor. phenobarbital rAmp rTTP rAUC n 0.1 mM 0.98 + 0.01 1.01 + 0.03 0.97 + 0.01 4 1 mM 0.94 + 0.03 1.10 + 0.04 0.96 + 0.02 4 3 mM 0.91 + 0.04 2.15 + 0.30 0.88 + 0.04 4 10 mM 0.77+ 0.05 3.86 + 0.57 0.71 + 0.04 4 p 0.006 <0.001 0.001

43

Results

3.6. Effect of valproic acid on GluR2L504Y receptor channels

1. Coapplication of valproic acid and agonist

1

rAmp 0.5 rTTP A rAUC B

0 1 mM 3 mM 10 mM 30 mM valproic acid valproic acid valproic acid valproic acid

Fig.3.6.1 (A) shows averaged original current curves after 200 ms pulses of 10 mM glutamate or 10 mM glutamate + increasing concentration of valproic acid (as indicated) applied to

GluR2L504Y channels. (B) shows the relative peak current amplitude (■), time to peak (×), rAUC (▲) in the presence of glutamate with valproic acid. The holding potential was -40 mV in all experiments.

Coapplication experiments were performed using fast application patch-clamp technique applying 10 mM glutamate and different concentrations of valproic acid (Fig.3.6.1 A). The data from different experiments were shown in Fig.3.6.1 B and Table 3.6.1. Fig.3.6.1 A shows averaged original current curves from outside-out patch expressing homomeric glutamate channels from the AMPA-type containing the GluR2L504Y -subunit. The relative peak current amplitude (Fig.3.6.1 A, B) was 1.01, 1.05, 0.99 and 1.06 respectively after coaplication of 10 mM glutamate and 1 mM , 3 mM, 10 mM or 30 mM valproic acid. The rAUC was 1.01, 1.06, 0.98 and 1.01 respectively. The relative time to peak was 0.97, 0.98, 0.97 and 0.98, respectively. As shown in table 3.6.1, all of them have not significantly changed with increasing concentrations of valproic acid.

44

Results

Table3.6.1 Coapplication of valproic acid and glutamate on GluR2L504Y receptor. rAmp rTTP rAUC n 1mM Val 1.01 + 0.01 0.98 + 0.02 1.01 + 0.02 5 3mM Val 1.04 + 0.01 1.02 + 0.05 1.07 + 0.03 5 10mM Val 1.02 + 0.02 1.03 + 0.04 1.04 + 0.04 4 30mM Val 1.06 + 0.03 1.00 + 0.06 1.04 + 0.04 4 p 0.41 0.26 0.64

2. Preincubation of patches with valproic acid. To exclude a competitive block effect we preincubated the patched cell with valproic acid before applying the test solution containing 10 mM glutamate and the corresponding concentration of valproic acid. In Fig.3.6.2 an averaged original current curve is shown before and after preincubation with a high concentration of 1 mM valproic acid. The rAUC and rAMP were 1.01 + 0.02 and 1.03 + 0.07 (n = 4) respectively. No significant effect between control current before preincubation and test current after preincubation was observed. A competitive block could be excluded by our results.

Fig.3.6.2 shows averaged original current curves before and after 30 s preincubation of outside-out patches transfected with GluR2L504Y channels with 1 mM valproic acid (as indicated).

45

Results

3.7 Effect of CNQX on GluR2L504Y receptor channels

Coapplication experiments with 10 mM glutamate and increasing concentrations of CNQX were performed on GluR2L504Y receptor as a control experiment (Fig.3.7.1 A, B, C and Table 3.7.1) Fig.3.7.1 A shows averaged original current curves from whole-cell patch expressing homomeric glutamate channels from the AMPA-type containing the GluR2L504Y -subunit. The relative Amplitude (Fig.3.7.1 A , B) was 0.99, 0.98 and 0.82 respectively after coaplication of 10 mM glutamate and 0.001 mM , 0.01 mM and 0.1 mM CNQX. The rAUC was 0.98, 0.98 and 0.71 respectively. The relative time to peak was 1.01, 1.03 and 0.90 respectively. The rCdes was 0.96, 0.96 and 0.74 respectively. As shown in table 3.6.1, all of them have reduced with the increasing concentration of CNQX. The deactivation time (Fig.3.7.A, C) remain unchanged (p = 0.97). And time constant of current decay could be measured as high as 33.68 + 2.50 ms when 0.1 mM CNQX was coapplied. A reopening phenomenon was not observed.

1

rAmp 0.5 rTTP rAUC rCdes B 0 0.001 mM CNQX 0.01 mM CNQX 0.1 mM CNQX

40

0 mM CNQX s m 0.001 mM CNQX 0.01 mM CNQX 0.1 mM CNQX 20

C A 0 τdec τB

Fig.3.7.1 (A) shows averaged original current curves after 200 ms pulses of 10 mM glutamate or

10 mM glutamate + increasing concentration of CNQX (as indicated) applied to GluR2L504Y channels. (B) shows the relative peak current amplitude (■), time to peak (×), rAUC (▲) and relative current amplitude of steady state (○) in the presence of glutamate with CNQX. (D) shows the deactivation time (τdec) and time constant of current decay (τB) in the presence of glutamate without or with CNQX. The holding potential was -40 mV in all experiments.

46

Results

Table3.7.1 Coapplication of CNQX and glutamate on GluR2L504Y receptor. CNQX rAmp rTTP rAUC rCdes 0.001 mM 0.99 + 0.01 1.00 + 0.04 0.98 + 0.01 0.96 + 0.01 0.01 mM 0.98 + 0.02 1.02 + 0.02 1.00 + 0.01 0.95 + 0.01 0.1 mM 0.83 + 0.02 0.91 + 0.04 0.73 + 0.01 0.75 + 0.03 p <0.001 0.02 <0.001 <0.001

47

Results

3.8 Effect of RPR119990 on AMPA-type receptor channels

3.8.1 RPR119990: GluR2L504Y receptor channels

1. Coapplication of RPR119990 and agonist

1

0.5 rAmp rTTP rAUC rCdes B

0 0.01 mM RPR119990 0.1 mM RPR119990 1 mM RPR119990

300 0 mM RPR119990 0.01 mM RPR119990 200 0.1 mM RPR119990 A ms 1 mM RPR119990 C 100

0 τdea τB

Fig.3.8.1 (A) shows averaged original current curves after 200 ms pulses of 10 mM glutamate or 10 mM glutamate + increasing concentration of RPR119990 (as indicated) applied to

GluR2L504Y channels. (B) shows the relative peak current amplitude (■), time to peak (×), rAUC

(▲) and relative current amplitude of steady state (○); (C) shows deactivation time (τdec) and time constant of current decay (τB1,2) in the presence of glutamate (as indicated) without or with RPR119990 . The holding potential was -40 mV in all experiments.

Coapplication experiments were performed using fast application patch-clamp technique applying 10 mM glutamate and different concentrations of RPR119990 (0.01, 0.1, and 1 mM; Fig.3.8.1 A). The data from different experiments were shown in Fig.3.8.1 B, C and Table 3.8.1. Fig.3.8.1 A showed averaged original current curves which were obtained from whole-cell patch expressing homomeric glutamate channels from the AMPA-type containing the GluR2L504Y -subunit. The relative peak current amplitude (Fig.3.8.1 A , B) was 0.98, 0.96 and 0.68, respectively, after coaplication of 10 mM glutamate and 0.01 mM , 0.1 mM or 1 mM RPR119990 . The rAUC was 48

Results 0.92, 0.76 and 0.31, respectively. As shown in table 3.8.1, both of them decreased significantly with the increasing concentration of RPR119990. The relative steady state was 0.94, 0.53, 0.08 respectively, decreased more obviously. At the same time, rTTP was 1.01, 0.98, and 1.01 respectively, remain unchanged.

A decrease of time constant of current decay (Fig.3.8.1 A, C), τB, as mono-exponential occurred at coapplication of 10 mM glutamate and 0.1 mM or 1 mM phenobarbital. τB decreased from 306.51 (305.70 + 45.77, n = 5) ms to 82.34 (85.08 + 3.85, n = 4) ms (p = 0.001). The deactivation time constant τdec (Fig.3.8.1 A, C) of 10 mM glutamate without or with 0.01, 0.1, 1 mM RPR119990 were 9.53 (10.87 + 1.51, n = 15), 8.66 (11.11 + 1.71, n = 5), 9.16 (11.17 + 2.44, n = 5), 9.41 (10.09 + 0.52, n = 4) ms respectively, with no significant difference (p = 0.98). RPR119990 blocks the

GluR2L504Y receptor channels dose-dependently.

Table3.5.1 Coapplication of RPR119990 and glutamate on GluR2L504Y receptor. PR119990 rAmp rTTP rAUC rCdes n 0.01 mM 0.97 + 0.01 0.97 + 0.03 0.91 + 0.02 0.93 + 0.01 5 0.1 mM 0.93 + 0.03 1.01 + 0.02 0.77 + 0.01 0.56 + 0.08 5 1 mM 0.72+ 0.04 0.99 + 0.01 0.34 + 0.03 0.10 + 0.05 4 p <0.001 0.52 <0.001 <0.001

2. Preincubation of patches with RPR119990 To different between competitive block effect and open-channel block effect we preincubated the patches with increasing concentrations of RPR119990 before applying 10 mM glutamate and the corresponding concentration of RPR119990. In Fig.3.8.2 an averaged original current curves were shown, which were obtained before and after preincubation whole-cell patch containing the GluR2L504Y subunit with increasing concentration RPR119990. The data from different experiments were showed in Fig.3.8.2 B. The relative peak current amplitude of preincubation tests with 0.1 nM, 1 nM, 10 nM, and 100 nM RPR119990 were 0.98 (0.99 + 0.03), 0.97 (0.95 + 0.03), 0.82 (0.79 + 0.04), and 0.16 (0.21 + 0.09) respectively, with significant difference (p < 0.001). When 1µM RPR119990 was used, the current was completely blocked. Compared to the coapplication experiment, the block effect is much stronger in the preincubation experiments. This indicates a competitive block effect.

49

Results

1 A B

0.5 rAmp

0 0.1 nM 1 nM 10 nM 100 nM 1000 nM RPR119990 RPR119990 RPR119990 RPR119990 RPR119990 Fig.3.8.2 (A) shows averaged original current curves before and after 30 s preincubation of wholc-cell patch transfected with GluR2L504Y channels with increasing concentration RPR119990 (as indicated). (B) shows the relative peak current amplitude (■) in the presence of glutamate after 30 s preincubation with RPR119990. The holding potential was -40 mV in all experiments. Currents were elicited by 200 ms pulses of 10mM glutamate.

3.8.2 RPR119990: GluR2flipGQ receptor channels

1. coapplication of RPR119990 and agonist

1

rAmp rTTP 0.5 B

0 0.01 mM 0.1 mM 1 mM RPR119990 RPR119990 RPR119990

15 τB

A 10

5 C

0 0 mM 0.01 mM 0.1 mM 1 mM RPR119990 RPR119990 RPR119990 RPR119990

Fig.3.8.3 (A) shows averaged original current curves after 50 ms pulses of 10 mM glutamate or 10

50

Results mM glutamate + increasing concentration of RPR119990 (as indicated) applied to GluR2flipGQ channels. (B) shows the relative peak current amplitude (■), time to peak (×) in the presence of glutamate with RPR119990. (C) shows the time constant of current decay (τB) in the presence of glutamate without or with RPR119990. The holding potential was -40 mV in all experiments.

To test the effect of RPR119990 in GluR2flipGQ receptor we measure the currents induced by 10 mM glutamate or 10 mM glutamate + increasing concentrations of RPR119990 (0.01, 0.1 or 1 mM).

The currents of Fig.3.8.3 A were obtained from one same cell which expressed GluR2flipGQ channels. Fig.3.8.3 B and C were the evaluation of the results from some independent experiments. After coapplication of 10 mM glutamate and 0.01, 0.1 or 1 RPR119990, the relative amplitude (Fig.3.8.3 A, B) was 0.99 (0.98 + 0.03), 0.99 (0.98 + 0.01) and 0.87 (0.87 + 0.03) respectively, since 1 mM has significant difference (p = 0.002). The rTTP was 0.99 (1.00 + 0.01), 0.98 (1.00 +

0.02), and 1.00 (0.98 + 0.02) respectively (p = 0.64). The time constant of current decay τB (Fig.3.8.3 A, B) of 10 mM glutamate without or with 0.01, 0.1 or 1mM RPR119990 was 15.60 (15.35 + 0.56), 14.97 (14.61 + 0.86), 12.4 (11.78 + 0.92) and 6.5 (6.31 + 0.37) ms respectively, with significant difference (p < 0.001).

2. Preincubation of patches with RPR119990

1

A B

0.5 rAmp

0 0.1 nM 1 nM 10 nM 100 nM 1000 nM RPR119990 RPR119990 RPR119990 RPR119990 RPR119990

51

Results Fig.3.8.4 (A) shows averaged original current curves before and after 30 s preincubation of whole-cell patch transfected with GluR2flipGQ channels with increasing concentration RPR119990 (as indicated). (B) shows the relative peak current amplitude (■) in the presence of glutamate after 30 s. preincubation with RPR119990. The holding potential was -40 mV in all experiments.

In the preincubation experiments, The rAmp (Fig.3.8.4) of 10 mM glutamate with 0.1, 1, 10, and 100 mM RPR119990 were 0.85 (0.86 + 0.03), 0.78 (0.82 + 0.04), 0.59 (0.63 + 0.06) and 0.11 (0.13 + 0.03) respectively, with significant difference (p < 0.001). When 1 mM RPR119990 was used, the currents were completely inhibited. It was apparent that the block effects of RPR119990 on AMPA-type receptor channels were mainly on preincubation experiments. This points to a competitive block mechanism.

52

Results 3.9. Comparison of the block effects of the different compounds tested

1

rAmp 0.5 rAUC

0 1 mM 1 mM 1 mM 1 mM 1 mM 1 mM 1 mM 0.1 mM minocycline riluzole memantine ketamine phenobarbital valproic acid RPR119990 CNQX Fig.3.9.1 shows the relative peak current amplitude and rAUC in the coapplication experiments of glutamate with different substances in GluR2L504Y receptor.

To compare the block effect among different substances, we used the rAmp and rAUC in coapplication experiments with 10 mM glutamate and 1 mM tested substances (except CNQX with 0.1 mM). As shown in Fig.3.9.1, when the parameter rAUC was compared to other tested drugs, RPR119990 showed the strongest effect, rAUC reached 0.34 + 0.03, and followed by memantine (0.72 + 0.03), riluzole (0.84 + 0.04), minocycline (0.93 + 0.04), phenobarbital (0.97 + 0.03) respectively. Ketamine and valproic acid showed no block effects. When rAmp was used to compare, the tendency is same, but the difference were not so obviously. The competitive effects of RPR119990 and riluzole must be considered additionally, which make the block effects of those substances much stronger in preincubation experiments. After preincubation with 1 mM riluzole, the rAUC was only 0.14 + 0.03; with 1µM RPR119990 the currents were completely inhibited.

53

Reference Discussion

1. Minocycline

Three effects of this substance were observed on AMPA-type receptor channels with the patch clamp technique in our experiments. On the one hand, it acts as an open-channel blocker on

GluR2L504Y, on the other hand, it has potentiating effect on GluR2flipGQ. As we know, these effects have not been reported so far. The block effect of minocycline can only be observed in coapplication measurements, not after preincubation of minocycline. This can be explained with an open-channel block effect. That means the block will be effective only after the channels open and can be proved by applying pulses of agonist and blocker together. It is characterized by a decrease of the time constant of current decay in fast application experiments104, 105, which is also shown in this experiment. Dose-dependent increasing reopening currents, which were registrated after removing application pulse due to an open-channel block mechanism. The reason for reopening currents can be explained by fast unbinding of blocker from the open state of the receptor after removing agonist and open-channel blocker, because the channel will stay for a short time in the open state before becoming deactivated. The potentiating effect of minocycline on AMPA-type receptor channels is out of expects. It was reported that some positive allosteric modulators improve performance of rats in the Morris water maze106. Also improvement of delayed recall in aged humans106, and gained special attention107 were reported. Currently, allosteric potentiators of AMPA receptors are divided into three groups according to structure107: the benzoylpiperidine-related potentiators (e.g., , such as 1-BCP, CX-516), the thiazide-related potentiators (e.g., , IDRA-21108) and PEPA (4-[2-(phenylsulphonylamino) ethylthio]- 2, 6- difluorophenoxya- cetamide50, 109). Among them, the strongest is cyclothiazide (CTZ), which has following effects: a fast block of desensitization; a slow current potentiation, a lengthening of the duration of single-channel openings (related to the increase of deactivation time); a shift in the relative weight of the conductance classes, and an increased apparent affinity for the agonist110. CTZ binds to AMPA receptors and ‘‘freezes’’ them in their existing state, i.e., nondesensitized or desensitized. Upon agonist application, the nondesensitized receptors shift to their open states, with a very small probability of undergoing desensitization, whereas the desensitized receptors slowly become available for activation, yielding a delayed potentiation.

54

Reference The potentiation effect of minocycline is different compared to that of CTZ. CTZ potentiated the AMPA current with increasing efficacy as the duration of CTZ preincubation was increased. When AMPA and CTZ were applied simultaneously, the time to peak did not change significantly, indicating that the desensitization induced by AMPA developed faster than the potentiating action of CTZ. But the potentiation effect of minocycline is stronger when it was coapplicated with glutamate than in the condition of preincubation. This proved that, the potentiating effect of minocycline shows a modulatory effect only in the opened channel. Probably, this interaction might take place after binding of the agonist to the receptor.

After coapplication of minocycline and 10 mM glutamate the τdes was not significantly changed, but the extent of desensitization. This suggests that the principal mechanism which might be explained by potentiating effect of minocycline on AMPA-type receptors is caused by the attenuating extent of receptor desensitization. The same effect on AMPA-type receptor was reported when the test drug

PEPA107 was used. This is different from cyclothiazide, which abolished the receptor’s desensitization; and different from aniracetam, which modulates both the onset and extent of desensitization in

AMPA-type receptor107.

The open-channel block effect and potentiation effect seems to be paradox. The explanation could be that the potentiation is depended from receptor desensitization, which exists in GluR2flipGQ and which is more effective than the open-channel block effect. Therefore only the potentiation effect could the oberserved when the desensitizing receptor GluR2flipGQ was used. While GluR2L504Y receptors show no desensitization, only the open-channel block was observed in the respective experiments. Recent studies have shown that minocycline is protective in different models of both acute and chronic neurodegeneration. It delayed disease onset and dose-dependently extended the survival of ALS mice (transgenic mice with the G93A human SOD1 mutation)70,71. Minocycline prevents glutamate-induced apoptosis in cerebellar granule neurons (CGNs)111. The mechanism of action is under investigation. With the patch clamp technique, we demonstrated an open-channel blockade of AMPA channels and a potentiating effect by minocycline, but both at very high concentration levels. The neuroprotective effects of minocycline might not be attributable to a direct effect on AMPA receptor channels. While the concentration of minocycline for open-channel block and potentiation is higher (≥ 1 mM) than the necessary concentration for neuroprotective effect (10–100 µM minocycline for regulations of p38 and Akt pathways111, 200 µM for inhibition of cytochrome C release in isolated brain mitochondria 71, 20 nM for inhibited microglial proliferation68) the role of the direct channel effects for neuroprotection has to be further characterized.

55

Reference 2 Riluzole

Our study shows that riluzole is able to modulate the AMPA receptors in a dose-dependent manner, showing two mechanism of action: competitive block and open-channel block.

In human GluR2flipGQ receptor, the effect of riluzole was only significant when the patches were preincubated with riluzole. In the non-desensitizing GluR2L504Y receptor, the effect was stronger when patches were preincubated, compared to single coapplication. When patches were preincubated with riluzole, the peak amplitude and rAUC were both reduced in a dose-dependent manner; the time to peak was prolonged at the same time. These are a typical competitive block effects.

In the coapplication experiments with GluR2L504Y receptors, at high concentration range of blocker (1 mM and 3 mM), riluzole showed a block effect, which was characterized with a increase rate of current decay with increasing concentrations105. Reopening currents after the coapplication of glutamate with 3 mM riluzole points to an open-channel block mechanism. The block effect of riluzole on AMPA-type receptor channels has been reported. A reduction of kainate-induced currents in spinal motor neurons in the presence of riluzole with IC50 value of 1.54 µM76 was described. This concentration could be arrived clinically (under the use of riluzole 50 mg twice daily, the peak serum concentration of riluzole is 0.99 + 0.85 µM (0.1-6.6 µM)112). But experiments with rat cortical neurons in primary culture revealed that riluzole decreased

77 kainate-induced currents dose-dependently with IC50 101 µM , and in another study it was reported that riluzole inhibits kainic acid-evoked responses in oocytes expression AMPA receptors

113 with IC50 of 167 µM . Those were more comparable to our results. In our experiments with

GluR2flipGQ, after preincubation with the lowest concentration of riluzole (0.1 mM), the rAMP and rAUC was 0.68 and 0.72, respectively. In experiments with GluR2L504Y, a significant reduction of the rAUC was observed when ≥ 0.1 mM riluzole was applied via preincubation. The difference might arise from the different receptor subtype and agonist which was used in the studies. Because riluzole reduced the kainate-induced currents in spinal motor neurons and in rat cortical neurons without changing the agonist EC50, it was presumed that riluzole blocks AMPA-type receptor in a non-competitive manner76, 77. But in one of the two experimental studies it was reported that riluzole reduced the probability of kainate-activated ionic channels to dwell in the open state with unchanged kinetic rate constants of desensitization and resensitisation; the block effect did not display any voltage-dependence77. Furthermore, if riluzole had only a non-competitive block effect, we would expect a stronger effect after coapplication compared to

56

Reference preincubation experiments. This suggestion could be excluded by our experiments: in our experiments with both GluR2flipGQ and GluR2L504Y receptors, the block effects were more efficient after preincubation. This points out the dominant competitive block effect of riluzole compared to the mild non-competitive block effect.

Riluzole has been the first compound that showed a therapeutic effect in the treatment of ALS2, 73. A neuroprotective effect has been shown in vivo and in vitro114, 115. Many mechanisms of action have been suggested: antagonism of neuronal glutamatergic transmission brought about by influencing different pre- and postsynaptic processes of glutamate transmission116, inhibit glutamate release, attenuate excitatory amino acid receptor activation, and decrease excitability of the postsynaptic cell membrane75, 117, decrease voltage-dependent sodium channels118, 119,

+ dose-dependently inhibited noninactivating (persistent) Na current (INa,p) up to a complete blocking of the current119, interact with channels120, decrease kainate induced currents in cortical neurons77, modulate small and large conductance Ca2+-activated K+ channels121, and potentiate postsynaptic-aminobutyric acid (GABA) responses78. Its effect may be multifactor. Among them, the influence on glutamatergic action is currently favored as explanation of neuroprotection, in which the directly effect on AMPA-type receptor channels may be also a factor.

3. Memantine

Our study shows that memantine acted dose-dependently on AMPA-type receptor channels as an open-channel blocker. The block effect of memantine was observed in the coapplication experiments with non-disensitisizing GluR2L504Y and human GluR2flipGQ receptor. Both experiments showed τB at coapplication 1 mM memantine is lower than that of 0.1 mM, which means a increase rate of current decay--an effect of open-channel block105. The reopening phenomenon was also observed after the pulse of coapplication glutamate with 1 mM memantine. In the preincubation experiments with both receptors, the block effect was also observed, but not stronger than that in the coapplication experiments. This excludes competitive block effect. It was reported that memantine (10–100 µM) had no effect on whole-cell inward currents (from cultured superior collicular and hippocampal neurons) to AMPA and kainate82, 122. This was accord with our result. But at high concentration (1 mM on human GluR2flipGQ receptor) it works as an open-channel blocker. On GluR2L504Y the block effect was observed since 0.1 mM 57

Reference memantine were used. This due to a higher sensitivity of memantine to the GluR2L504Y receptor. Memantine has been widely used for the treatment of dementia with excellent safety and efficacy profiles81, which antagonized NMDA receptor-mediated inward currents in a use and strongly voltage-dependent manner with IC50 of 1–3 µM (at -100 to -70 mV). It was reported that memantine protected cultured cortical neurons from the toxic effects of glutamate (100 µM for 20

82 83 h) with an IC50 of 1.4 µM , and prolongs survival in an ALS mouse model . But both the two concentrations are obviously lower than it can act on AMPA-type receptor channels. With daily doses between 5 and 30 mg the serum levels of memantine range from 0.025 to 0.529 µM123. In rats by acute i.p. administration of memantine 5–10 mg/kg leads to plasma levels of 1.0–3.2 µM124. These means that 0.1 mM or 1 mM concentrations memantine are too difficult to be reached in vitro, even memantine has excellent safety84. So the open-channel block effect of memantine on AMPA-type receptor channels is unlikely to be of therapeutic relevance.

4. Ketamine

Our study shows that ketamine, at as high as 1 mM concentration, has no effect on AMPA-type receptor channels.

Both in GluR2L504Y and GluR2flipGQ receptor channels the amplitude, area and time to peak of currents remain unchanged after the use of 1 mM ketamine. More than 1 mM ketamine we have not used, because anaesthetic plasma concentration of ketamine was 9.3 ± 0.8 µmol/l125 only. Neuroprotection by ketamine has been described in a variety of different experimental settings, including transient focal and global, as well as permanent ischemia, traumatic brain injury, and in-vitro hypoxia/ischemia86. As a well known open-channel blocker of NMDA receptor, 0.1µM or 1.0µM ketamine soon produced a pronounced reduction in the frequency of NMDA channel opening. A recent in-vitro study126 showed that 100 µM ketamine during or after 1 h of ischemic injury protected the cellular integrity in striatal slice cultures, although it did not affect neurotransmitter release from these cells. From our experiment, the direct effect of ketamine on AMPA-type receptors could be excluded.

5. Phenobarbital

Our study implies that Phenobarbital affects the AMPA receptors in a dose-dependent manner, showing two mechanism of action: open-channel block and competitive block.

58

Reference

In the coapplication experiments with GluR2L504Y receptors, with increasing concentration, phenobarbital showed a block effect, which was characterized with an obviously reduced current of steady-state and rAUC, a decrease of peak current before desensitization, and an increasing reopening currents amplitude after the coapplication of glutamate with Phenobarbital. These might be explained with mechanism of open-channel block. But the value of time constant of current decay does not change during experiments, which is argument against open-channel block. Additionally, the block effect of phenobarbital was more effective in preincubation experiments than in coapplication experiments. Both, the peak amplitude and the rAUC, were reduced in dose-dependent manner; the time to peak was prolonged at the same time. These results implicated an additional competitive block effect. The block effect of barbiturates on AMPA-type receptor channels has been reported, while there is no consistent hypothesis on the molecular mechanism. A study in cultured cortical neurons revealed that the inhibition of AMPA-type receptor channels by pentobarbital is use-dependent, an effect interpreted as being consistent with an open-channel block mechanism127. However, another study suggests that pentobarbital does not act as an open-channel blocker of AMPA-type receptor channels128. Rather, the sensitivity, use dependence, and trapping of inhibition by pentobarbital are determined by AMPA-type receptor channels desensitization128. From our results, the block mechanism of Phenobarbital on AMPA-type receptor seems to be mainly an open-channel block while there is an additional competitive block effect, too. As a traditional barbiturate type drug, phenobarbital has been widely used for its sedative, anesthetic, and anticonvulsant properties. It was reported that pre-treatment with phenobarbital results in a behavioral neuroprotection against kainic acid-induced neurotoxicity in animal models of toxic brain damage. Daily phenobarbital treatment (20 mg/kg i.p.) initiated 45 minutes prior to the kainic acid injection blocked the kainic acid-induced deficits in acquisition learning89. In our experiments, phenobarbital could block AMPA-type receptor channels, but at very high concentrations beyond clinically reached plasma levels. The concentration range for the block effect is according to the results of pentobarbital application on AMPA receptors from GluR2 null allele (-/-) neurons129, which are relatively insensitive to pentobarbital inhibition129. It was

130 131 reported that recombinant receptors that lack the GluR2 subunit or its Q/R edited site demonstrate a fivefold to 10-fold reduced sensitivity to pentobarbital, respectively. In our experiments, we used the GluR2L504Y, which is also permeable, which could be the reason why it seems to be insensitive to phenobarbital. Normally, the anticonvulsant therapeutic serum levels of phenobarbital is 10-40 µg/ml132,133. If there is an interaction on AMPA-type receptor with edited GluR2 subunit in this concentration range needs to be clarified in further

59

Reference studies.

6. Valproic acid

Our study shows that valproic acid has no direct effect on AMPA-type receptor channels.

In GluR2L504Y receptor channels, the current amplitude, area under the current curve and time to peak were not influenced when 30 mM valproic acid were coapplied or after preincubation with 1 mM valproic acid. Considering the antiepileptic therapeutic serum levels of 50-100 µg/ml134, 135, further concentrations were not tested in our experiments. Neuroprotection by valproic acid has been described. It is known to protect spinal motoneurons against death from glutamate toxicity in vitro. Furthermore, valproic acid can significantly prolong the disease duration in the ALS mouse model without changing the time point of onset92. As widely used antiepileptic drug, its pharmacological effects involve a variety of mechanisms, including increased gamma-aminobutyric acid (GABA)-ergic transmission, reduced release and/or effects of excitatory amino acids, blockade of voltage-gated sodium channels and modulation of dopaminergic and serotoninergic transmission90. Which of them might be help to protect motoneurons need to be clarified. From our experiments, a direct effect of valproic acid on AMPA-type receptors could be excluded.

7. CNQX

We test CNQX (FG 9065, 6-cyano-7-nitroquinoxaline-2,3-dione)，a well known competitive blockers of AMPA receptors, as a control drug.

In GluR2L504Y receptor channels, currents amplitude, area of current curve and time to peak reduced in coapplication experiments with 10 mM glutamate and 0.1 mM CNQX. τB was measured as 33.68 ms. The deactivation time remain unchanged. No reopening phenomenon was observed. The competitive block of CNQX on AMPA-type receptor channels has been widely researched. CNQX potently diminishes inward currents produced by kainate and quisqualate, and shifts the two dose-response curves to the right, with no change in the maximal response. CNQX applied alone did not evoke inward or outward currents and did not affect the current-voltage relationship

136 at membrane potentials between -90 and -30 mV . In preincubation experiments, its IC50 (the half maximal inhibitory concentration) was 0.3 + 0.15 µM93. As the first generation competitive

60

Reference blockers of non- NMDA receptors, it showed poor selectivity between AMPA and kainate receptors, and a non-competitive block against NMDA receptors. CNQX specifically prevented neurotoxicity induced by AMPA on adult rat spinal94. It could also block the neurotoxicity from cerebrospinal fluid from patients with ALS95. But its clinical use was restricted by the very low solubility, low selectivity, and nephrotoxicity in vivo137.

8. RPR119990

Our study shows that 9-carboxy methyl-4-oxo-5H, 10H-imidazo [1,2-a]-indeno[1,2-e]pyrazin-2- phosphonic acid (RPR 119990), block AMPA-type receptor channels competitively with a very high affinity.

Both in human GluR2flipGQ receptor and in nondisensitisized GluR2L504Y receptor, the effect of RPR119990 was much stronger when it was preincubated than it was coapplied. In coapplication experiments, since 0.1 mM RPR119990 reduced the currents amplitude and the time of current decay. But in preincubation experiments, after 1 µM RPR119990 preincubation the receptors were completely blocked. These are a typical competitive block effect. The block effect of RPR119990 on AMPA-type receptor channels has been reported64, 98, 99. Binding assays revealed a marked selectivity of AMPA/kainate-type receptor channels; recombinant AMPA-type receptor channels showed a competitive inhibition of kainate induced membrane currents by a parallel shift of the concentration-response curve to higher kainate concentrations without depression of the maximal responses. Preliminary pharmacological characterization revealed that it inhibit AMPA-type receptor channels in the nanomolar concentration range64, 98, 99. One binding site at the unliganded state of the receptor and another binding site connected to the mono-liganded AR state of the receptor were presumed64. This corresponds well with our data. After preincubation of patches with the blocker, a strong concentration dependent inhibition was shown. RPR 119990 was effective in different disease models like ischemic stroke, traumatic brain injury and epileptic seizures. It was reported that RPR119990 prolonged survival of transgenic ALS mouse model where it was able to improve grip muscle strength and glutamate uptake from spinal synaptosomal preparations98. The compound passes rapidly into the plasma after subcutaneous administration (achieved 534 ng/ml at 1 h in the case of the 1-mg/kg dose), although passage into the healthy brain was low, its half-life in the brain was quite long. The plasma pharmacokinetics appeared to be roughly dose proportional between 1 and 30 mg/kg. After oral administration at 3 mg/kg, plasma levels could reach 18 ng/ml at 1 h98. Consider its very high affinity as indicated in

61

Reference the low nanomolar range; this makes the compound extremely interesting for a therapeutic use in different neurological disorders such as neurodegenerative diseases. Future studies should settle the issue of AMPA-antagonism in clinical neuropharmacology.

62

Reference

Summarize

Direct neurotoxic effects of glutamate may play a role in the selective vulnerability of motoneurons in ALS. The AMPA-type receptor channels are promising targets for pharmacological neuroprotection. However, none of the AMPA-type receptor channel blockers developed so far has passed clinical testing. Therefore, we selected 8 substances to test their action on AMPA-type receptor channels to screen new antagonists and to analyse the pharmacological mechanisms of interaction.

1. Minocycline has both current enhancing and open-channel block effects on AMPA-type receptor channels. The current enhancing effect arises from attenuating the extent of receptor desensitization, and is stronger than its block effect. Because GluR2L504Y receptors do not desensitise, on it minocycline works only as an open-channel blocker. 2. Riluzole dose-dependently modulates the AMPA receptors in mainly competitive block manner and additionally open-channel block manner. 3. Memantine acts dose-dependently on AMPA-type receptor channels by an open-channel block like mechanism. 4. Phenobarbital interacts with the AMPA receptors in a high concentration range, in a dose-dependent manner, mainly via an open-channel block and, additionally, via a weak competitive block. 5. RPR119990 blocks AMPA-type receptor channels competitively, with a very high affinity in nanomolar range. Similar effects were observed for CNQX. 6. Ketamine and valproic acid have no effect on AMPA-type receptor channels. 7. The block effect of these substances on AMPA-type receptor channels is in turns RPR119990, CNQX, riluzole, minocycline, and phenobarbital. Only RPR119990 and CNQX inhibit AMPA-type receptor channels in the nanomolar concentration range, which could be reached in clinical use.

These information could help to better understand the effects of those compounds and thus to develop better therapeutic strategies that could protect motor neurons from excitotoxicity. Especially with respect to the missing success of translation of preclinical neuroprotection achieved by a great variety of compounds during the last decade to the patients suffering from progressive neurodegenerative diseases like amyotrophic lateral sclerosis, the pharmacological

63

Reference research has to be addressed back to the molecular mechanisms of action being relevant for neuroprotection, and exact drug profiles have to be studied and compared to biological effects. Only a systematic research approach seems to be promising for the development of molecular drug design in this field of research.

64

Reference Reference

1. Van Damme P, Dewil M, Robberecht W, Van Den Bosch L. Excitotoxicity and amyotrophic lateral sclerosis. Neurodegener Dis. 2005;2:147-159 2. Bensimon G, Lacomblez L, Meininger V. A controlled trial of riluzole in amyotrophic lateral sclerosis. ALS/Riluzole Study Group. N Engl J Med. 1994;330:585-591 3. Lacomblez L, Bensimon G, Leigh PN et al. Dose-ranging study of riluzole in amyotrophic lateral sclerosis. Amyotrophic Lateral Sclerosis/Riluzole Study Group II. Lancet. 1996;347:1425-1431 4. Rosen DR, Siddique T, Patterson D et al. Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature. 1993;362:59-62 5. Yang Y, Hentati A, Deng HX et al. The gene encoding alsin, a protein with three guanine-nucleotide exchange factor domains, is mutated in a form of recessive amyotrophic lateral sclerosis. Nat Genet. 2001;29:160-165 6. Chen YZ, Bennett CL, Huynh HM et al. DNA/RNA helicase gene mutations in a form of juvenile amyotrophic lateral sclerosis (ALS4). Am J Hum Genet. 2004;74:1128-1135 7. Nishimura AL, Mitne-Neto M, Silva HC et al. A mutation in the vesicle-trafficking protein VAPB causes late-onset spinal muscular atrophy and amyotrophic lateral sclerosis. Am J Hum Genet. 2004;75:822-831 8. Cleveland DW, Rothstein JD. From Charcot to Lou Gehrig: deciphering selective motor neuron death in ALS. Nat Rev Neurosci. 2001;2:806-819 9. Brown RH, Jr., Robberecht W. Amyotrophic lateral sclerosis: pathogenesis. Semin Neurol. 2001;21:131-139 10. Heath PR, Shaw PJ. Update on the glutamatergic neurotransmitter system and the role of excitotoxicity in amyotrophic lateral sclerosis. Muscle Nerve. 2002;26:438-458 11. Julien JP. Amyotrophic lateral sclerosis. unfolding the toxicity of the misfolded. Cell. 2001;104:581-591 12. Benveniste H, Drejer J, Schousboe A, Diemer NH. Elevation of the extracellular concentrations of glutamate and aspartate in rat during transient cerebral ischemia monitored by intracerebral microdialysis. J Neurochem. 1984;43:1369-1374 13. Kvamme E, Schousboe A, Hertz L et al. Developmental change of endogenous glutamate and gamma-glutamyl transferase in cultured cerebral cortical interneurons and cerebellar granule cells, and in mouse cerebral cortex and cerebellum in vivo. Neurochem Res. 1985;10:993-1008 14. Meldrum B, Garthwaite J. Excitatory amino acid neurotoxicity and neurodegenerative disease. Trends Pharmacol Sci. 1990;11:379-387 15. Rosenberg PA, Amin S, Leitner M. Glutamate uptake disguises neurotoxic potency of glutamate agonists in cerebral cortex in dissociated cell culture. J Neurosci. 1992;12:56-61 16. Novelli A, Reilly JA, Lysko PG, Henneberry RC. Glutamate becomes neurotoxic via the N-methyl-D-aspartate receptor when intracellular energy levels are reduced. Brain Res. 1988;451:205-212 17. Henneberry RC, Novelli A, Cox JA, Lysko PG. Neurotoxicity at the N-methyl-D-aspartate receptor in energy-compromised neurons. An hypothesis for cell death in aging and disease.

65

Reference Ann N Y Acad Sci. 1989;568:225-233 18. Carriedo SG, Yin HZ, Weiss JH. Motor neurons are selectively vulnerable to AMPA/kainate receptor-mediated injury in vitro. J Neurosci. 1996;16:4069-4079 19. Van Den Bosch L, Vandenberghe W, Klaassen H et al. Ca(2+)-permeable AMPA receptors and selective vulnerability of motor neurons. J Neurol Sci. 2000;180:29-34 20. Van den Bosch L, Van Damme P, Vleminckx V et al. An alpha-mercaptoacrylic acid derivative (PD150606) inhibits selective motor neuron death via inhibition of kainate-induced Ca2+ influx and not via inhibition. Neuropharmacology. 2002;42:706-713 21. Simeone TA, Sanchez RM, Rho JM. Molecular biology and ontogeny of glutamate receptors in the mammalian central nervous system. J Child Neurol. 2004;19:343-360; discussion 361 22. Hollmann M, Heinemann S. Cloned glutamate receptors. Annu Rev Neurosci. 1994;17:31-108 23. Collingridge GL, Lester RA. Excitatory amino acid receptors in the vertebrate central nervous system. Pharmacol Rev. 1989;41:143-210 24. Nicoll RA, Malenka RC. Expression mechanisms underlying NMDA receptor-dependent long-term potentiation. Ann N Y Acad Sci. 1999;868:515-525 25. Van Den Bosch L, Robberecht W. Different receptors mediate motor neuron death induced by short and long exposures to excitotoxicity. Brain Res Bull. 2000;53:383-388 26. Kruman, II, Pedersen WA, Springer JE, Mattson MP. ALS-linked Cu/Zn-SOD mutation increases vulnerability of motor neurons to excitotoxicity by a mechanism involving increased oxidative stress and perturbed calcium homeostasis. Exp Neurol. 1999;160:28-39 27. Hugon J, Vallat JM, Spencer PS et al. Kainic acid induces early and delayed degenerative neuronal changes in rat spinal cord. Neurosci Lett. 1989;104:258-262 28. Ikonomidou C, Qin Qin Y, Labruyere J, Olney JW. Motor neuron degeneration induced by excitotoxin agonists has features in common with those seen in the SOD-1 transgenic mouse model of amyotrophic lateral sclerosis. J Neuropathol Exp Neurol. 1996;55:211-224 29. Saroff D, Delfs J, Kuznetsov D, Geula C. Selective vulnerability of spinal cord motor neurons to non-NMDA toxicity. Neuroreport. 2000;11:1117-1121 30. Rothstein JD, Jin L, Dykes-Hoberg M, Kuncl RW. Chronic inhibition of glutamate uptake produces a model of slow neurotoxicity. Proc Natl Acad Sci U S A. 1993;90:6591-6595 31. Wu TY, Liu CI, Chang YC. A study of the oligomeric state of the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-preferring glutamate receptors in the synaptic junctions of porcine brain. Biochem J. 1996;319 (Pt 3):731-739 32. Mano I, Teichberg VI. A tetrameric subunit stoichiometry for a glutamate receptor-channel complex. Neuroreport. 1998;9:327-331 33. Safferling M, Tichelaar W, Kummerle G et al. First images of a glutamate receptor ion channel: oligomeric state and molecular dimensions of GluRB homomers. Biochemistry. 2001;40:13948-13953 34. Keinanen K, Wisden W, Sommer B et al. A family of AMPA-selective glutamate receptors. Science. 1990;249:556-560 35. Rosenmund C, Stern-Bach Y, Stevens CF. The tetrameric structure of a glutamate receptor channel. Science. 1998;280:1596-1599 36. Palmer CL, Cotton L, Henley JM. The molecular pharmacology and cell biology of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors. Pharmacol Rev. 2005;57:253-277

66

Reference 37. Rogers SW, Hughes TE, Hollmann M et al. The characterization and localization of the glutamate receptor subunit GluR1 in the rat brain. J Neurosci. 1991;11:2713-2724 38. Stern-Bach Y, Russo S, Neuman M, Rosenmund C. A point mutation in the glutamate binding site blocks desensitization of AMPA receptors. Neuron. 1998;21:907-918 39. Stern-Bach Y, Bettler B, Hartley M et al. Agonist selectivity of glutamate receptors is specified by two domains structurally related to bacterial amino acid-binding proteins. Neuron. 1994;13:1345-1357 40. Jayaraman V, Keesey R, Madden DR. Ligand--protein interactions in the glutamate receptor. Biochemistry. 2000;39:8693-8697 41. Okubo Y, Kakizawa S, Hirose K, Iino M. Cross talk between metabotropic and ionotropic glutamate receptor-mediated signaling in parallel fiber-induced inositol 1,4,5-trisphosphate production in cerebellar Purkinje cells. J Neurosci. 2004;24:9513-9520 42. Johansen TN, Greenwood JR, Frydenvang K et al. Stereostructure-activity studies on agonists at the AMPA and kainate subtypes of ionotropic glutamate receptors. Chirality. 2003;15:167-179 43. Henley JM. Proteins interactions implicated in AMPA receptor trafficking: a clear destination and an improving route map. Neurosci Res. 2003;45:243-254 44. Malenka RC. and AMPA receptor trafficking. Ann N Y Acad Sci. 2003;1003:1-11 45. Sommer B, Keinanen K, Verdoorn TA et al. Flip and flop: a cell-specific functional switch in glutamate-operated channels of the CNS. Science. 1990;249:1580-1585 46. Mosbacher J, Schoepfer R, Monyer H et al. A molecular determinant for submillisecond desensitization in glutamate receptors. Science. 1994;266:1059-1062 47. Koike M, Tsukada S, Tsuzuki K et al. Regulation of kinetic properties of GluR2 AMPA receptor channels by alternative splicing. J Neurosci. 2000;20:2166-2174 48. Partin KM, Patneau DK, Mayer ML. Cyclothiazide differentially modulates desensitization of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor splice variants. Mol Pharmacol. 1994;46:129-138 49. Kessler M, Rogers G, Arai A. The norbornenyl moiety of cyclothiazide determines the preference for flip-flop variants of AMPA receptor subunits. Neurosci Lett. 2000;287:161-165 50. Sekiguchi M, Takeo J, Harada T et al. Pharmacological detection of AMPA receptor heterogeneity by use of two allosteric potentiators in rat hippocampal cultures. Br J Pharmacol. 1998;123:1294-1303 51. Shen Y, Yang XL. modulation of AMPA receptors may be relevant to splice variants in carp retina. Neurosci Lett. 1999;259:177-180 52. Karkanias NB, Papke RL. Lithium modulates desensitization of the glutamate receptor subtype gluR3 in Xenopus oocytes. Neurosci Lett. 1999;277:153-156 53. Arvola M, Keinanen K. Characterization of the ligand-binding domains of glutamate receptor (GluR)-B and GluR-D subunits expressed in Escherichia coli as periplasmic proteins. J Biol Chem. 1996;271:15527-15532 54. Sommer B, Kohler M, Sprengel R, Seeburg PH. RNA editing in brain controls a determinant of ion flow in glutamate-gated channels. Cell. 1991;67:11-19 55. Hollmann M, Hartley M, Heinemann S. Ca2+ permeability of KA-AMPA--gated glutamate receptor channels depends on subunit composition. Science. 1991;252:851-853

67

Reference 56. Burnashev N, Villarroel A, Sakmann B. Dimensions and ion selectivity of recombinant AMPA and kainate receptor channels and their dependence on Q/R site residues. J Physiol. 1996;496 (Pt 1):165-173 57. Swanson GT, Kamboj SK, Cull-Candy SG. Single-channel properties of recombinant AMPA receptors depend on RNA editing, splice variation, and subunit composition. J Neurosci. 1997;17:58-69 58. Utz AL, Verdoorn TA. Recombinant AMPA receptors with low Ca2+ permeability increase intracellular Ca2+ in HEK 293 cells. Neuroreport. 1997;8:1975-1980 59. Liu SQ, Cull-Candy SG. Synaptic activity at calcium-permeable AMPA receptors induces a switch in receptor subtype. Nature. 2000;405:454-458 60. Lomeli H, Mosbacher J, Melcher T et al. Control of kinetic properties of AMPA receptor channels by nuclear RNA editing. Science. 1994;266:1709-1713 61. Krampfl K, Schlesinger F, Zorner A et al. Control of kinetic properties of GluR2 flop AMPA-type channels: impact of R/G nuclear editing. Eur J Neurosci. 2002;15:51-62 62. Krampfl K, Schlesinger F, Wolfes H et al. Functional diversity of recombinant human AMPA type glutamate receptors: possible implications for selective vulnerability of motor neurons. J Neurol Sci. 2001;191:19-23 63. Schlesinger F, Tammena D, Krampfl K, Bufler J. Two mechanisms of action of the adamantane derivative IEM-1460 at human AMPA-type glutamate receptors. Br J Pharmacol. 2005;145:656-663 64. Krampfl K, Schlesinger F, Cordes AL, Bufler J. Molecular analysis of the interaction of the pyrazine derivatives RPR119990 and RPR117824 with human AMPA-type glutamate receptor channels. Neuropharmacology. 2006;50:479-490 65. Mansour M, Nagarajan N, Nehring RB et al. Heteromeric AMPA receptors assemble with a preferred subunit stoichiometry and spatial arrangement. Neuron. 2001;32:841-853 66. Klein NC, Cunha BA. Tetracyclines. Med Clin North Am. 1995;79:789-801 67. Barza M, Brown RB, Shanks C et al. Relation between lipophilicity and pharmacological behavior of minocycline, doxycycline, tetracycline, and oxytetracycline in dogs. Antimicrob Agents Chemother. 1975;8:713-720 68. Tikka TM, Vartiainen NE, Goldsteins G et al. Minocycline prevents neurotoxicity induced by cerebrospinal fluid from patients with motor neurone disease. Brain. 2002;125:722-731 69. Tikka T, Fiebich BL, Goldsteins G et al. Minocycline, a tetracycline derivative, is neuroprotective against excitotoxicity by inhibiting activation and proliferation of microglia. J Neurosci. 2001;21:2580-2588 70. Van Den Bosch L, Tilkin P, Lemmens G, Robberecht W. Minocycline delays disease onset and mortality in a transgenic model of ALS. Neuroreport. 2002;13:1067-1070 71. Zhu S, Stavrovskaya IG, Drozda M et al. Minocycline inhibits cytochrome c release and delays progression of amyotrophic lateral sclerosis in mice. Nature. 2002;417:74-78 72. Kriz J, Nguyen MD, Julien JP. Minocycline slows disease progression in a mouse model of amyotrophic lateral sclerosis. Neurobiol Dis. 2002;10:268-278 73. Lacomblez L, Bensimon G, Leigh PN et al. A confirmatory dose-ranging study of riluzole in ALS. ALS/Riluzole Study Group-II. Neurology. 1996;47:S242-250 74. Gurney ME, Cutting FB, Zhai P et al. Benefit of vitamin E, riluzole, and in a transgenic model of familial amyotrophic lateral sclerosis. Ann Neurol. 1996;39:147-157

68

Reference 75. Doble A. The pharmacology and mechanism of action of riluzole. Neurology. 1996;47:S233-241 76. Albo F, Pieri M, Zona C. Modulation of AMPA receptors in spinal motor neurons by the neuroprotective agent riluzole. J Neurosci Res. 2004;78:200-207 77. Zona C, Cavalcanti S, De Sarro G et al. Kainate-induced currents in rat cortical neurons in culture are modulated by riluzole. Synapse. 2002;43:244-251 78. He Y, Benz A, Fu T et al. Neuroprotective agent riluzole potentiates postsynaptic GABA(A) receptor function. Neuropharmacology. 2002;42:199-209 79. Mohammadi B, Krampfl K, Moschref H et al. Interaction of the neuroprotective drug riluzole with GABA(A) and receptor channels. Eur J Pharmacol. 2001;415:135-140 80. Chen HS, Pellegrini JW, Aggarwal SK et al. Open-channel block of N-methyl-D-aspartate (NMDA) responses by memantine: therapeutic advantage against NMDA receptor-mediated neurotoxicity. J Neurosci. 1992;12:4427-4436 81. Reisberg B, Doody R, Stoffler A et al. Memantine in moderate-to-severe Alzheimer's disease. N Engl J Med. 2003;348:1333-1341 82. Parsons CG, Panchenko VA, Pinchenko VO et al. Comparative patch-clamp studies with freshly dissociated rat hippocampal and striatal neurons on the NMDA receptor antagonistic effects of and memantine. Eur J Neurosci. 1996;8:446-454 83. Wang R, Zhang D. Memantine prolongs survival in an amyotrophic lateral sclerosis mouse model. Eur J Neurosci. 2005;22:2376-2380 84. Parsons CG, Danysz W, Quack G. Memantine is a clinically well tolerated N-methyl-D-aspartate (NMDA) receptor antagonist--a review of preclinical data. Neuropharmacology. 1999;38:735-767 85. MacDonald JF, Bartlett MC, Mody I et al. Actions of ketamine, and MK-801 on NMDA receptor currents in cultured mouse hippocampal neurones. J Physiol. 1991;432:483-508 86. Koerner IP, Brambrink AM. Brain protection by anesthetic agents. Curr Opin Anaesthesiol. 2006;19:481-486 87. Kamiya Y, Andoh T, Furuya R et al. Comparison of the effects of convulsant and depressant barbiturate stereoisomers on AMPA-type glutamate receptors. Anesthesiology. 1999;90:1704-1713 88. Zhu H, Cottrell JE, Kass IS. The effect of thiopental and propofol on NMDA- and AMPA-mediated glutamate excitotoxicity. Anesthesiology. 1997;87:944-951 89. Brown-Croyts LM, Caton PW, Radecki DT, McPherson SL. Phenobarbital pre-treatment prevents kainic acid-induced impairments in acquisition learning. Life Sci. 2000;67:643-650 90. Perucca E. Pharmacological and therapeutic properties of valproate: a summary after 35 years of clinical experience. CNS Drugs. 2002;16:695-714 91. Dou H, Birusingh K, Faraci J et al. Neuroprotective activities of sodium valproate in a murine model of human immunodeficiency virus-1 encephalitis. J Neurosci. 2003;23:9162-9170 92. Sugai F, Yamamoto Y, Miyaguchi K et al. Benefit of valproic acid in suppressing disease progression of ALS model mice. Eur J Neurosci. 2004;20:3179-3183 93. Honore T, Davies SN, Drejer J et al. Quinoxalinediones: potent competitive non-NMDA glutamate receptor antagonists. Science. 1988;241:701-703 94. Kwak S, Nakamura R. Acute and late neurotoxicity in the rat spinal cord in vivo induced by

69

Reference glutamate receptor agonists. J Neurol Sci. 1995;129 Suppl:99-103 95. Couratier P, Hugon J, Sindou P et al. Cell culture evidence for neuronal degeneration in amyotrophic lateral sclerosis being linked to glutamate AMPA/kainate receptors. Lancet. 1993;341:265-268 96. Doble A. The role of excitotoxicity in neurodegenerative disease: implications for therapy. Pharmacol Ther. 1999;81:163-221 97. Dingledine R, Borges K, Bowie D, Traynelis SF. The glutamate receptor ion channels. Pharmacol Rev. 1999;51:7-61 98. Canton T, Bohme GA, Boireau A et al. RPR 119990, a novel alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid antagonist: synthesis, pharmacological properties, and activity in an animal model of amyotrophic lateral sclerosis. J Pharmacol Exp Ther. 2001;299:314-322 99. Mignani S, Bohme GA, Birraux G et al. 9-Carboxymethyl-5H,10H-imidazo[1,2-a]indeno[1,2-e]pyrazin-4-one-2-carbocyl ic acid (RPR117824): selective anticonvulsive and neuroprotective AMPA antagonist. Bioorg Med Chem. 2002;10:1627-1637 100. Hamill OP, Marty A, Neher E et al. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflugers Arch. 1981;391:85-100 101. Franke C, Hatt H, Dudel J. Liquid filament switch for ultra-fast exchanges of solutions at excised patches of synaptic membrane of crayfish muscle. Neurosci Lett. 1987;77:199-204 102. Heckmann M, Bufler J, Franke C, Dudel J. Kinetics of homomeric GluR6 glutamate receptor channels. Biophys J. 1996;71:1743-1750 103. Jahn K, Bufler J, Franke C. Kinetics of AMPA-type glutamate receptor channels in rat caudate-putamen neurones show a wide range of desensitization but distinct recovery characteristics. Eur J Neurosci. 1998;10:664-672 104. Bufler J, Franke C, Parnas H, Dudel J. Open channel block by physostigmine and procaine in embryonic-like nicotinic receptors of mouse muscle. Eur J Neurosci. 1996;8:677-687 105. Bufler J, Wilhelm R, Parnas H et al. Open channel and competitive block of the embryonic form of the nicotinic receptor of mouse myotubes by (+)-tubocurarine. J Physiol. 1996;495 (Pt 1):83-95 106. Staubli U, Rogers G, Lynch G. Facilitation of glutamate receptors enhances memory. Proc Natl Acad Sci U S A. 1994;91:777-781 107. Sekiguchi M, Nishikawa K, Aoki S, Wada K. A desensitization-selective potentiator of AMPA-type glutamate receptors. Br J Pharmacol. 2002;136:1033-1041 108. Yamada KA. Modulating excitatory synaptic : potential treatment for neurological disease? Neurobiol Dis. 1998;5:67-80 109. Sekiguchi M, Fleck MW, Mayer ML et al. A novel allosteric potentiator of AMPA receptors: 4--2-(phenylsulfonylamino)ethylthio--2,6-difluoro-phenoxyaceta mide. J Neurosci. 1997;17:5760-5771 110. Fucile S, Miledi R, Eusebi F. Effects of cyclothiazide on GluR1/AMPA receptors. Proc Natl Acad Sci U S A. 2006;103:2943-2947 111. Pi R, Li W, Lee NT et al. Minocycline prevents glutamate-induced apoptosis of cerebellar granule neurons by differential regulation of p38 and Akt pathways. J Neurochem. 2004;91:1219-1230

70

Reference 112. Groeneveld GJ, Van Kan HJ, Kalmijn S et al. Riluzole serum concentrations in patients with ALS: associations with side effects and symptoms. Neurology. 2003;61:1141-1143 113. Debono MW, Le Guern J, Canton T et al. Inhibition by riluzole of electrophysiological responses mediated by rat kainate and NMDA receptors expressed in Xenopus oocytes. Eur J Pharmacol. 1993;235:283-289 114. Peluffo H, Estevez A, Barbeito L, Stutzmann JM. Riluzole promotes survival of rat motoneurons in vitro by stimulating trophic activity produced by spinal monolayers. Neurosci Lett. 1997;228:207-211 115. Couratier P, Sindou P, Esclaire F et al. Neuroprotective effects of riluzole in ALS CSF toxicity. Neuroreport. 1994;5:1012-1014 116. Bryson HM, Fulton B, Benfield P. Riluzole. A review of its pharmacodynamic and pharmacokinetic properties and therapeutic potential in amyotrophic lateral sclerosis. Drugs. 1996;52:549-563 117. Centonze D, Calabresi P, Pisani A et al. of the neuroprotective agent riluzole on striatal spiny neurons. Neuropharmacology. 1998;37:1063-1070 118. Benoit E, Escande D. Riluzole specifically blocks inactivated Na channels in myelinated nerve fibre. Pflugers Arch. 1991;419:603-609 119. Urbani A, Belluzzi O. Riluzole inhibits the persistent sodium current in mammalian CNS neurons. Eur J Neurosci. 2000;12:3567-3574 120. Xu L, Enyeart JA, Enyeart JJ. Neuroprotective agent riluzole dramatically slows inactivation of Kv1.4 potassium channels by a voltage-dependent oxidative mechanism. J Pharmacol Exp Ther. 2001;299:227-237 121. Cao YJ, Dreixler JC, Couey JJ, Houamed KM. Modulation of recombinant and native neuronal SK channels by the neuroprotective drug riluzole. Eur J Pharmacol. 2002;449:47-54 122. Parsons CG, Gruner R, Rozental J et al. Patch clamp studies on the kinetics and selectivity of N-methyl-D-aspartate receptor antagonism by memantine (1-amino-3,5-dimethyladamantan). Neuropharmacology. 1993;32:1337-1350 123. Kornhuber J, Quack G. Cerebrospinal fluid and serum concentrations of the N-methyl-D-aspartate (NMDA) receptor antagonist memantine in man. Neurosci Lett. 1995;195:137-139 124. Danysz W, Parsons CG, Kornhuber J et al. Aminoadamantanes as NMDA receptor antagonists and antiparkinsonian agents--preclinical studies. Neurosci Biobehav Rev. 1997;21:455-468 125. Idvall J, Ahlgren I, Aronsen KR, Stenberg P. Ketamine infusions: pharmacokinetics and clinical effects. Br J Anaesth. 1979;51:1167-1173 126. Basagan-Mogol E, Buyukuysal RL, Korfali G. Effects of ketamine and thiopental on ischemia reoxygenation-induced LDH leakage and amino acid release from rat striatal slices. J Neurosurg Anesthesiol. 2005;17:20-26 127. Marszalec W, Narahashi T. Use-dependent pentobarbital block of kainate and quisqualate currents. Brain Res. 1993;608:7-15 128. Jackson MF, Joo DT, Al-Mahrouki AA et al. Desensitization of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors facilitates use-dependent inhibition by pentobarbital. Mol Pharmacol. 2003;64:395-406 129. Joo DT, Xiong Z, MacDonald JF et al. Blockade of glutamate receptors and barbiturate anesthesia: increased sensitivity to pentobarbital-induced anesthesia despite reduced inhibition

71

Reference of AMPA receptors in GluR2 null mutant mice. Anesthesiology. 1999;91:1329-1341 130. Taverna FA, Cameron BR, Hampson DL et al. Sensitivity of AMPA receptors to pentobarbital. Eur J Pharmacol. 1994;267:R3-5 131. Yamakura T, Sakimura K, Mishina M, Shimoji K. The sensitivity of AMPA-selective glutamate receptor channels to pentobarbital is determined by a single amino acid residue of the alpha 2 subunit. FEBS Lett. 1995;374:412-414 132. Schmidt D, Haenel F. Therapeutic plasma levels of phenytoin, phenobarbital, and carbamazepine: individual variation in relation to seizure frequency and type. Neurology. 1984;34:1252-1255 133. Babaei A, Eslamai MH. Evaluation of therapeutic drug level monitoring of phenobarbital, phenytoin and carbamazepine in Iranian epileptic patients. Int J Clin Pharmacol Ther. 2007;45:121-125 134. Cramer JA, Mattson RH, Bennett DM, Swick CT. Variable free and total valproic acid concentrations in sole- and multi-drug therapy. Ther Drug Monit. 1986;8:411-415 135. ES ELD, Fuseau E, S ELDA, Cosson V. Pharmacokinetic modelling of valproic acid from routine clinical data in Egyptian epileptic patients. Eur J Clin Pharmacol. 2004;59:783-790 136. Yamada KA, Dubinsky JM, Rothman SM. Quantitative physiological characterization of a non-NMDA receptor antagonist. J Neurosci. 1989;9:3230-3236 137. Weiser T. AMPA receptor antagonists for the treatment of stroke. Curr Drug Targets CNS Neurol Disord. 2005;4:153-159

72

Curriculum vitae

Curriculum vitae

Personal data: Name: Lingjing Jin Gender: Male Nationality: Chinese Birthplace : Nei Mongol, P.R. China Birthday: 09-12-1975 Marital status: Married E-mail: [email protected]

Education:

Year Institution Degree Field of study

9.1989-7.1992 Wu Da Mineral Affairs Bureau No.2 Senior high ______Senior High School school

9.1992-7.1997 Medical College of Shanghai Bachelor Medicine(Clinical Railway University, P.R.China Medicine)

9.2000-5.2003 Medical College of Shanghai Tongji Master Image Medicine and University, P.R.China Nuclear Medicine (Neuroradiology)

Clinical training and experiences

7.1997-9.2000 Resident in the department of neurology, Tongji hospital of Shanghai Tongji University, P.R.China

7.2001-5.2003 Resident in the department of neuroradiology(MRI), Tongji hospital of Shanghai Tongji University

5.2003-5.2004 Doctor-in-charge in the department of neurology, Tongji hospital of Shanghai Tongji University

5.2004-10.2005 Doctor-in-charge in the department of neurology and neurointervention, Cerebral Vascular Disease Center of Shanghai Tongji University

10.2005- now Resident in the department of neurology and neurophysiology, Medical school of Hannover, Germany

73

Curriculum vitae Publications 1) Jin L.J., Yang Z.Y., Chen F., et al. Magnetic resonance chemical shift imaging for acute cerebral infarction. Chin J Neurol, 2004,37(4):8-11

2) Jin L.J., Chen Z.Q.， Han H.J.. Drug Eluting Stent for Treatment of Vertebral Stenosis

(Report of one case). Chin J Cerebrovase Dis,2005,2(1):35-36

3) Jin L.J., Yang Z.Y.. The value of MR diffusion-weighted imaging in diagnosing acute cerebral infarction. Journal of Tongji University(medical science),2003,24(1):34-37

4) Jin L.J.. Development of new magnetic resonance imaging technique in acute ischemic stroke, Journal of Tongji University(medical science),2002,23(5):445-448

5) Jin L.J.. Hou H.G.. Development of diffusion- and perfusion-weighted imaging technique in acute ischemic stroke, Chinese journal of neurology and neurorehabilitation, 2002,7(2)58-61

6) Jin L.J., Yang Z.Y., Wu Y.Z.. CT findings of subcortical arteriosclerotic encephalopathy(SAE), Chinese journal of medical imaging technology. 2001,17(sup):45-46

7) Yang Z.Y., Jin L.J.. Magnetic Resonance Chemical Shift Imaging for the Metabolites

Evolution in Cerebral Infarction, Chinese journal of medical imaging. 2005,13(6):401-404

8) Dai, G.C. Jin L.J., Yang Z.Y.. An evaluation of diagnosing cerebral infarction by diffusion-weighted imaging and fluid-attenuated-inversion-recover, J Med Imaging, 2004,

14(8):609-611

9) Jin R.X., Jin L.J., Wang G.Y.,et al. Preliminary Detect ion on the Mutation of Alpha-synucle in Gene in Sporadic Parkinson’s Disease Patients. Newsletter on Neurochemistry,

1999,12(1):10-13

10) Chen Y.H., Wang G.Y., Jin L.J.. Analysis of the risk factors of cerebral hemorrhage patients accompanied by arrhythmic, Henan Journal of Practical Nervous Diseases, 2003, 6(5):7-8

11) Wang G.Y., Jiao R., Jin L.J.. Clinical study of identifying the stability of atherosclerotic plaque in carotid with3D MOTSA, Chin J Crit Care Med,2004,24(3):198-199

74

Declaration

Declaration (Erklärung nach §2 Abs. 2 Nr. 5 und 6)

Ich erkläre, daß ich die der Medizinischen Hochschule Hannover zur Promotion eingereichte Dissertation mit dem Titel:

Antagonists of AMPA-type receptor channels: A Patch Clamp Study im Zentrum Neurologie und Neurophysiologie der Medizinischen Hochschule Hannover (Direktor：Prof. Dr. Med. Reinhard Dengler) unter Betreuung von Herrn PD. Dr. med. Klaus Krampfl ohne sonstige Hilfe durchgeführt und bei der Abfassung der Dissertation keine anderen als die dort aufgeführten Hilfsmittel benutzt habe. Ich habe bisher an keiner in- oder ausländischen Medizinischen Fakultät ein Gesuch um Zulassung zur Promotion eingereicht noch die vorliegende oder eine andere Arbeit als Dissertation vorgelegt.

Hannover, den ……………..

...... Lingjing Jin

75

Acknowledgements Acknowledgements

My sincere gratitude goes to my advisor, all my friends, and my colleagues in the neurological department of medical school of Hannover for their support and patience over the last two years.

I would like to thank Prof. Dr. med. R. Dengler, who gave me the chance to learn in MHH and gave me so much help during my work and study in Hannover. He cares of my doctor works constantly and gives me great confidence.

I want to say special thanks to my advisor, PD. Dr. med. K. Krampfl. He gave me the chance to learn basic electrophysiological methods, especially the patch clamp technique. Addionally thanks for support and the excellent education during the whole two years.

Special thanks go to Dr. med F.Schlesinger who is a patient guidance and taught me not only the knowledge of patch clamp, but also how to use different devices. He gave me valuable advices and gave me support during the dissertation writing and study.

Dr. med E. Ziegler, Dr.med. K. Jahn, had shown me basic technique for molecular experiments.

I sincerely thank Dr. med. S. Petri, K. Kollewe and all colleges in our labor for their concerns.

I am grateful to Mr. J.Kilian and Mr. A. Niesel for technical help.

Many people on the faculty and staff of the department assisted and encouraged me in various ways during my course.

I want to say thanks to my teachers and colleges in Tongji University for so much help to me and my family.

Special thanks go to Frau Dr. H. Guo, Frau Y.H. Ji, Frau Dr. Pi and all the colleges from Shanghai for supporting my project in Germany.

At last I want to say thanks to my wife W.J. Ding, my daughter D.Y Jin, my parents, parents-in-law and my sisters.

76