| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Medical Chemistry, University of Szeged;
Department of Comparative Physiology, University of Szeged;
Department of Analytical and Inorganic Chemistry, University of Szeged;
Institute of Biophysics, Biological Research Centre, Hungarian Academy of Sciences;
|| Department of Biology, Juhász Gyula College, University of Szeged; and
¶ Institute of Biochemistry, Biological Research Center of the Hungarian Academy of Sciences, Szeged, Hungary
1Correspondence: Department of Medical Chemistry, University of Szeged, Dóm Square 8, Szeged 6720, Hungary. E-mail: szegv{at}mdche.szote.u-szeged.hu
ABSTRACT
The underlying cause of Alzheimer’s disease (AD) is thought to be the ß-amyloid aggregates formed mainly by Aß1–42 peptide. Protective pentapeptides [e.g., Leu-Pro-Phe-Phe-Asp (LPFFD)] have been shown to prevent neuronal toxicity of Aß1–42 by arresting and reversing fibril formation. Here we report that an endogenous tetrapeptide, endomorphin-2 (End-2, amino acid sequence: YPFF), defends against Aß1–42 induced neuromodulatory effects at the cellular level. Although End-2 does not interfere with the kinetics of Aß fibrillogenesis according to transmission electron microscopic studies and quasielastic light scattering measurements, it binds to Aß1–42 during aggregation, as revealed by tritium-labeled End-2 binding assay and circular dichroism measurements. The tetrapeptide attenuates the inhibitory effect on cellular redox activity of Aß1–42 in a dose-dependent manner, as measured by 3-(4,5-dimethylthiazolyl-2)-2,-5-diphenyltetrazolium bromide (MTT) assay. In vitro and in vivo electrophysiological experiments show that End-2 also protects against the field excitatory postsynaptic potential attenuating and the NMDA-evoked response-enhancing effect of Aß1–42. Studies using [D-Ala (2)
, N-Me-Phe (4)
, Gly (5)
-ol]-enkephalin (DAMGO), a µ-opioid receptor agonist, show that the protective effects of the tetrapeptide are not µ-receptor modulated. The endogenous tetrapeptide End-2 may serve as a lead compound for the drug development in the treatment of AD.—Szegedi, V., Juhász, G., Rózsa, E., Juhász-Vedres, G., Datki, Z., Fülöp, L., Bozsó, Z., Lakatos, A., Laczkó, I., Farkas, T., Kis, Z., Tóth, G., Soós, K., Zarándi, M., Budai, D., Toldi, J., Penke, B. Endomorphin-2, an endogenous tetrapeptide, protects against Aß1–42 in vitro and in vivo.
Key Words: Alzheimer’s disease • fibrillogenesis • circular dichroism • single-unit • transmission electronmicroscopy • rat
ALZHEIMER’S DISEASE (AD) is the leading cause of senile dementia (1)
. The disease is histopathologically characterized by the deposition of a mainly 42 amino acid residue peptide (Aß1–42) in senile plaques in the brain, a massive neurite dystrophy, and neuronal loss chiefly in the basal forebrain and hippocampus (2)
. The underlying cause of the disease seems to be the extracellular accumulation and aggregation of amyloid beta peptides, encoded by a gene for the much larger amyloid precursor protein (APP) (3
, 4)
. Aß deposition is an invariable feature of the disease, and the heritable form of AD is mainly linked with mutations in the APP-related genes (5)
. Aggregated Aß1–42 has been shown to exert toxic and disruptive effect on neurons in vitro (6
, 7)
and in vivo (8
, 9)
. Impaired cellular ion homeostasis (10
, 11)
, Ca2+ influx (12
13
14)
, oxidative stress (15)
, and NMDA receptor enhancement (16
, 17)
has been reported after Aß1–42 application. Additionally, hippocampal and cortical excitatory postsynaptic potential (EPSP; refs 17
, 18
) and long-term potentiation (LTP; ref 19
) are attenuated both in vivo and in vitro by Aß1–42. The toxicity of Aß peptides proved to correlate well with their aggregation properties (20)
; therefore, a promising approach in the treatment of AD would be the inhibition of amyloid aggregation (21)
. The introduction of so called beta-sheet-breaker (BSB) peptides (22
23
24)
allowed the design of rational, putative protective peptides. These small molecules arrest Aß fibrillogenesis and/or amyloid neurotoxicity in vitro, but only a few compounds have been reported to be active in vivo (17)
. One such compound is Leu-Pro-Phe-Phe-Asp (LPFFD), which after in vitro tests, successfully reduced amyloid load and cerebral damage in transgenic animals (25)
.
A recently isolated endogenous tetrapeptide, endomorphin-2 (End-2, YPFF, Tyr-Pro-Phe-Phe; ref 26
), shows high structural similarity with LPFFD: the C-terminal tripeptide sequence (PFF) of End-2 perfectly matches with the middle tripeptide sequence of LPFFD. In addition, the high End-2 containing central nervous system (CNS) areas (27
, 28)
seem to be unaffected both in human AD and in transgenic animals (29
30
31)
. Those areas that suffer massive cell loss in AD, including the hippocampal formation and neocortex, show no or little End-2 like immunoreactivity. Therefore, it is intriguing to suppose that this endogenous tetrapeptide might have some protective action against Aß1–42.
We have used several independent methods to evaluate the putative neuroprotective effect and the mechanism of End-2 against aggregated Aß1–42. End-2 and Aß1–42 interaction was measured by tritium-labeled End-2 binding assay. Transmission electron microscopy (TEM), quasielastic light-scattering (QLS), and circular dichroism (CD) were used to elucidate whether the tetrapeptide has an antifibrillogenic effect. Neuroprotection against the cellular redox activity attenuation and neuromodulatory effects of Aß1–42 were tested by an in vitro cell viability assay and by in vitro and in vivo electrophysiology.
End-2 possesses some antioxidant effect alone (32)
and protective action through the activation of the µ-opioid receptor (MOR; refs 33
, 34
); therefore, we used a synthetic peptidomimetic agonist of the same receptor, [D-Ala (2)
, N-Me-Phe (4)
, Gly (5)
-ol]-enkephalin (DAMGO), as a control substance in the biological measurements. Here, we report that End-2, an endogenous tetrapeptide has a MOR-independent protective action against the neuromodulatory effects of Aß1–42 both in vitro and in vivo; however, it has no antifibrillogenic action.
MATERIALS AND METHODS
Peptide synthesis
Aß1–42 was synthesized in our laboratory by a solid-phase procedure involving the use of Wang-resin and Fmoc chemistry. The peptide was purified on a C-4 RP-HPLC column with an acetonitrile gradient; pure fractions were pooled and lyophilized. Aß1–42 used for biological experiments was repeatedly lyophilized from aqueous solution resulting in assemblies with higher aggregation rate. Purity control and proof of structure were achieved by amino acid analysis and mass spectrometry (ESI MS, FinniganMat TSQ 7000). End-2 and H-Tyr-[3H]Pro-Phe-Phe-NH2 radioligand ([3H]End-2) were also prepared in our laboratory using End-2 containing dehydro proline in position 2 and tritium gas in the presence of Pdo/BaSO4. DAMGO was purchased from BACHEM (Budendorf, Switzerland).
TEM
To destroy the preformed aggregates, Aß1–42 was dissolved in HFIP and incubated for 24 h at ambient temperature. Then the organic solvent was removed in vacuo. Solutions of Aß1–42 (100 µM), either alone or coincubated with 500 µM End-2, were prepared by dissolving the peptide(s) in deionized water. Dissolution was promoted by constant agitation of the solution for 2 min with an automatic pipette, followed by a sonication for 10 min. To initiate and facilitate the nucleation-dependent aggregation of Aß1–42, peptide samples were seeded with a definite volume of preaggregated Aß1–42 solution (c=0.5 mg/ml, 10 µl were added to 1 ml sample). Electron micrographs were taken from the freshly prepared peptide solutions (0 h) and after incubation for 24 and 120 h at 37°C. Ten microliter droplets of solutions were placed on 400 mesh carbon-coated copper grids (Electron Microscopy Sciences, Washington, PA), incubated for 2 min, fixed with 0.5% (v/v) glutaraldehyde solution, washed three times with deionized water, and finally stained with 2% (w/v) uranyl acetate. Specimens were studied with a Philips CM 10 transmission electron microscope (FEI Company, Hillsboro, OR) operating at 100 kV. Images were taken by a Megaview II Soft Imaging System routinely at magnifications of x25,000, x46,000, and x64,000 and analyzed by an AnalySis 3.2 software package (Soft Imaging System, Münster, Germany).
QLS measurements
All experiments were performed at 25°C with a Malvern Zetasizer Nano ZS Instrument (Malvern Instruments, Worcestershire, UK) equipped with a He-Ne laser (633 nM), applying the Non-Invasive Back Scatter (NIBS) technology, which means detection of the scattered light at an angle of 173°. The translational diffusion coefficients were obtained from the measured autocorrelation functions using the regularization algorithm CONTIN built in the software package Dispersion Technology Software 4.0 (Malvern Instruments). With the assumption of the scattering particles as hard spheres, their apparent hydrodynamic radius can be calculated from the diffusion parameters by using the Stokes-Einstein equation Rh= kBT/(6
DT), in which kB is the Boltzmann constant, T is the absolute temperature, is the viscosity of the medium, and DT is the translational diffusion coefficient. Samples were prepared as given above, and the aqueous ones were filtered before incubation through a 0.45 µm PVDF sterile membrane filter (Roth, Karlsruhe, Germany) to remove the possibly interfering dust particles. Correlation function and distribution of the apparent hydrodynamic radii over the scattered intensity of the samples were monitored for 2 days.
Radioactive binding assay
Twice lyophilized Aß1–42 (1.0 mg/ml) was aged in a pH 7.4 buffer [Tris (56 mM), NaCl (112 mM), EGTA (1.12 mM), and MgCl2 (3.36 mM)] for a week. Aß1–42 suspension was diluted to 0.2 mg/ml before use. The assays were performed in 96-well plates in which the wells contained 0.01 mg/ml Aß1–42; 8 x 10–8 [3H]End-2 and different concentrations of nonlabeled End-2. After 60 min incubation at room temperature, the content of the plate was transferred to a Multiscreen HTS filtration plate (Millipore Corporation, Bedford, MA). After filtration, the wells were washed four times with 200 µl buffer and dried, and 75 µl scintillation cocktail (OptiPhase Supermix, Perkin-Elmer) were added into each well. Radioactivity was measured in a scintillation counter. Experiments were performed in duplicates.
CD measurements
CD measurements were performed at 25°C on a Jobin-Yvon Mark VI dichrograph using a 1 mm pathlength quartz cuvette. All spectra were the average of four scans, the resolution was 0.2 nM. Aß1–42 peptide was dissolved in tridistilled water (pH 
5) at a concentration of 0.2 mg/ml (44 µM) and sonicated for 5 min. After sonication, the samples were immediately measured (0 h) and then after 1, 3, 6, 24, and 48 h. The samples were incubated during the aging period at 25°C. The band intensities were expressed as mean residue ellipticity, [
]MR, and given in units of deg · cm–2 · dmol–1 using a mean residual wt 114.
3-(4,5-Dimethylthiazolyl-2)-2,-5-diphenyltetrazolium bromide assay for cell viability
The standard method of (3-(4,5-dimethylthiazolyl-2)-2,-5-diphenyltetrazolium bromide) (MTT) assay modified by Datki et al. (7)
was used to measure the effect of End-2 and DAMGO against Aß1–42 in a dose-dependent manner, using differentiated SH-SY5Y cells. Cells were incubated with the following peptides: 1) 10 µM aggregated Aß1–42; 2) 1, 10, 100, and 200 µM End-2/DAMGO; and 3) a mixture of 10 µM aggregated Aß1–42 and 1, 10, 100, and 200 µM End-2/DAMGO in a 24-well plate for 24 h. One-hundred milliliters of MTT stock solution (4 mg/ml) were added to each well, containing 1 ml medium, and the mixture was incubated for 3 h. The MTT solution was carefully decanted off, and formazan was extracted from the cells with 1 ml of a DMSO/EtOH (4:1) mixture in each well. The formazan color intensity was measured with a 96-well ELISA plate reader at 550 nM with the reference filter set to 620 nM. Experiments were done in triplicates, with seven measurements within each trial (n=21).
In vitro electrophysiology
Fifty-six young (postnatal days 20–30) Wistar rats were decapitated, and 400 µm thick coronal slices were prepared from their primary motor cortices. Slices were placed in a recording chamber and were perfused continuously with oxygenated artificial cerebrospinal fluid (ACSF) at 34°C. Bipolar stainless-steel microelectrodes were used to stimulate the layer II/III horizontal pathways. The stimulus intensity was adjusted between 20 and 70 µA to evoke the half-maximum response. Glass micropipettes used for recordings were filled with ACSF and broken off so that the impedance was between 1.0 and 1.5 M
. The recordings were carried out at 300–500 µm from the stimulation.
Application of peptides
DAMGO or End-2 was applied either alone (5x10–5 M) or in a mixture with Aß1–42 at a molar ratio of 5:1. Aß1–42 was used alone in a concentration of 10–5 M. Fifteen microliters of solution were applied in the close vicinity of the recording electrode with a Hamilton syringe equipped with a micromanipulator. The signals were amplified (SEC-LX05, npi), filtered (1–3 kHz), acquired at a 10 kHz sampling rate on a pClamp8 system and Digidata 1320 A/D board (Axon Instruments), and analyzed off-line with Origin 6.0 software (Microcal Software). The averaged amplitude (n=4) of the initial negative component of the field excitatory postsynaptic potentials (fEPSPs) evoked by stimuli was used as a measure of the population excitatory synaptic current.
In vivo electrophysiology
Extracellular single-unit recordings were made in 41 chloral hydrate-anesthetized (4 g/kg initial dose, ip, supplemental doses as required) male Wistar rats weighing 300–360 g. The head of the animal was mounted in a stereotactic frame, the skull was opened above the hippocampus (antero-posterior: –2.8 to –3.8 from bregma; lateral: 2 mm on either side from the midline), and the dura mater was carefully removed. Structures were localized according to the stereotactic atlas of Paxinos and Watson (1986). The location of the electrode was verified by iontophoretic Pontamine sky blue (4% in 100 mM sodium acetate) ejection at –2 µA for 20 min followed by conventional histology using neutral red contrastaining.
Extracellular recordings and iontophoresis
Single unit activity was extracellularly recorded by means of a low impedance (<1 M) 7 µm carbon fiber-containing microelectrode (Kation Scientific, Minneapolis, MN) from the hippocampus between the depths of 2–3 mM. Drugs were delivered from the surrounding outer barrels. The action potentials were amplified, filtered (ExAmp-20K, Kation Scientific; ref 35
), and monitored with an oscilloscope. A window discriminator was used for discriminating the spikes. The amplified signals were sampled and digitalized at 50 kHz frequency. The number of action potentials per second was counted by the computer, and peristimulus time histograms were calculated, displayed in line, and digitally stored for off-line analysis. Iontophoretic drug delivery and collection of experimental data were performed by a multifunction instrument control and data acquisition board (National Instruments PCI-1200) placed in a computer along with Union-36 iontophoretic pumps (Kation Scientific).
The following solutions were ejected: 1) 100 mM NMDA Na (Sigma-Aldrich, Budapest, Hungary) in 100 mM NaCl (pH 8.0); 2) 5 x 10–5 M Aß1–42; 3) 2.5 x 10–4 M DAMGO/End-2 dissolved in saline (pH 6.4); and 4) a mixture containing Aß1–42 (5x10–5 M) and DAMGO/End-2 (2.5x10–4 M) stored for 24 h at 4°C. To lessen the degree of aggregation, the Aß1–42 and the mixture solutions were sonicated (Merck Eurolab 120 W apparatus) for 15 min before use.
Neurons were excited by repetitive iontophoresis of NMDA at 1 min intervals for 5 s by applying negative iontophoretic currents ranging from 11 to 100 nA, and the ejection current was selected so that the maximum firing rate fell between 30 and 80 spikes/s. Retaining current of the opposite direction in the range 2–21 nA was used. The peristimulus time histograms of the neurons were recorded. After establishment of a stable control (at least 4 successive peaks), the solution to be examined was ejected: DAMGO/End-2 containing solution for 3 min at +100 nA; Aß1–42 and the mixtures for 1 min at –0.5 µA.
Data analysis
The results of the MTT assay were analyzed by using ANOVA with post hoc Bonferroni test (P<0.01). For in vitro electrophysiological studies, statistical significance was determined by means of Student’s t test. In all cases, statistical significance was set at P 
0.05. Statistical evaluations of the in vivo results were performed by using the total number of spikes evoked during each epoch of excitation by iontophoretic application of the excitatory amino acid NMDA. The background neuronal discharge was calculated by averaging a 15 s period of ongoing activity preceding each epoch of excitation, and this value was subtracted from all evoked responses. The total spike number during each epoch of excitation was calculated and expressed as a percentage of the mean (±SE) and compared statistically with the data obtained after Aß1–42 application by using one-way ANOVA (with the Bonferroni test for post hoc analysis). A P value of at least 0.05 was considered significant in all cases.
RESULTS
TEM
The aggregation process of Aß1–42 was followed with TEM during 5 days. The putative ß-sheet-breaker effect of End-2 was also studied coincubating Aß1–42 with the tetrapeptide. A few protofibrils and spherical oligomers could be observed at 0 h in both solutions (Fig. 1
, A and B, top). After 1 day of incubation, both samples contained long, mature, smooth amyloid fibrils emanating from the seeding centers without any observable difference between the two time-matched different samples (Fig. 1, A
and B, middle). Although the concentration of fibrils increased further after a 5 day incubation period, the End-2 containing solution resembled the control one: the tetrapeptide did not affect the fibrillogenesis of Aß1–42 (Fig. 1, A
and B, bottom).
|
QLS measurements
QLS measurements were performed to characterize the aggregation altering effect of End-2 compared with the reference peptide LPFFD. Figure 2
represents the distribution curves of Rh values over the scattered intensity in these experiments. According to our results, Aß1–42 can be characterized with a bimodal curve having the maxima at Rh = 91 and 684 nm. LPFFD showed the expected BSB effect on the aggregation course: beside the two maxima falling into either the protofibrillar (81 nm) or the fibrillar size range (549 nm), a new peak appeared representing the small oligomers (Rh=20 nm). The ratio of the integrated intensities is 1:3.9:13.2 (oligomers/protofibrils/fibrils, respectively). Because the scattering intensity of a particle is proportional to the sixth power of its diameter, it can be assumed that the sample contains predominantly small oligomers together with the fibrillar aggregates. On the contrary, End-2 was not capable of changing the aggregation profile of Aß1–42 considerably; in the case of coincubation with End-2, its intensity distribution curve possessed the same bimodal characteristic (maxima at 157 and 759 nm) as without the tetrapeptide in the control sample.
|
Radioligand binding studies
To demonstrate the actual binding of End-2 to Aß1–42, we have used a radioligand binding assay with [3H]-labeled End-2. The amount of [3H]End-2 bound to aggregated Aß1–42 was decreased in a concentration-dependent manner by the addition of unlabeled End-2, indicating that the binding was competitive. However, the steady-state condition was not reached even after 7 days of incubation. The possible cause for that could be the ongoing fibrillogenesis of Aß1–42.
CD measurements
The CD spectra of Aß1–42 alone and in the presence of End-2 at 1:1 M ratio in the function of aging time are shown in Fig. 3
, B–D. The shape of the CD curve of Aß1–42 (Fig. 3B
) at t = 0 time reflects the mixture of various conformations: the broad negative band is contributed by the n* transition (215 nm) of ß-sheet conformation and the class C spectrum of ß-turns type I/III or 310 helix (36)
. The positive band, which is expected to appear below 200 nm in the case of both ß-sheet (strand) or ß-turn structures, is suppressed by the intensive negative band of the unordered (random coil) conformation. Indeed, at t = 0, the low (negative) values of the spectrum here are indicative of the presence of a considerable amount of unordered secondary structures. As the peptide ages, a gradual spectral change can be observed: the intensity of both the positive and negative band increases reaching their maximal values after 48 h incubation. This reflects a conformational transition from a random-coil-rich form to a ß-sheet-rich form. In the presence of End-2, the shape of the spectrum measured immediately after dissolution (at 0 h) is different (Fig. 3C
): the appearance of a positive band at 200 nm is due to the positive band of End-2 in this spectral range. The strong interaction between Aß1–42 and End-2 (at 0 h) is reflected by the considerable difference between the measured spectrum of the complex and the sum spectrum of the free Aß1–42 and free End-2 (Fig. 3D
). This is also supported by the smaller increase of the [] value at 200 nm of the complex than that of the free Aß1–42 in the aging period of 48 h (for Aß1–42–End-2 complex, [+2110]0 h and [+5290]48 h and for Aß1–42, [-440]0 h and [+3985]48 h). The CD spectra of the mixture of Aß1–42 and End-2 after standing for 1–3 h are shown in Fig. 3C
: the intensity of the positive band at 200 nm is increasing, reflecting a higher amount of ß-sheet content. Practically no further intensity increase can be observed between 3 and 24 h. A typical ß-sheet spectrum appears only after 48 h incubation; all the other spectra reflect the presence of a small amount turn conformation in addition to ß-sheet.
|
MTT cell viability assay
To elucidate the putative neuroprotective action of End-2 against Aß, MTT assay was used. Both End-2 and DAMGO were applied in 1, 10, 100 and 200 µM either alone or in a mixture with 10–5 M Aß1–42. Application of 10–5 M Aß1–42 resulted in a decrease of cell viability. Reduction of MTT into formazan in the SH-SY5Y cell line decreased to 58 ± 8% (n=21), as described previously (17
, 37)
. DAMGO, applied in gradually increasing concentration, was unable to interfere with Aß1–42 induced decrease of MTT reduction. In contrast, End-2 could significantly attenuate the inhibition of MTT reduction of Aß1–42 in a dose-dependent manner. The effect was significant when the tetrapeptide was applied in 100 and in 200 µM concentration with 10–5 M Aß1–42 (Fig. 4
, 77±7 and 79±10%, respectively, P<0.01, n=21). DAMGO and End-2 did not affect cell viability alone at the concentrations used (data not shown).
|
In vitro electrophysiology: fEPSP recordings
The amplitude of the initial negative component of the fEPSP evoked by stimuli applied through horizontally displaced electrodes in layers II/III was recorded. Aß1–42 at a concentration of 10–5 M decreased the amplitude of the fEPSPs to 58.9 ± 10% (P<0.05, n=15; Fig. 3
). Application of End-2 in a concentration of 5 x 10–5 M did not change the amplitudes (100.3±17.6%, n=13). The Aß1–42 and End-2 (1:5 M ratio) mixture application resulted only in a slight decrease of the fEPSP amplitude: the value of 91.7 ± 11.4% (n=14) did not differ significantly from the control concentration; 5 x 10–5 M DAMGO did not alter the measured fEPSPs (92.8±16%, n=6). In contrast, application of Aß1–42 and DAMGO mixture in 1:5 M ratio resulted in significant fEPSP amplitude attenuation (57±14%, n=8, P<0.05).
In vivo iontophoresis and single-unit activity: modulation of the NMDA-evoked neuronal firing
Effects of End-2 and DAMGO
End-2 alone was ejected for 3 min after establishing a stable evoked neuronal firing (control). The NMDA-evoked responses increased up to a maximum of 162 ± 14% (n=9, P<0.05), which proved to be significant compared to the control response concentration (Fig. 4B
and Fig. 5
). Response enhancement lasted for 18–20 min, and then the concentration of triggered neuronal firing returned to control concentration. DAMGO application for 3 min resulted in similar significant response enhancement to 167 ± 5% (n=5, P<0.05). However, the evoked responses did not return to the control concentration within the time frame of the experiment, and the enhancement proved to be long lasting. Application of the vehicle had no effect on the neuronal firing (n=6, data not shown).
|
Effect of aggregated Aß1–42
After Aß1–42 application, the NMDA-evoked responses (frequency of spikes) significantly enhanced (Fig. 4A
) to 269 ± 16% (n=14, P<0.05) as compared with the preAß1–42 control. The enhancement gradually increased up to a maximum value and did not return to the control concentration within the 40-min time frame of the experiments, as described by Molnar et al. (16)
and Szegedi et al. (17)
. In contrast, iontophoresis of the saline without peptide caused no change in the baseline activity or in the NMDA-evoked firing rate (n=6, data not shown).
Effect of mixtures containing Aß1–42 and End-2 or Aß1–42 and DAMGO
The peptide mixtures were administered iontophoretically for 1 min at –0.5 µA. In that way, only the negatively charged Aß1–42 was ejected; End-2/DAMGO could not leave the micropipette, unless absorbed on the surface of different Aß1–42 assemblies. Application of the End-2 and Aß1–42 mixture did not result in enhancement of the NMDA responses (Fig. 4D
): the concentration of neuronal firing remained at 115 ± 21% (n=6; Fig. 5
). In contrast, ejection of the DAMGO and Aß1–42 containing solution enhanced NMDA-evoked firing. The maximum response to 269 ± 37% remained, and it did not return to control concentration (n=6, P<0.05).
Coiontophoresis of Aß1–42 and End-2 or Aß1–42 and DAMGO
To elucidate the protective mechanism of End-2 against the effect of Aß1–42, coiontophoretic administration was used. End-2 and DAMGO were ejected right before Aß1–42 application with the same current and time frame as described above to ensure the effective concentration in the close vicinity of the neuron. In that way, MORs will be activated, and the drugs will not exert their putative protective effect on the surface of aggregated Aß1–42 species. Immediately after ejection of End-2, Aß1–42 was applied. The maximum NMDA-evoked response reached 348 ± 65%, and the enhancement remained during the measurements (n=7, P<0.05; Figs. 4C
and 5
). Coiontophoresis of DAMGO with Aß1–42 resulted in similar response enhancement. The maximum evoked response was 238 ± 37%, and the effect did not diminish during the time frame of the experiment (n=6, P<0.05).
DISCUSSION
Our results show that End-2 possesses protective properties against Aß1–42 induced attenuation of cell redox activity and neuromodulation in vitro and in vivo. End-2 was able to attenuate the decrease of MTT reduction by ß-amyloid in an in vitro cell viability test in a dose-dependent manner (Fig. 4)
. Inhibition of MTT reduction by Aß is not an indicator of actual cell death (38
, 39)
. Although the decrease of cellular redox activity after Aß application is not due to massive cell loss, the MTT test is a widely used assay for indicating the overall metabolic rate of cultured neurons. Synaptic changes emerge well before actual neuronal death in early AD (40
41
42)
. Because inhibition of MTT reduction by Aß is supposed to be due to changes in membrane properties (43)
and exocytosis (44)
, which can significantly affect normal synaptic working, the MTT test with Aß may be exploited as an indicator of possible synaptotoxicity. In accord to this phenomenon, End-2 offered protection against the fEPSP attenuation caused by Aß1–42 in in vitro brain slice experiments (Fig. 5)
. The amyloid-induced enhancements of NMDA-evoked responses were also diminished by End-2 in vivo (Figs. 6
and 7
). These protective effects were not mediated through the activation of MOR, because DAMGO, a synthetic MOR agonist was not able to attenuate the Aß1–42 mediated effects. Results of the two kinds of in vivo iontophoretic experiments ("mixture" and "coiontophoresis") provided additional evidence that the activation of MORs does not yield protection against Aß1–42. Coiontophoretic application of Aß1–42 with either End-2 or DAMGO resulted in NMDA-induced response enhancement, similar to the effect of ß-amyloid. However, when Aß1–42 was preincubated with End-2, no ß-amyloid induced enhancement could be measured (Figs. 6
and 7)
. In coiontophoretic experiments, the ejection of MOR agonists preceded Aß1–42 application to activate the µ-receptors. In contrast, at application of the mixture solutions, the pH values and the ejection polarities were chosen in that way that only Aß1–42 and the possibly bound End-2 and DAMGO could leave the capillary. Under these conditions, presumably no significant MOR activation occurred due to the ejection of the mixtures.
|
|
End-2 does not arrest fibril formation or disassemble preformed aggregates, as revealed by TEM and QLS measurements (Figs. 1
and 2)
. The particle size distribution of End-2 incubated Aß1–42 resembles that of Aß1–42 alone after 48 h, indicating that there was no interference with aggregation kinetics. In contrast, BSB compounds, such as LPFFD, a pentapeptide that is capable of arresting and reversing fibril formation (25
, 45)
, shifted the particle size distribution toward small aggregates and dimers (Fig. 2
, and, i.e., ref 24
). However, End-2 could obviously bind to Aß1–42, as seen by radioligand binding studies and CD measurements. The binding is reversible, as End-2 could replace [3H]End-2 (Fig. 3)
. In addition, data from the CD measurements suggest that Aß1–42 and End-2 formed a complex in the initial state of the aggregation process.
It is controversial whether all stages of aggregated Aß1–42 exert action leading to neural compromise. Previous studies provided evidence that the small-size aggregates (protofibrils, oligomers) are likely to affect neurons (46
, 47)
, and fibrillar species may trigger glial cells to produce toxic mediators (e.g, 48
, 49
). The applied Aß1–42 in our biological measurements contained a wide range of different sized aggregated particles; however, End-2 was capable of inhibiting their biological effects. Wogulis et al. (38)
suggested that the process of fibril formation causes the neuronal damage. Fibrillogenesis was evident under our experimental conditions; nevertheless, Aß1–42 induced changes were diminished in the presence of End-2; therefore, the tetrapeptide may have protected against this possible phenomenon as well.
Recent reports indicated that not all types of aggregated assemblies are neurotoxic (38
, 50)
or different conformational states may have different biological activity (51)
. End-2 might be bound to the biological harmful assemblies, transforming their conformation into a harmless structure, or by inhibiting the interaction between cells and Aß aggregates. By doing so, the tetrapeptide may offer protection against assemblies of different aggregation states. The novel mechanism of protection prones the tetrapeptide to be a very promising candidate for future research in the combat against AD. A recent report (52)
underlies the risk of approaches aimed at "bustering" or inhibiting the formation of Aß fibrils. Because oligomers are supposed to have greater inherent toxicity than fibrils, arresting or reversing fibril formation may exaggerate brain damage in AD.
Being an endogenous substance, End-2 might play a role in the actual AD progression and possibly assent to the area selective cell loss seen in the disease. CNS areas with high End-2 content were found to be unaffected by Aß-induced damage (29
30
31)
. Those areas that suffer cell loss in AD, including the hippocampal formation and neocortex, show no or little End-2 like immunoreactivity. Only a few endogenous molecules have been reported or suggested to have protective effect against ß-amyloid (i.e., taurine (53)
and kynurenic acid (54)
), which target the excitotoxic imbalance either by GABAA receptor activation or by NMDA receptor inhibition. Shifting the net concentration of neuronal excitability toward inhibition attenuates Ca2+ entry. Decreased intracellular Ca2+ concentration provides less probability of apoptotic cascade initiation. Another endogenous compound, melatonin, has a more complex neuroprotective mechanism. In addition to its antioxidant properties, it also binds to the GABAA receptor (53)
. Additionally, melatonin blocks fibrillogenesis of Aß1–42 in vitro (24)
.
The effect of MOR activation on the in vivo NMDA-evoked neuronal firing indicated that the action potentials of pyramidal cells were recorded (Figs. 6
and 7)
. The distribution of MOR in the hippocampus displays cell type specificity: only the GABA-ergic interneurons express this type of inhibitory, metabotropic receptor (55
, 56)
. Therefore, excitation after End-2 and DAMGO ejection indicates a disinhibitory mechanism (57)
: µ-receptor bearing interneurons might be inhibited by agonists, resulting in an attenuation of the inhibitory input to the pyramidal cells, increasing their excitability.
Approaches aimed to increase End-2 concentration in the brain presumably would not offer rational treatment of AD patients. The activation of MORs increases the net excitability in the CA1 region and may result in focal epileptic seizures (57)
. Consequently, in the treatment of AD, the sequence of End-2 is what may be exploited as a lead sequence for development of peptidomimetic drugs that will not activate µ-opioid receptors.
ACKNOWLEDGMENTS
This work was supported by the National Bureau of Research and Development (NKTH RET 08/2004), OTKA TS 049817, NKFP 1A/005/2004, and a János Bolyai Research Fellowship (Dr. Tamás Farkas). Multibarrel electrodes and technical background were provided by Kation Europe Bt.
Received for publication August 26, 2005. Accepted for publication January 31, 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
