| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Neuropharmacology Section,
Laboratory of Structural Biology,
Chemistry Section, Laboratory of Pharmacology and Chemistry, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, USA;
Department of Bioscience and Bioengineering, Dalian University of Technology, Dalian, P.R. China; and
|| Department of Neurology, First Clinical Hospital,
Department of Physiology, Dalian Medical University, Dalian, China
1Correspondence: MD F1-01 NIEHS, P.O. Box 12233, Research Triangle Park, NC 27709, USA. E-mail: Hong3{at}niehs.nih.gov
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: dynorphin pepide • ROS • GGF • Parkinson’s disease
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. Although the concept is intriguing to scientists, pharmaceutical companies, and clinicians, the acceptance of biologically active compounds in the femtomolar dose range has remained controversial due to the scarcity of defined mechanisms and the lack of plausible, putative high-affinity receptors (4
5
6)
. The purpose of this study was to identify two neuroprotective compounds with a similar bimodal dose response, to discern whether the neuroprotective characteristics of each compound could converge into a single mechanism, and through this process, to elucidate a common femtomolar site of action.
Parkinson’s disease (PD) is characterized by the specific and progressive death of dopaminergic neurons in the substantia nigra (SN); other neuronal cell types are much less affected. Recent reports have linked inflammation to neurodegenerative disease, where microglia, cells of myeloid lineage responsible for innate immunity in the brain, are considered to be the major cell type underlying the inflammation-mediated neurotoxicity (7
8
9)
. The activation of microglia is a complex process involving the release of several soluble proinflammatory factors [tumor necrosis factor 
(TNF-), PGE2, IL-1] and free radicals (nitric oxide, superoxide) (7)
. Current replacement therapy with L-dopa is able to alleviate disease symptoms, but is unable to alter the disease course. Thus, therapeutic interventions designed to inhibit the microglial inflammatory response offer hope for attenuation of the neurodegenerative disease process. The current anti-inflammatory treatments available, including steroids and nonsteroidal anti-inflammatory drugs, are limited by the ability to influence only a small portion of the microglial response (10)
. Thus, identification of compounds acting on novel targets to inhibit the release of a wide range of proinflammatory factors from overactivated microglia is of paramount importance. In the ensuing study, we report that femtomolar concentrations of naloxone and the peptide fragment glycine-glycine-phenylalanine (GGF) attenuate a broad spectrum of the microglia inflammatory response (reactive oxygen species (ROS) and proinflammatory factors) and are neuroprotective with extremely potent efficacy through the inhibition of microglial NADPH oxidase.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents
WST-1 was purchased from Dojindo Laboratories (Gaithersburg, MD, USA). Cell culture ingredients were obtained from Life Technologies (Grand Island, NY, USA). The polyclonal antibody against tyrosine hydroxylase (TH) was a kind gift from Dr. John Reinhard of GlaxoSmithKline (Research Triangle Park, NC, USA). Lipopolysaccharide (strain O111:B4) was purchased from Calbiochem (San Diego, CA, USA). The Biotinylated horse anti-mouse and goat anti-rabbit secondary antibodies were purchased from Vector Laboratories (Burlingame, CA, USA). [3H] Dopamine (DA, 28 Ci/mmol) was purchased from NEN Life Science (Boston, MA, USA). 2', 7'-Dichlorofluorescin diacetate (DCFH-DA) was obtained from Calbiochem (La Jolla, CA, USA). TNF- ELISA kits were purchased from R&D Systems Inc. (Minneapolis, MN, USA).
Mesencephalic neuron-glia cultures
The rat and mouse ventral mesencephalic neuron-glia cultures were prepared following a described protocol (11)
. Cultures were used for treatment 7 days after seeding.
Microglia-enriched cultures
Primary enriched-microglia cultures were prepared from the whole brains of 1-day-old Fisher 344 rat pups, using a described procedure (12)
. Cells were treated 24 h after seeding the microglia.
DA uptake assay
The ability of cells to take up [3H] DA was performed using the described protocol (11)
. At 7 days postseeding, mesencephalic neuron-glia cultures were pretreated with MEM medium containing either dynorphin peptides, naloxone, or control treatment medium for 30 min, then exposed to MEM containing 10 ng/mL lipopolysaccharide (LPS). Neurotoxicity was measured at 7 days after treatment. Cells were incubated for 20 min at 37°C with 1 µM [3H]-DA in Krebs-Ringer buffer (16 mM NaH2PO4, 16 mM Na2HPO4, 119 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO4, 1.3 mM EDTA, pH 7.4). Nonspecific uptake was measured in the presence of 10 µM mazindol. After the cells were washed (3x) with ice-cold Krebs-Ringer buffer (1 mL/well) and lysed with 1 N NaOH (0.5 mL/well), the lysate was mixed with 15 mL of scintillation fluid and radioactivity was determined with a liquid scintillation counter. The specific [3H] DA was calculated by subtracting the amount of radioactivity obtained in the presence of mazindol from that obtained in the absence of mazindol.
Immunostaining
At 7 days postseeding, mesencephalic neuron-glia cultures were pretreated with a dynorphin peptide or naloxone, then exposed to 10 ng/mL LPS. Neurotoxicity was measured at 7 days after treatment. Dopamine neurons were stained with the polyclonal antibody against TH in the manner described previously (13)
. To quantify cell numbers, nine representative areas/well in the 24-well plate were counted under the microscope at 100x magnification by three individuals in blind. The average of these scores was reported.
Superoxide assay
Extracellular superoxide (O2–) production from microglia was determined as reported (14)
by measuring the superoxide dismutase (SOD) inhibitable reduction of tetrazolium salt, WST-1 (15
16
17)
. Primary rat microglia enriched cultures were plated at 5 x 104/well in 200 µL culture medium in 96-well plates, and incubated for 24 h at 37°C in a humidified atmosphere of 5% CO2 and 95% air. Cells were washed twice with Hanks’ balanced salt solution (HBSS) and pretreated with GGF or naloxone for 30 min before adding 10 ng/mL of LPS. To each well, 50 µL of peptide or naloxone, 50 µL of HBSS with or without SOD (600 U/mL), 50 µL of vehicle or LPS, and 50 µL of WST-1 (1 mM) in HBSS were added. The cultures were incubated for 30 min at 37°C and 5% CO2 and 95% air. The absorbance at 450 nm was read with a Spectra Max Plus microtiter plate spectrophotometer (Molecular Devices, Sunnyvale, CA, USA). The amount of SOD-inhibitable superoxide was calculated and expressed a percentage of vehicle-treated control cultures.
Intracellular ROS assay
The production of intracellular ROS was measured by DCFH oxidation as described (13)
. For this assay, 10 mM DCFH-DA was dissolved in methanol and diluted 500-fold in HBSS without serum or other additives to give a 20 µM concentration of DCFH-DA, as earlier reported (13
, 18)
. Enriched-microglia cultures seeded (5x104) in 96-well plates were washed twice with HBSS, then exposed to DCFH-DA for 1 h. After DCFH-DA loading, cells were then pretreated with HBSS containing either GGF or naloxone for 30 min before adding 10 ng/mL of LPS. After LPS treatment, cells were incubated at 37°C and 5% CO2 and 95% air for 2 h. After incubation, the fluorescence was read at the 485 nm excitation and 530 nm emission on a fluorescence plate reader. Increases in intracellular ROS are expressed as the difference of the treatment group from the control group.
TNF- assay
The production of TNF- was measured with a commercial enzyme-linked immunosorbent assay (ELISA) kit from R&D Systems, as described (12)
.
Pharmacophore analysis
The Accelrys DS MedChem Explorer version 2.1 software was used for pharmacophore analysis. All the molecules were built within the Accelrys DS Modeling software. The peptides were constructed as linear chains. Conformational analysis was performed within the DS MedChem Explorer software using the Catalyst ConFirm algorithm. The maximum number of conformers generated for each molecule was 250. Conformers with relative energies >20 kcal/mol were removed. The best (most rigorous and time-consuming) conformer generation method was used. Eight chemical features were included in the development of the pharmacophore model: HBA (hydrogen bond acceptor), green ball; HBD (hydrogen bond donor), pink ball; hydrophobic, teal ball; negative charge or negative ionizable, dark blue ball; positive ionizable, red ball; aromatic ring, orange; hydrophobic-aromatic or aliphatic, light blue ball; positive charge, red ball.
The given active compounds (ligands) that have been determined experimentally to elicit neuroprotection were superimposed to determine their commonly shared chemical features. Conformational flexibility in each ligand that is aligned to define the pharmacophore, is performed by precomputing a series of low energy conformers for each. The molecule with the least number of conformations is used as a reference (naloxone) and all the conformations of the other active ligands are compared with the reference. This is repeated for the each conformation of the reference. The output is a series of hypothesis for pharmacophores with a mapping score given: the higher the score, the more of the features and properties matched each of the given ligands.
Statistical analysis
The data are expressed as the mean ± SE and statistical significance was assessed with an ANOVA followed by Bonferroni’s t test. A value of P <0.05 was considered statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The dynorphin (2
3
4)
peptide GGF was found to be the minimal peptide sequence required for neuroprotection, where scrambling the sequence (GFG) proved to be ineffective (Table 1)
. Indeed, all of the peptide fragments tested that contained the GGF sequence were shown to be neuroprotective (Table 1)
.
|
Naloxone and GGF are neuroprotective with similar efficacy and dose response
To discern whether naloxone exhibited a neuroprotective dose response similar to GGF, femtomolar concentrations of naloxone were assessed for the ability to prevent the LPS-induced death of DA neurons in neuron-glia cultures. Here, we show the comparison of the minimum dynorphin peptide sequence required for neuroprotection (GGF) and naloxone, where both compounds exhibit similar neuroprotective qualities at femtomolar concentrations (Fig. 1
A). The ability of DA neurons in mesencephalic cultures to take up [3H] DA after exposure to LPS was enhanced by 
35% with 30 min pretreatment of either naloxone or GGF, with the greatest level of protection conferred at the concentration of 10–14 M for both compounds. Immunocytochemical staining with an anti-tyrosine hydroxylase antibody (TH), a specific marker for DA neurons, demonstrated that 10–14 M naloxone and 10–14 M GGF both protected TH immunoreactive neurons from LPS-induced damage, such as loss of dendrites, axon disintegration, and loss of DA neurons (Fig. 1C
). Upon counting the number of TH-positive cells present after LPS treatment, it was apparent that GGF and naloxone pretreatment protected against LPS-induced DA neuron cell loss, with the peak protection occurring at 10–14 M for GGF and naloxone (Fig. 1B
). Taken together, these results indicate that the peptide GGF and the alkaloid naloxone both protected neurons from LPS-induced DA neuron cell death and loss of function with a similar efficacy and dose response.
|
Naloxone and GGF inhibit microglial ROS production
To test the effect of naloxone and GGF on the microglial generation of ROS, enriched-microglial cultures were pretreated with naloxone or GGF, then exposed to LPS. Both 10–14 M naloxone and 10–14 M GGF reduced intracellular ROS concentrations by 65% (Fig. 2
A) and reduced microglial superoxide response to nearly control levels (Fig. 2B
). These results demonstrate a similar efficacy and dose response of GGF and naloxone on microglial ROS levels, one of the pivotal signaling mechanisms governing microglia-mediated neurotoxicity (13)
.
|
Naloxone and GGF are neuroprotective through NADPH oxidase
In an effort to identify the mechanism through which naloxone and GGF are influencing the microglial inflammatory response to afford neuroprotection, the ability of GGF and naloxone to protect DA neurons from LPS-induced neurotoxicity in mesencephalic cultures from NADPH oxidase deficient mice (PHOX–/–) was determined. Microglia from PHOX–/– mice are unable to produce superoxide in response to LPS (13)
. Naloxone and GGF failed to show neuroprotection in PHOX–/– cultures (Fig. 3
A), supporting that inhibition of this enzyme is critical to the mechanism of action. The TNF- production was measured in response to LPS in PHOX–/– and PHOX+/+ mesencephalic neuron/glia cultures pretreated for 30 min with GGF and naloxone. Again, PHOX–/– mice failed to show any TNF- reduction in response to LPS with pretreatment of either neuroprotective compound, whereas the control mice showed a reduction of TNF- with naloxone and GGF treatment (10–14 M) (Fig. 3B
), demonstrating that these femtomolar-acting compounds inhibit the ROS-induced amplification of TNF- expression. Together, these results support the conclusion that GGF and naloxone afford neuroprotection through inactivation of NADPH oxidase.
|
Chemical similarity of femtomolar-acting neuroprotective compounds
Pharmacophore modeling and analysis were performed on the compounds experimentally identified as exhibiting neuroprotective activity: 1) naloxone; 2) DynA 1-17 YGGFLRRIRPKLKWDNQ; 3) DynA 2-17 GGFLRRIRPKLKWDNQ; 4) DynA 1-5 YGGFL; 5) DynA 2-5 GGFL; 6) DynA 2-4 GGF. The pharmacophore with the highest fit value is shown in Fig. 4
. Figure 4
depicts all neuroprotective molecules superimposed with the pharmacophore. This pharmacophore illustrates the common chemical properties that are shared among all the molecules with a common relationship in 3-dimensional space, which exhibit neuroprotective activity experimentally. The shared common properties identified by pharmacophore analysis that are necessary to elicit neuroprotection at femtomolar concentrations were: hydrogen bond acceptor, hydrogen bond donor, positive ionizable, and hydrophobic.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The ability to inhibit the microglial source of oxidative damage is critical to halting disease progression. Figure 5
depicts the critical role of NADPH oxidase in the generation of neurotoxic oxidative stress and demonstrates the therapeutic utility of inhibition of this enzyme complex. Several toxins, such as LPS (13)
, paraquat (20)
, particulate matter (21)
, and rotenone (22)
are selectively toxic to DA neurons through activation of NADPH oxidase. The microglial response to dying neurons treated with MPTP further propagates DA neurotoxicity through the activation of microglial NADPH oxidase (23)
. Postmortem analysis of the brains of PD patients reveals an up-regulation of the catalytic subunit of NADPH oxidase, gp91 (24)
. Together, this suggests that NADPH oxidase is critical to DA neurodegeneration. The targeting of gp91 and NADPH oxidase provides a novel and ideal neuroprotective strategy, as the production of neurotoxic ROS (Fig. 2A, B
) and proinflammatory factors, such as TNF- (Fig. 3B
), are both reduced. Here, we show that femtomolar concentrations of GGF and naloxone are neuroprotective through the inhibition of NADPH oxidase (Fig. 3A
), offering a novel neuroprotective approach for inhibition of microglial-derived oxidative stress.
|
The femtomolar dose response has been reported in multiple compounds (25
26
27
28
29)
where the pattern of the entire dose curve depends largely on the ligand and the cell type being tested (30)
. There are reported cases where the micromolar concentrations elicit responses in the opposite direction from the femtomolar concentrations and cases where responses from both ranges of the dose curve occur in the exact same direction (31)
. These reported differences in the dose curves from several femtomolar-acting compounds and cell types likely reflect the differences in mechanisms driving the phenomena (30)
. However, at this time relatively few definitive mechanisms for femtomolar-acting compounds are described. In the current study, we demonstrate that the femtomolar-acting peptide (dynorphin) can be reduced to the smallest biologically active fragment (GGF) (Table 1)
, which shares a similar neuroprotective dose response (Fig. 1A
) and mechanism (Fig. 3A
) with naloxone.
Pharmacophores are a graphical description of the 3-dimensional relationship of chemical features of ligands that are recognized by the receptors to which they bind. Pharmacophore modeling categorizes and generalizes the properties required for active ligands. Pharmacophores can then be used to predict other active ligands that will bind to a receptor when the tertiary molecular structure of the receptor is unknown. In the current study, we have identified common chemical properties required for the femtomolar-acting neuroprotection of the compounds tested in this study, as depicted by the pharmacophore superimposed on the structures of all compounds testing neuroprotective, as seen in Fig. 4
. Further supporting the validity of the pharmacophore in Fig. 4
, GFG failed to be neuroprotective and failed to fit the pharmacophore generated for the femtomolar-acting neuroprotective compounds. Future work in our laboratory will focus on using the pharmacophore shown in Fig. 4
to predict and test other femtomolar-acting neuroprotective compounds.
We have identified NADPH oxidase as a possible critical high-affinity target for the femtomolar regulation of microglial activation. We show that the neuroprotection (Fig. 3A
) and reduction of proinflammatory gene expression (Fig. 3B
) conferred by naloxone and GGF is dependent on presence of functioning NADPH oxidase. In another study, we have show that the opiate peptides leucine enkephalin and the kappa receptor inactive form, des-tyrosine-leucine enkephalin (DTLE), are also neuroprotective through inhibition of microglial NADPH oxidase (32)
, further supporting that several opiate peptides are neuroprotective through the common mechanism of NADPH oxidase attenuation. There is significant support that the femtomolar-acting mechanism of NADPH oxidase inhibition is independent of a traditional opiate receptor-mediated event. Whereas GGF is a component of dynorphin that binds preferentially to the kappa opiate receptor, the GGF peptide fragment is unable to bind the kappa receptor. As mentioned earlier, DTLE is unable to bind the kappa receptor and is neuroprotective at femtomolar concentrations through inhibition of NADPH oxidase (32)
. In the current study we report that naloxone, a nonspecific opiate antagonist, is neuroprotective at femtomolar concentrations, indicating a femtomolar-acting mechanism independent of traditional opiate receptor-mediated mechanisms. Naloxone has been shown to be neuroprotective at micromolar concentrations (33)
. In an earlier publication from our laboratory, (+) naloxone was extensively studied at micromolar concentrations both in vivo and in vitro, revealing an anti-inflammatory neuroprotective mechanism involving the suppression of superoxide (33)
. Though this indeed supports the involvement of NADPH oxidase in naloxone neuroprotection at micromolar concentrations, the similarities and differences of the mechanisms of actions for the neuroprotection at micromolar concentrations compared with femtomolar concentrations is unknown. We hypothesize that femtomolar and micromolar concentrations of GGF and naloxone inhibit the enzymatic activity of NADPH oxidase by binding to the gp91 subunit. Ongoing research in our laboratory is working to test this hypothesis.
The bimodal dose response is a common phenomenon seen in femtomolar-acting neuroprotective compounds tested in our laboratory (11
, 34)
. This bimodal response can be categorized by three unique sections of the dose curve. The physiological effects seen in the higher, micromolar concentrations are often assumed to be the traditional receptor mediated effects, but this assumption may not be applicable for all cases. Specifically, GGF is unable to bind the opiate receptors, suggesting an alternative mechanism. Naloxone had been identified as neuroprotective at micromolar concentrations through a mechanism independent of traditional opiate receptors (33)
. The femtomolar region of the curve is attributed to an atypical mechanism that is independent of the receptor-mediated event. The third component of the neuroprotective bimodal response is the middle portion of the curve (1010–1012 M), where the neuroprotective response is absent. There is now no clear explanation for the inability of these compounds to show any protective effect in the picomolar range (1010–1012 M). One interpretation is that multiple sites of action exist for the same drug, depending on the dose. In the case of two sites of action, the inhibition seen in the picomolar range may be due to an effect similar to "substrate inhibition," where the site of action for femtomolar concentrations is overwhelmed with increasing dose. Alternatively, a hypothetical third binding site in the picomolar range may mask the effect of these agents at femtomolar concentrations. This bimodal response is seen in several peptides for a multitude of functions. More research investigating the potential site of action for the femtomolar-acting compounds will provide valuable insight into the mechanism of the bimodal dose response.
Several neuroprotective compounds, such as delta opioids (35)
, naltrexone (36)
, the peptide NAPVSIPQ (37
, 38)
, vasoactive intestinal peptide (3)
, activity-dependent neuroprotective protein (39)
, and pituitary adenylate-activating peptide (37
, 40)
, have been reported to be biologically active at femtomolar concentrations. Although these compounds are reported to be neuroprotective and some have even demonstrated anti-inflammatory characteristics, several studies have reported femtomolar-acting mechanisms that likely are independent of NADPH oxidase activity. Thus, it seems likely that a complex pharmacology occurring at femtomolar concentrations is responsible for the determination of neuronal survival. In the current study, we add NADPH oxidase inhibition to the list of femtomolar-acting neuroprotective mechanisms and suggest that compounds chemically similar to the generated pharmacophore may be neuroprotective through this same mechanism.
The potential therapeutic utility of femtomolar-acting neuroprotective compounds is extensive. First, the neuroprotective characteristics of GGF and naloxone provide great hope for the treatment of neurodegenerative disease, as compounds effective at such low doses suggest a low potential for side effects. GGF and naloxone inhibit production of neurotoxic extracellular ROS and of proinflammatory cytokines, and thus represent a new class of compounds that will inhibit a broad spectrum of inflammatory profiles. Yet the targets of these compounds are limited to cells that express NADPH oxidase, which provides cell type selectivity, a substantial advantage for anti-inflammatory and neuroprotective drug design. The NADPH oxidase activation occurs within minutes in the microglial inflammatory response (13)
. Thus, naloxone and GGF inhibition of this enzyme will influence the early component of the microglial inflammatory response, offering a possible opportunity to inhibit the microglia before significant neuronal damage occurs.
The convergence of the functional, mechanistic, and chemical characteristics of the dynorphin peptides and naloxone provides the specificity necessary to test and to begin to understand how molecules at femtomolar concentrations exert biological effects. Thus, this work suggests that dynorphin is an endogenous femtomolar-acting compound and that naloxone acts as a GGF mimetic. Indeed, recent work from our laboratory has identified that another morphinan structurally similar to naloxone, dextromorphan (DM), is neuroprotective with the same bimodal dose response (34)
. DM possesses chemical characteristics similar to the femtomolar-acting, neuroprotective pharmacophore depicted in Fig. 4
., suggesting there may be an entire class of femtomolar-acting GGF mimetics with similar function, mechanism, and chemical properties. Here, we provide substantial support for the theory of ultra-low pharmacology and suggest an alternate mechanism of inflammatory control, presenting an ideal drug target for treatment of neurodegenerative disease.
|
ACKNOWLEDGMENTS |
|---|
Received for publication August 4, 2004. Accepted for publication December 2, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S.-L. Liu, Y.-H. Li, G.-Y. Shi, Y.-H. Chen, C.-W. Huang, J.-S. Hong, and H.-L. Wu A Novel Inhibitory Effect of Naloxone on Macrophage Activation and Atherosclerosis Formation in Mice J. Am. Coll. Cardiol., November 7, 2006; 48(9): 1871 - 1879. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Yang, J. Yang, Z. Yang, P. Chen, A. Fraser, W. Zhang, H. Pang, X. Gao, B. Wilson, J.-S. Hong, et al. Pituitary Adenylate Cyclase-Activating Polypeptide (PACAP) 38 and PACAP4-6 Are Neuroprotective through Inhibition of NADPH Oxidase: Potent Regulators of Microglia-Mediated Oxidative Stress J. Pharmacol. Exp. Ther., November 1, 2006; 319(2): 595 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Block, G. Li, L. Qin, X. Wu, Z. Pei, T. Wang, B. Wilson, J. Yang, and J. S. Hong Potent regulation of microglia-derived oxidative stress and dopaminergic neuron survival: substance P vs. dynorphin FASEB J, February 1, 2006; 20(2): 251 - 258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. QIN, Y. LIU, X. QIAN, J.-S. HONG, and M. L. BLOCK Microglial NADPH Oxidase Mediates Leucine Enkephalin Dopaminergic Neuroprotection Ann. N.Y. Acad. Sci., August 1, 2005; 1053(1): 107 - 120. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
