ANTISENSE oligonucleotides: a new tool in neuroscience

Clinical development of passive tau-based immunotherapeutics for treating primary and secondary tauopathies

INTRODUCTION

Tauopathies are clinicopathological entities with increased and pathological deposition in glia and/or neurons of hyperphosphorylated aggregates of the microtubule-binding protein tau. In secondary tauopathies, i.e. Alzheimer's disease (AD), tau deposition can be observed, but tau coexists with another protein (amyloid-β). In the last 20 years, little progress has been made in developing disease-modifying drugs for primary and secondary tauopathies and available symptomatic drugs have limited efficacy.

AREAS COVERED

The present review summarized recent advances about the development and challenges in treatments for primary and secondary tauopathies, with a focus on passive tau-based immunotherapy.

EXPERT OPINION

Several tau-targeted passive immunotherapeutics are in development for treating tauopathies. At present, 14 anti-tau antibodies have entered clinical trials, and 9 of them are still in clinical testing for progressive supranuclear palsy syndrome and AD (semorinemab, bepranemab, E2814, JNJ-63733657, Lu AF87908, APNmAb005, MK-2214, PNT00, and PRX005). However, none of these nine agents have reached Phase III. The most advanced anti-tau monoclonal antibody for treating AD is semorinemab, while bepranemab is the only anti-tau monoclonal antibody still in clinical testing for treating progressive supranuclear palsy syndrome. Further evidence on passive immunotherapeutics for treating primary and secondary tauopathies will come from ongoing Phase I/II trials.

Association of PSD-95 with ErbB4 facilitates neuregulin signaling in cerebellar granule neurons in culture

The growth factor neuregulin 1 (NRG) selectively induces an increase in the gamma-aminobutyric acid (GABA)(A) receptor beta2 subunit protein in rat cerebellar granule neurons in culture. We previously demonstrated that NRG acts by triggering ErbB4 receptor phosphorylation and subsequent signaling through the mitogen-activated kinase (MAPK), phosphatidyl inositol-3 kinase (PI-3K) and cyclin-dependent kinase 5 (cdk5) pathways. In this report we show that the scaffolding protein, PSD-95, plays a key role in mediating the effects of NRG and that reducing its level attenuates the NRG-induced increase in beta2 subunit expression. PSD-95 appears to facilitate the effects of NRG through its association with ErbB4, an interaction that is augmented by NRG-activated cdk signaling. Inhibition of cdk activity with roscovitine attenuates the association of PSD-95 with ErbB4. The effects of cdk5 are not blocked by U0126, an inhibitor of MAPK signaling, indicating that cdk5 functions independently of cross-talk with this pathway. These findings raise the possibility that NRG-induced activation of cdk5 works in part by recruiting PSD-95, a protein involved in regulating synaptic plasticity, to associate with ErbB4. This interaction may be a positive feedback loop that augments NRG signaling and its downstream effects on GABA(A) receptor beta2 subunit expression.

Signaling by bone morphogenetic proteins and Smad1 modulates the postnatal differentiation of cerebellar cells

Previous studies have demonstrated that bone morphogenetic proteins (BMPs) activate the Smad1 signaling pathway to regulate cell determination and differentiation in the embryonic nervous system. Studies examining gene and protein expression in the rat cerebellum suggest that this pathway also regulates postnatal differentiation. Using microarrays, we found that Smad1 mRNA expression in the cerebellum increases transiently at postnatal day 6 (P6). Immunohistochemistry and Western blots showed that Smad1 and BMP4 proteins are present in the cerebellum, and that their expression also changes postnatally. The proteins are detectable at P4-P6, a stage at which most cerebellar cells reside in the external germinal layer (EGL), where they extensively differentiate. The levels become maximal at P8-P10, when neurons begin to migrate from the EGL into their mature positions in the internal granule layer. In cerebellar cultures prepared at P6 or P10, BMP4 activates Smad1 signaling to modulate cell differentiation. Brief BMP4 application caused Smad1 translocation from the neuronal cytoplasm into the nucleus, where it is known to regulate transcription in association with Smad4. Longer BMP4 treatment promoted the differentiation of both neuronal and non-neuronal cells. By 3 d, neuronal processes appeared more fasciculated, and the level of synaptotagmin, a protein found in synaptic vesicles, increased. In addition, many astroglial cells became more branched and stellate in morphology. The BMP-induced changes were reduced by treatment with antisense oligonucleotides to Smad1 or Smad4. These findings in vivo and in culture suggest that BMP4 and Smad1 signaling participate in regulating postnatal cerebellar differentiation.

Nerve growth factor plays a divergent role in mediating growth of rat C6 glioma cells via binding to the p75 neurotrophin receptor

Dysregulation of proliferation, differentiation and cell death play a major role in glial tumors, and there is evidence for regulatory mechanisms involving nerve growth factor (NGF) and its receptors in various CNS-derived tumor cell lines. The aim of our study was to observe the effect of exogenous recombinant NGF on C6 rat glioma growth, to characterize the role of endogenous NGF and the p75 neurotrophin receptor (p75) and to rule out whether p75 is necessary to mediate the effect of exogenous NGF. Recombinant exogenous NGF (1-100 ng/ml) was applied under different serum conditions (0%, 1%, 5%) and knockdown of endogenous NGF and p75 was achieved by lipid-mediated antisense oligonucleotide treatment. In presence of serum, NGF had a positive whereas in absence of serum NGF produced a negative effect on C6 cell number. A knockdown of NGF or p75 increased cell numbers and enhanced BrdU incorporation. In p75-knocked down cells NGF did not enhance C6 glioma growth in presence of serum. We conclude that (1) exogenous recombinant NGF enhances C6 glioma growth under serum conditions but decreases cell number in absence of serum, that (2) the effect of exogenous NGF is mediated by p75 alone or by heterodimers containing p75 and that (3) either basal levels of endogenous NGF or basal levels of p75 receptor moderate C6 glioma growth and represent an autoregulatory potential of C6 glioma cells.

Modulation of the gene expression of N-methyl-D-aspartate receptor NR2B subunit in the rat neostriatum by a single dose of specific antisense oligodeoxynucleotide

N-methyl-D-aspartate receptors (NRs) are a group of ionotropic glutamate receptors in the brain and they are composed of heteromeric subunits (NR1, NR2A-D and NR3). In the neostriatum, a brain region that is associated with movement in animals, NMDA channels are known to involve in the motor control. Our previous report (Lai et al., 2000, Neuroscience 98, 493-500) has shown that a single dose of antisense oligodeoxynucleotides that are specific to NR1 subunit results in blockage of the gene expression of NR1 as well as NR2A subunits in the neostriatum. In the present study, antisense oligodeoxynucleotides that are specific to NR2B (ANR2B) were then employed as molecular tools to further investigate the molecular interactions of NMDA receptor subunits in the neostriatum. A single dose of ANR2B was injected unilaterally into the rat neostriatum. After one day of injection, no modification of motor behavior was found in the ANR2B-injected rats. The mRNA level of NR2B in the ANR2B-injected neostriatum was found to be decreased (-20.4%) by reverse transcriptase polymerase chain reaction (RT-PCR). However, the mRNA levels of NR1, NR2A, NR2C and NR2D in the ANR2B-treated neostriatum were found to be unchanged. After two days of injection, NR2B immunoreactivity was found to decrease in the ANR2B-treated neostriatum by immunofluorescence (-35.1%). At higher magnification, NR2B immunoreactivity was found to decrease in presumed spiny neurons of the neostriatum (-23.4%). No change in NR1 immunoreactivity was observed. These results indicate that a single dose of ANR2B can successfully block the gene expression of NR2B in neurons of the neostriatum and there is less effect on NR1 and other NR2 subunits. The blockage of the gene expression of NR2B is therefore specific and the present results may provide important implications in applications of antisense in research and in clinical therapy of neurological diseases.

Glial cell line-derived neurotrophic factor (GDNF) is a proliferation factor for rat C6 glioma cells: evidence from antisense experiments

Growth factors play an important role in proliferation and differentiation of malignant brain gliomas in humans. Glial cell line-derived neurotrophic factor (GDNF) has been shown recently to be highly expressed in human glioblastomas and in rat glial cell lines B49 and C6. The aim of the present study was to knockdown GDNF, its receptor GFR-alpha1, and the related family member persephin by using antisense oligonucleotides and to observe the effects on cell proliferation. To enhance cellular uptake into C6 glioma cells, 15-mer phosphorothioate oligonucleotides were complexed with the cationic lipid Lipofectamine. The complex was applied for 3 x 12 hours to C6 glioma cells, and cells were allowed to recover for 24 hours after each transfection and then analyzed. This protocol markedly reduced GDNF and GFR-alpha1 protein levels in C6 glioma cells compared with control oligonucleotides. Knockdown of C6 cells with GDNF and GFR-alpha1 but not with persephin antisense oligonucleotides significantly decreased the number of C6 glioma cells and also inhibited the incorporation of bromodeoxyuridine as a sign of reduced DNA synthesis. In conclusion, it is shown that GDNF but not persephin is a potent proliferation factor for rat glioma cells. Knockdown of GDNF using antisense oligonucleotides complexed with lipids as carriers may be useful in gene therapeutic approaches in vitro and possibly also in vivo.

Dopamine transporter function assessed by antisense knockdown in the rat: protection from dopamine neurotoxicity

The plasma membrane dopamine transporter is located on presynaptic nerve terminals and is responsible for the termination of dopaminergic neurotransmission via dopamine reuptake. The dopamine transporter may also contribute to the pathogenesis of Parkinson disease. Dopamine transporter expression correlates well with susceptibility to neuronal degeneration in 1-methyl-4-phenyl-1,2,3,6 -tetrahydropyridine (MPTP)-induced parkinsonism. Recent studies have implicated the dopamine transporter in the uptake of both this neurotoxin and its metabolite, MPP(+), as well as another experimental neurotoxin, 6-hydroxydopamine. In these studies we examined the role of the dopamine transporter in the neurotoxicity of both MPP(+) and 6-hydroxydopamine in the rat brain using in vivo administration of phosphorothioate antisense oligonucleotides targeting dopamine transporter mRNA. Infusion of dopamine transporter antisense (1 nmol/day, 7 days) into the left substantia nigra pars compacta resulted in reduced (3)H-WIN 35-428 binding in the left striatum and significant levodopa and amphetamine-induced contralateral rotations. Unilateral pretreatment with dopamine transporter antisense prior to bilateral intrastriatal infusion of either MPP(+) or 6-hydroxydopamine resulted in asymmetrical striatal (3)H-WIN 35-428 binding and dopamine content as well as significant apomorphine-induced ipsilateral rotations, suggesting neuroprotection of nigrostriatal neurons on the antisense-treated side. Thus, the dopamine transporter appears to play a critical role in determining susceptibility to the experimental neurotoxins MPP(+) and 6-hydroxydopamine. In light of this, the dopamine transporter may prove useful, both as a marker for susceptibility to Parkinson's disease and as a target for therapeutic intervention.

Effects of oligonucleotide antisense to dopamine D(1A) receptor messenger RNA in a rodent model of levodopa-induced dyskinesia

Dyskinesias are abnormal involuntary movements which develop as a side-effect of long-term treatment with levodopa in patients with Parkinson's disease. The pathophysiology underlying these dyskinesias remains unclear, although, it has been suggested that heightened activity of dopamine D(1) receptor-bearing striatonigral neurons may play a key role. Chronic pulsatile levodopa administration to hemiparkinsonian rats results in sensitization of rotational responses to apomorphine. This sensitization is thought to be analogous to levodopa-induced dyskinesias in humans. In these studies, we further clarify the role of the dopamine D(1A) receptor in this rodent model of levodopa-induced dyskinesias using an in vivo oligonucleotide antisense approach. Hemiparkinsonian rats received twice daily injections of levodopa for three weeks followed by intrastriatal infusion of dopamine D(1A) receptor antisense (7nmol/day, three days), a scrambled missense control sequence, or saline. Those animals treated with antisense displayed significantly fewer apomorphine-induced rotations than saline- or missense-treated controls.By reducing dopamine D(1A) receptor expression, we were able to attenuate sensitization of the response to apomorphine resulting from chronic pulsatile levodopa treatment. Thus, the dopamine D(1A) receptor appears to play a significant role in levodopa-induced dyskinesias and warrants further examination. These findings may have important implications for the development of selective treatment strategies designed to alleviate parkinsonian symptoms, while minimizing motor complications.

Changes in expression of N-methyl-D-aspartate receptor subunits in the rat neostriatum after a single dose of antisense oligonucleotide specific for N-methyl-D-aspartate receptor 1 subunit

In order to investigate the process of gene expression of N-methyl-D-aspartate glutamate receptor (NMDAR) subunits in the rat neostriatum and how this relates to motor behaviors, a single dose of antisense phosphodiester oligodeoxynucleotide specific for NMDAR1 was unilaterally applied in the neostriatum in a stereotaxic apparatus. After one day of antisense treatment, ipsilateral rotation behaviors that were induced by apomorphine were found in the treated animals. Reductions in the levels of expression of NMDAR1 and NMDAR2A messenger RNAs (NMDAR1: 20.6%; NMDAR2A: 19.7%) were found in the antisense-treated striatal tissues by reverse transcriptase-polymerase chain reaction. There was no change in the levels of NMDAR2B, NMDAR2C and NMDAR2D messenger RNAs. After two days, western blotting experiments showed that there were decreases in the levels of expression of NMDAR1 (decreased 27.6%) and NMDAR2A (decreased 19.2%) proteins in the NMDAR1 antisense-treated striatal tissues. In addition, NMDAR1 immunoreactivity was found to decrease in intensity in the NMDAR1 antisense-treated neostriatum. At the cellular level, the intensity of NMDAR1 immunoreactivity in perikarya of presumed medium spiny neurons was found to decrease. These results indicate that a single dose of NMDAR1 antisense modifies the expression of NMDAR1 messenger RNA and protein in neurons in the neostriatum. The modification in the expression of NMDAR1 has differential effects in the expression of NMDAR2 subunits. Gene expression of the native NMDAR subunits is likely to be a dynamic process. The change in gene expression of the NMDAR subunits in the neostriatum may have a profound effect on the motor behaviors of rats.

Antidepressant effects on GABA-stimulated 36Cl(-) influx in rat cerebral cortex are altered after treatment with GABA(A) receptor antisense oligodeoxynucleotides

Antidepressants act at the GABA(A) receptor to inhibit GABA-stimulated 36Cl(-) influx and GABA reduction of [35S]TBPS binding. This study examined how selective knock-down (via antisense oligodeoxynucleotides, aODNs) of GABA(A) receptor subunits modified antidepressant activity. The specific aODNs used were for the alpha1, beta1, beta2 or gamma2 subunits of the GABA(A) receptor. The aODN microinjections reduced corresponding GABA(A) receptor subunit mRNA levels by 30-40% as assessed by RT-PCR. The inhibitory effect of the antidepressants amitriptyline and mianserin on GABA-stimulated 36Cl(-) influx was decreased after microinjections of alpha1, beta1, or beta2 subunit aODNs but potentiated after microinjections of gamma2 subunit aODNs. This pattern of aODNs effect on amitriptyline and mianserin modulation of GABA-stimulated 36Cl(-) influx was the same for both antidepressants and similar to GABA but different than that of diazepam and bicuculline. We conclude that multiple subunits of the GABA(A) receptor regulate the effect of amitriptyline and mianserin on the GABA(A) receptor chloride ionophore complex. However, the exact identity of the subunit mediating the direct or allosteric modulation of the antidepressant effect on GABA-stimulated 36Cl(-) influx remains unclear.

Effects of treatment with GABA(A) receptor subunit antisense oligodeoxynucleotides on GABA-stimulated 36Cl- influx in the rat cerebral cortex

GABA(A) receptor function was studied in cerebral cortical vesicles prepared from rats after intracerebroventricular microinjections of antisense oligodeoxynucleotides (aODNs) for alpha1, gamma2, beta1, beta2 subunits. GABA(A) receptor alpha1 subunit aODNs decreased alpha1 subunit mRNA by 59+/-10%. Specific [3H]GABA binding was decreased by alpha1 or beta2 subunit aODNs (to 63+/-3% and 64+/-9%, respectively) but not changed by gamma2 subunit aODNs (94+/-5%). Specific [3H]flunitrazepam binding was increased by alpha1 or beta2 subunit aODNs (122+/-8% and 126+/-11%, respectively) and decreased by gamma2 subunit aODNs (50+/-13%). The "knockdown" of specific subunits of the GABA(A )receptor significantly influenced GABA-stimulated 36Cl- influx. Injection of alpha1 subunit aODNs decreased basal 36Cl- influx and the GABA Emax; enhanced GABA modulation by diazepam; and decreased antagonism of GABA activity by bicuculline. Injection of gamma2 subunit aODNs increased the GABA Emax; reversed the modulatory efficacy of diazepam from enhancement to inhibition of GABA-stimulation; and reduced the antagonist effect of bicuculline. Injection of beta2 subunit aODNs reduced the effect of diazepam whereas treatment with beta1 subunit aODNs had no effect on the drugs studied. Conclusions from our studies are: (1) alpha1 subunits promote, beta2 subunits maintain, and gamma2 subunits suppress GABA stimulation of 36Cl- influx; (2) alpha1 subunits suppress, whereas beta2, and gamma2 subunits promote allosteric modulation by benzodiazepines; (3) diazepam can act as an agonist or inverse agonist depending on the relative composition of the receptor subunits: and (4) the mixed competitive/non-competitive effects of bicuculline result from activity at alpha1 and gamma2 subunits and the lack of activity at beta1 and beta2 subunits.

Downregulation of brain mineralocorticoid and glucocorticoid receptor by antisense oligodeoxynucleotide treatment fails to alter spatial navigation in rats

Adult male Brown Norway rats were long-term intracerebroventricularly (i.c.v.) infused with antisense oligodeoxynucleotides (18-mer, double endcapped phosphorothioate protected) targeting either mineralocorticoid or glucocorticoid receptor mRNA, or received the respective mixed bases sequence or vehicle. Mineralocorticoid receptor-mixed bases and glucocorticoid receptor-mixed bases oligodeoxynucleotide infusion (1 microg/0.5 microl/h) over a time period of seven days did not alter hippocampal mineralocorticoid receptor and glucocorticoid receptor binding when compared to vehicle treatment. In contrast, i.c.v. administration of mineralocorticoid receptor, as well as glucocorticoid receptor-antisense over the same time period resulted in a significantly reduced binding of mineralocorticoid receptor and glucocorticoid receptor in the hippocampus [mineralocorticoid receptor-antisense group approx. 72% of mineralocorticoid receptor-mixed bases and vehicle groups (100%); glucocorticoid receptor antisense group approx. 77% of glucocorticoid receptor-mixed bases and vehicle]. The specificity of these antisense effects is indicated by the finding that rats treated with mineralocorticoid receptor-antisense did not show any changes in glucocorticoid receptor and vice versa. Animals treated according to this infusion protocol and tested in the Morris water maze for their spatial navigation abilities failed to show significant differences among the groups. These data indicate that a reduction of hippocampal mineralocorticoid receptor or glucocorticoid receptor binding capacity by 20-30% does not interfere with spatial navigation.

Drug delivery of antisense molecules to the brain for treatment of Alzheimer's disease and cerebral AIDS

Antisense oligonucleotides (ODNs) and peptide nucleic acids (PNAs) are potential therapeutics for eradication of malignancies, viral infections, and other pathologies. However, ODNs and PNAs in general are unable to cross cellular membranes and blood-tissue barriers, such as the blood-brain barrier (BBB), which is only permeable to lipophilic molecules of molecular weight <600 Da. Cellular delivery systems based on conjugates of streptavidin (SA) and the OX26 monoclonal antibody directed to the transferrin receptor may be employed as a universal carrier for the transport of mono-biotinylated peptides, ODNs, or PNAs. 3'-Biotinylation of phosphodiester (PO)-ODN produces complete protection of ODN against serum and cellular 3'-exonucleases, facilitating the conjugation to avidin-based delivery systems and maintaining the activation of RNase H. These delivery systems markedly increased the cellular uptake and antisense efficacy of 3'-biotinylated ODNs in models of Alzheimer's disease and HIV-AIDS. In vivo brain delivery studies demonstrated that 3'-protected PO-ODNs and PO-phosphorothioate(PS)-ODN hybrids containing a single PO linkage are subjected to endonuclease degradation in vivo. On the contrary PS-ODNs, which were also protected at 3'-terminus by biotinylation, are metabolically stable in vivo and resistant to exo/endonuclease degradation. However, because of the strong binding of these oligomers to plasma protein, PS-ODNs are poorly transported into the brain through the BBB by the OX26-SA delivery vector following intravenous administration. PNAs are also resistant to exo/endonuclease and protease degradation, and these molecules biotinylated at the amino terminal group were transported into the brain by the OX26-SA delivery system with brain uptake levels comparable to that of morphine. Using the rev gene of HIV as a model target, RNase protection assays and cell-free translation arrest showed that the PNA-OX26-SA conjugate maintained active recognition and inactivation of target mRNA, respectively. The overall experimental evidence suggests that PNA-OX26-SA conjugates represent optimal antisense molecules for drug delivery to the brain.

Antisense oligonucleotides against ‘cardiac’ and ‘skeletal’ DHP-receptors reveal a dual role for the ‘skeletal’ isoform in EC coupling of skeletal muscle cells in primary culture

Two dihydropyridine receptor mRNA isoforms (cardiac and skeletal) are expressed in rat skeletal muscle cells in primary culture. The progressive changes in excitation-contraction coupling mode from dual mode (‘skeletal’ and ‘cardiac’) to predominant ‘skeletal’ one during in vitro myogenesis are thought to be linked to the developmental changes in the relative expression of the two types of molecular entity previously observed in this preparation. In order to test this hypothesis, myotube cultures (5- to 7-day-old) were treated with antisense phosphorothioated oligodeoxynucleotides against cardiac or skeletal alpha1 subunit of L-type calcium channel. The oligodeoxynucleotide uptake by cells was checked by means of imaging of fluorescent oligodeoxynucleotide derivatives within the cells. Optimum concentration used (10 microM in the extracellular medium) and incubation time (70 hours) were empirically determined. Antisense directed against the cardiac type led to a 54% decrease in the averaged L-type calcium current peak density at −10 mV. The same type of experiment was performed with antisense against the skeletal isoform and led to a same order of inhibition (46%). This result clearly shows that the two isoforms can work as a calcium channel. Conversely, analysis of the shape of T-V (relative contractile amplitude versus membrane potential) curves shows that the treatment with ‘skeletal’ antisense depressed the contractile response in the medium membrane potential range whereas treatment with ‘cardiac’ antisense had no effect. This and other results taken together suggest that the skeletal isoform of dihydropyridine receptor is involved in both ‘cardiac’ and ‘skeletal’ types of EC coupling mechanisms at work in early stages of myotubes in vitro development. The type of coupling probably depends on the proximity of the skeletal dihydropyridine receptor and the ryanodine receptor.

Intracerebroventricular administration of antisense oligodeoxynucleotide against GLUT2 glucose transporter mRNA reduces food intake, body weight change and glucoprivic feeding response in rats

The GLUT2 glucose transporter, which may play a glucose-sensing role in hepatocyte and islet beta cells because of its low affinity and high Km for glucose, has been identified in some discrete brain areas that are related to feeding behavior and energy metabolism. We tested the hypothesis that brain GLUT2 may play a role in the control of food intake by antisense technology loss-of-function analysis. Antisense oligonucleotides directed against GLUT2 mRNA were administered intracerebroventricularly to eight rats daily for 13 days. Another eight rats were administered intracerebroventricularly with missense oligonucleotides as the control. Food intake was monitored by a computerized feeding system. Data were analyzed using a one-way general linear model or Mann-Whitney U test when appropriate. Cumulative food intake and body weight change in antisense-treated rats were significantly lower (18 and 160%, respectively) in the group treated with antisense oligonucleotides than in the group treated with missense control oligonucleotides. There was no increase in cumulative food intake in response to 2-deoxyglucose challenge in rats treated with antisense oligonucleotide, but in those treated with missense control oligonucleotide, cumulative food intake was fivefold greater in response to 2-deoxyglucose. These data suggest a possible role of brain GLUT2 in the regulation of food intake and body energy stores.

Therapeutic potential and mechanism of action of oligonucleotides and ribozymes

Specific inactivation of gene expression is an attractive approach for rational drug design to combat degenerative diseases and infectious agents. Oligonucleotide-directed triple-helix formation at cis-acting elements of gene promoters, short oligonucleotides containing base sequences that are complementary to the messenger RNA (antisense oligos), and RNA enzymes (ribozymes) that specifically cleave messenger RNA molecules are currently being used both as experimental tools and as therapeutic agents. Mechanisms of action of various oligonucleotide-based drugs, recent developments in the drug-delivery approaches, and future potentials are discussed in this review.

The use of in vivo antisense oligonucleotide technology for the investigation of brain GABA receptors

Antisense oligodeoxynucleotides (ODN) can be used as selective inhibitors of in vivo gene expression in the central nervous system (CNS) of experimental animals. The gamma-aminobutyric acid type A (GABAA) receptor is a member of the ligand-gated ion channel superfamily of neurotransmitter receptors. GABAA receptor function is allosterically modulated by several clinically important compounds, e.g. 1,4-benzodiazepines, barbiturates and certain neurosteroids, which recognize binding sites within the receptor complex. GABAA receptor chloride channel complexes are probably pentamers of different polypeptide subunits. The number of known subunit families and isoforms (six alpha s, four beta s, three gamma s, one delta and two rho s) indicates an extensive heterogeneity of GABAA receptors. The gamma 2 subunit is a functionally integral part of the GABAA receptor, necessary for the high affinity binding of benzodiazepines. The infusion of phosphorothioate ODN antisense to the gamma 2 subunit mRNA, but not control sense or mismatch ODN, into the lateral cerebral ventricle or into the hippocampus of rats leads to significant decreases in benzodiazepine receptor radioligand binding. In the hippocampus this is accompanied by a decrease in the number of GABAA receptors and by a loss of neurones, the latter possibly being due to reduced GABAergic inhibitory neurotransmission. Autoradiographic analysis following continuous intrahippocampal infusion of antisense ODN shows the regional extent of the effect on [3H]flunitrazepam binding. The continuous infusion of antisense ODN, but not of mismatch control ODN, into the right lateral cerebral ventricle induced a significant decrease in benzodiazepine binding and [3H]muscimol binding to membranes of the right cortex. Antisense ODN infused into the striatum decreased benzodiazepine binding and binding to the GABA binding site of the GABAA receptor to an extent similar to that found in the hippocampus. It is concluded that the preferred route of administration of antisense ODN for in vivo studies of the GABAA receptor may be by infusion into defined rat brain regions. The reported data support the idea that antisense ODN can be used as a valuable tool for the investigation of the contribution of individual GABAA receptor subunits to the properties of the receptor complex and of mechanisms of receptor subunit assembly.

Antisense strategies in neurobiology

The use of antisense oligodeoxynucleotides, targeted to the transcripts encoding biologically active proteins in the nervous system, provides a novel and highly selective means to further our understanding of the function of these proteins. Recent studies of these agents also suggest the possibility of their being used therapeutically for a variety of diseases involving neuronal tissue. In this paper we review studies showing the in vitro and in vivo effects of antisense oligodeoxynucleotides as they relate to neurobiological functions. Particular attention is paid to the behavioral and biochemical effects of antisense oligodeoxynucleotides directed to the various subtypes of receptors for the neurotransmitter dopamine. An example is also provided showing the effects of a plasmid vector expressing an antisense RNA targeted to the calmodulin mRNAs in the PC12 pheochromocytoma cell line. The advantages of antisense oligodeoxynucleotides over traditional pharmacological treatments are assessed, and the advantages of using vectors encoding antisense RNA over the use of antisense oligodeoxynucleotides are also considered. We also describe the criteria that should be used in designing antisense oligodeoxynucleotides and several controls that should be employed to assure their specificity of action.

Diazepam protects against rat hippocampal neuronal cell death induced by antisense oligodeoxynucleotide to GABA(A) receptor gamma2 subunit

Antisense oligodeoxynucleotides (ODNs) are used for the selective inhibition of gene expression. Antisense ODNs are promising tools for the investigation of physiological implications of proteins in the central nervous system of rodents in vivo. We have previously demonstrated that a phosphorothioate antisense ODN to the GABA(A) receptor gamma2 subunit, but not sense or mismatch control ODNs, induces a decrease in ex vivo benzodiazepine receptor radioligand binding in rat hippocampus when infused into the hippocampus in vivo [Karle et al., Neurosci. Lett., 202 (1995) 97-100]. This effect is parallelled by a decrease in the number of GABA(A) receptors and an extensive loss of hippocampal neurones. There is increasing awareness of risks of toxic 'non-antisense' effects induced by ODNs, and in particular phosphorothioate ODNs. The present experiments were designed to investigate the specificity of effects induced by the gamma2 subunit antisense ODN. The temporal development of changes in [3H]flunitrazepam and [3H]quinuclidinyl benzilate binding as well as in tissue protein levels supports the notion that the antisense ODN primarily acts by blocking the expression of the targeted receptor subunit protein. Furthermore, it is shown that a threshold for the elicitation of neurodegenerative changes exists. Finally, it is demonstrated that diazepam treatment of rats protects against the development of neuronal cell death induced by the antisense ODN. Collectively, the results support the hypothesis that the neurodegeneration induced by the antisense ODN is a consequence of diminished GABAergic inhibitory tonus following a selective down-regulation of gamma2 subunit-containing GABA(A) receptor complexes.

Angiotensinogen antisense oligonucleotides and fluid intake

The effectiveness of antisense oligonucleotides (ODNs) to angiotensinogen on intracerebrovenricularly injected renin induced thirst was investigated. As a corollary, information would be gained about the role of centrally synthesised angiotensinogen in the neural mechanisms subserving water drinking in rats. Stable, easily synthesised phosphorothioate antisense oligonucleotides (18 mer), one of which included the sequence encompassing the translation start site, were injected into the lateral ventricle of rats. The drinking response to a number of dipsogenic stimuli was tested. Antisense significantly reduced (by about 50%) the volume of water drunk in response to intracerebroventricular (icv) renin or isoproterenol but did not reduce drinking in response to the physiological challenge of icv angiotensin II, icv carbachol, intravenous hypertonic saline, water deprivation or subcutaneous injection of polyethylene glycol. Only one out of four antisense probes gave positive results, while mismatch or scrambled oligonucleotides did not inhibit water intake. This finding reduces the probability that the results observed are non-specific. In these experiments, an ODN specific for angiotensinogen was discovered and was produced easily in large enough amounts and stabilised against intracellular nucleases without floss of cellular access or biological effect.

Selective attenuation of mu-opioid receptor-mediated effects in rat sensory neurons by intrathecal administration of antisense oligodeoxynucleotides

To test the hypothesis that the expression of specific proteins on peripheral terminals of primary afferents can be attenuated by intrathecal administration of antisense oligodeoxynucleotides (ODNs), we administered ODNs antisense to the mu-opioid receptor to male Sprague-Dawley rats via chronically implanted intrathecal cannulae. Antisense but not mismatch ODN treatment significantly decreased peripheral (D-Ala2, N-Me-Phe4, Gly5-ol)-enkephalin (DAMGO) inhibition of prostaglandin E2 (PGE2) hyperalgesia. Antisense treatment affected neither the magnitude of PGE2 hyperalgesia nor the antinociception produced by a peripherally administered adenosine A1-agonist. The antinociceptive effects of DAMGO was fully recovered 8 days after cessation of ODN treatment. DAMGO-induced inhibition of voltage-gated Ca2+ currents (VGCC), in cultured dorsal root ganglion (DRG) neurons from rats treated with ODNs, was also significantly reduced by antisense but not mismatch ODNs. Taken together, these observations suggest that intrathecal administration of antisense ODNs can be used to study the function of proteins present in the peripheral terminals of primary afferent neurons.

The role of angiotensin II in ingestive behaviour: a brief review of angiotensin II, thirst and Na appetite

From the outset, the study of angiotensin II (Ang II) in body fluid homeostasis has been both complicated and intriguing. Since the publication of an early report of the dipsogenic action of this peptide, the pursuit of the role of Ang II in thirst and Na appetite has continued for the last 25 years. This pursuit captured the attention of all workers interested in the behavioural/physiological regulation of body fluid balance, with major contributions being made by James T. Fitzsimons and his colleagues. In spite of its powerful dipsogenic actions, delineation of its precise role in physiological thirst has been elusive and difficult to demonstrate. The influence of Ang II on Na intake took longer to show convincingly. However, in contrast to thirst, the role of Ang II in physiological Na appetite has been demonstrated clearly. The technological advances made during the recent years have greatly increased our ability to delineate the neurobiological context of Ang II-mediated responses. Thus, the future is promising in regard to illuminating the subtleties of the role of Ang II in body fluid balance.

Uptake and distribution of fluorescein-labeled D2 dopamine receptor antisense oligodeoxynucleotide in mouse brain

To determine the uptake and distribution of oligodeoxynucleotides in brain, a 20-mer phosphorothioated oligodeoxynucleotide complementary to a portion of the D2 dopamine receptor mRNA was fluorescently labeled with fluorescein isothiocyanate (FITC) and injected into the lateral cerebral ventricles of mice. At various survival times after the injection, the brains were removed, fixed, sectioned, and viewed under a fluorescent microscope. The results showed that the oligodeoxynucleotide was rapidly taken up into the brain. Initially the label was relatively diffusely spread throughout the interstitial spaces of the brain, then became redistributed to the cellular compartments. The signal extended from those forebrain nuclei located immediately in contact with the ventricles, such as the corpus striatum, septum, and hippocampus, to areas further removed from the ventricles, such as the cerebral cortex, nucleus accumbens, and substantia nigra. When the FITC-labeled D2 antisense oligodeoxynucleotide was given once daily for 4 d, the signal intensity seen 24 h after the last injection appeared to be of greater intensity overall compared to that seen after a single injection. At early time-points the oligodeoxynucleotide signals appeared to be punctuated and were found in cell bodies as well as in proximal dendritic processes. However, not all cells were equally labeled, suggesting an uneven uptake and accumulation of the D2 antisense into the various cell types. At later time-points the fluorescent signal appeared granular; at these times the injected material was largely degraded. These studies show that a D2 dopamine receptor antisense oligodeoxynucleotide is rapidly taken up from cerebral ventricles into brain, becomes widely distributed throughout the brain tissue to areas far removed from direct contact with the ventricles, and appears to accumulate to a different extent in the various brain areas and cell types.

Antisense oligonucleotide to GABAA receptor gamma 2 subunit induces loss of neurones in rat hippocampus

The binding site for 1,4-benzodiazepines in the brain is part of the hetero-oligomeric gamma-aminobutyric acid (GABA)A receptor complex which regulates a chloride ion channel. The presence of the gamma 2 subunit in the complex is necessary for the binding of benzodiazepines to their binding site. This study demonstrates a reduction of benzodiazepine receptor radioligand binding by 43% compared to control following infusion of phosphorothioate antisense oligodeoxynucleotide to gamma 2 subunit into rat hippocampus. Reduction of benzodiazepine binding sites was paralleled by a decrease in [35S]tert-butyl-bicyclo-phosphorothionate ([35S]TBPS) binding (51%) and [3H]muscimol binding (37%), indicating a reduction in the number of GABAA receptors. Changed macroscopic appearance, reduced protein content and severe loss of neurones in antisense-treated hippocampi suggests that the reduced formation of GABAA receptors leads to neuronal cell death.

Antisense to thyrotropin releasing hormone receptor reduces arterial blood pressure in spontaneously hypertensive rats

We report in the present study the effect of intrathecal treatment with antisense oligonucleotides complementary to thyrotropin releasing hormone (TRH) receptor mRNA on the pressor response to intrathecal administration of TRH and on resting arterial blood pressure in Wistar-Kyoto (WKY) rats and spontaneously hypertensive rats (SHR). In 16-week-old male WKY rats, 18-base phosphodiester antisense or mismatch oligonucleotides to TRH receptor mRNA (100 micrograms per day) were injected intrathecally for 3 days. Twenty-four hours after the last injection, the magnitude of the pressor response to intrathecal TRH (10 micrograms) was significantly smaller in the antisense-treated group (n = 7) compared with mismatch-treated controls (n = 7) (change in mean arterial pressure, +20.3 +/- 3.0 versus +32.6 +/- 2.5 mm Hg, P < .01). No differences were observed in the pressor responses to injection of N-methyl-D-aspartic acid. Resting arterial blood pressure was unaffected by antisense treatment in WKY rats. In separate experiments, 16-week-old male SHR were treated with antisense (n = 7) or mismatch (n = 6) oligonucleotides for 3 days. Mean resting arterial blood pressure was significantly reduced by treatment with antisense oligonucleotides (from 157 +/- 4.8 to 119 +/- 8.8 mm Hg, P < .01), but no significant changes were observed in mismatch-treated animals. Our results suggest that the expression of TRH receptors in spinal sympathetic preganglionic neurons can be selectively reduced by intrathecal treatment with antisense oligonucleotides and that TRH projections to sympathetic preganglionic neurons play an important role in the elevation of arterial blood pressure in SHR.