The uptake of a fluorescently labelled antisense oligonucleotide in vitro and in vivo

The uptake of a fluorescently labelled antisense oligonucleotide in vitro and in vivo

Journal of Neuroscience Methods 147 (2005) 48–54 The uptake of a fluorescently labelled antisense oligonucleotide in vitro and in vivo Emma S.J. Robi...

178KB Sizes 1 Downloads 87 Views

Journal of Neuroscience Methods 147 (2005) 48–54

The uptake of a fluorescently labelled antisense oligonucleotide in vitro and in vivo Emma S.J. Robinson a,∗ , David J. Nutt b , Helen C. Jackson c,1 , Alan L. Hudson b b

a Department of Pharmacology, School of Medical Sciences, University Walk, Clifton, Bristol BS8 1TD, UK Psychopharmacology Unit, University of Bristol, Dorothy Hodgkin Building, Whitsun Street, Bristol BS8 1TD, UK c Knoll Pharmaceuticals Research and Development, Nottingham NG1 1GF, UK

Received 10 November 2004; received in revised form 8 March 2005; accepted 9 March 2005

Abstract Antisense oligonucleotides have been used to target a range of different gene products in the CNS including neurotransmitter receptors. Previous studies using antisense oligonucleotides to target the rat ␣2A/D -adrenoceptor revealed changes in receptor expression in specific brain areas following i.c.v. administration but no reduction was observed following antisense treatment in primary cortical neurones. In order to resolve these discrepant results, the uptake and distribution of the antisense sequence has been determined. In vivo, the fluorescent signal was detected close to the site of injection (2–3 mm) and on the same side of the brain as the injection. Although the oligonucleotides (ODN) were distributed throughout the CSF, the ODN was not widely distributed within the mid or hindbrain parenchyma. In vitro uptake studies revealed the antisense was poorly taken up into primary cortical neurones but a higher level of fluorescence was detected in a small sub-population of cells. These studies demonstrate that antisense is rapidly taken up into cells in vivo but poorly taken up into primary cortical neurones in culture. These data provide further evidence for the uptake and distribution of antisense oligonucleotides in neuronal tissue in vivo. © 2005 Elsevier B.V. All rights reserved. Keywords: Antisense; Uptake; Fluorescence microscopy; Primary cortical neurones; In vivo; Rat

1. Introduction Antisense oligonucleotides provide a highly selective and reversible means for inhibiting protein expression and thus, enable specific receptor-mediated functions to be investigated (for review, see Weiss et al., 1997). The use of antisense in the CNS has become a recognised method for the investigation of novel receptor sub-types particularly where highly selective ligands are not available (Robinson et al., 1997). However, there is still limited information available regarding the uptake and distribution of antisense Abbreviations: BV, blood vessel; Cb, cerebellum; CC, corpus callosum; DG, dentate gyrus; FC, frontal cortex; Hippo, hippocampus; LV, lateral ventricle; ODN, oligonucleotides; Sep, septum; Str, striatum; 4V, fourth ventricle ∗ Corresponding author. Tel.: +44 117 928 8666; fax: +44 117 925 0168. E-mail address: [email protected] (E.S.J. Robinson). 1 Present address: RenaSci Consultancy Ltd., Pennyfoot Street, Nottingham, UK. 0165-0270/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jneumeth.2005.03.003

in neuronal tissue. The results from our own experiments using antisense to target the ␣2A/D -adrenoceptor in the rat raised questions about the uptake and distribution of the antisense sequence both in vitro and in vivo. Preliminary studies using primary cortical neurones failed to show an antisense-mediated reduction in binding to ␣2 -adrenoceptors (Robinson et al., 1999a). In contrast, binding studies in vivo showed significant reductions in binding to ␣2 -adrenoceptors in the septum and hypothalamus (Robinson et al., 1999b). Furthermore, reductions of similar magnitude (20%) were also observed in brain regions close to the site of infusion including the rostral hippocampus, thalamus and anterior amygdaloid area. The results from the binding studies suggest a limited distribution of the antisense following i.c.v. administration. Subsequent behavioural studies also suggested the antisense-mediated inhibition in ␣2 -adrenoceptor function was limited to specific brain areas (Robinson et al., 2000). Given the efficacy of these sequences in vivo, the in vitro uptake of the oligonucleotides was in

E.S.J. Robinson et al. / Journal of Neuroscience Methods 147 (2005) 48–54

question following their lack of effect in primary cortical neurones. The aims of the present uptake study were firstly to determine the anatomical distribution of an 18 base phosphorothioate modified antisense following i.c.v. administration and secondly, to determine the uptake of the antisense sequence into primary cortical neurones in culture. The uptake of antisense oligonucleotides into cells, both in culture and in vivo, has previously been investigated, although studies comparing in vitro and in vivo uptake are limited (Yakubov et al., 1989; Agrawal et al., 1991; Akhtar and Juliano, 1992; Yee et al., 1994; Szklarczyk and Kaczmarek, 1995; Yaida and Nowak, 1995). Phosphorothioate oligonucleotides are thought to enter the cells through receptor-mediated endocytosis (Stein et al., 1988), and a number of different methods have been used to detect the cellular uptake of oligonucleotide sequences in vitro and in vivo. These include fluorescent labelling of the oligonucleotide, immunocytochemical assays and the use of radiolabelled oligonucleotides (Whitesell et al., 1993; Tischmeyer et al., 1994; Yaida and Nowak, 1995; Zhang et al., 1996; Schlingensiepen and Klinger, 1997). These methods have all been shown to produce a similar pattern of in vivo uptake and distribution. The fluorescently labelled oligonucleotide is more convenient to use than a radiolabelled probe as it is commercially synthesised and is easier to handle for in vivo experiments. Thus, the modification used for the current studies was a 5 carboxy-fluorescein molecule, which was added to the fully modified phosphorothioate oligonucleotide antisense. The protocols used for these studies were adapted from the literature (Schlingensiepen and Klinger, 1997). Time course studies into the uptake and the stability of phosphorothioate oligonucleotide in vivo suggest the phosphorothioate oligonucleotide is stable in the brain for approximately 1 day (Zhang et al., 1996; Weiss et al., 1997). Therefore, a time point of 10 h was used to identify the pattern of distribution in the brain before the majority of the oligonucleotide had been degraded and distribution of metabolites had occurred, which may have given a false picture of the distribution of the intact oligonucleotide sequence. The same time point was also used to investigate the uptake in vitro.

49

2. Materials and methods 2.1. Synthesis of the fluorescently labelled oligonucleotide The fluorescently labelled oligonucleotide used for the in vitro and in vivo uptake studies was commercially synthesised on an ABI 394 DNA synthesiser by Oswel DNA services (Southampton, UK). The fluorescent modification used was a 5 FAM dye, carboxy-fluorescein (Fig. 1), maximum emission spectra 494 nm and maximum excitation spectrum 520 nm. The uptake study was performed with a fully modified phosphorothioate antisense sequence (Nunes, 1995; Robinson et al., 1999a,b, 2000). 2.2. In vitro administration of the fluorescently labelled oligonucleotide The procedure used for culturing the cortical neurones was adapted from that described by Sumners et al. (1991), which had been shown to result in cultures consisting of ∼90% neurones and ∼10% glial cells. Cortices were obtained from newborn Wistar rat pups of both sexes and the neurones were dissociated by trypsination. The resulting cell suspension was plated on poly-l-lysine (Sigma) pre-coated wells, containing a glass coverslip, at a density of 4 × 106 cells per well. The medium used was Basal Eagle’s medium (Gibco) containing 10% foetal bovine serum (heat inactivated, Gibco), 2 mM l-glutamine (Sigma), 25 mM KCl (Sigma) and 50 U/50 ␮g penicillin/streptomycin (Gibco). 1-Beta-d-arabino-furanosylcytosine (AraC, 1 mM, Sigma) was added to the cells 24 h after plating to curtail the proliferation of non-neuronal cells. The cells were grown for 6 days without a medium change to allow the neuronal cells to fully develop before oligonucleotide treatment. On day 6 after plating, the fluorescent oligonucleotide was added, to give a final concentration of 1 ␮M in the medium. The plates were returned to the incubator for 10 h, then removed and the cells rinsed in ice cold buffer (20 mM HEPES, 146 mM NaCl, 4.2 mM KCl, 0.5 mM MgCl2 ) before mounting onto clean glass slides using fluorescence mounting solution (Vector, UK). The coverslips were fixed in place using

Fig. 1. (A) Structure of carboxy-fluorescein (FAM) dye used to label the 5 end of the antisense oligonucleotide for the uptake study. (B) Representative graph for the absorption and emission spectra for this fluorescent label.

50

E.S.J. Robinson et al. / Journal of Neuroscience Methods 147 (2005) 48–54

clear nail varnish and the cells imaged using a fluorescence microscopy. 2.3. In vivo administration of fluorescently labelled oligonucleotide Male Wistar rats (Bantin and Kingman, UK), weighing 270–310 g, were housed in a group of four on a 12-h light:12-h dark cycle (lights on at 07:00 h) at 21 ± 1 ◦ C and 55% humidity and were allowed free access to food (standard rat diet) and water throughout the experiment. Rats were anaesthetised with sodium pentabarbitone (60 mg kg−1 , i.p.), then placed in a stereotaxic frame (Kopf) with the incisor bar set at −3.5 mm. The surface of the skull was exposed, a small burr hole made at the site of injection, then the dura was punctured. A unilateral injection of the fluorescence labelled oligonucleotide (2 nmol in 2 ␮l) was made into the left lateral ventricle, co-ordinates: 0.92 mm caudal to bregma, 1.4 mm lateral and 3.5 mm below the surface of the dura (Paxinos and Watson, 1983). Injections were made using a Hamilton syringe over a 5 min period and the syringe was left in place for a further 5 min before being slowly withdrawn. The burr hole was then sealed using bone wax and the region closed using suture clips and the animals were allowed to recover. Rats were re-anaesthetised 10 h after the i.c.v. injection with sodium pentobarbitione (60 mg kg−1 , i.p.) and perfused, intra-cardiac, with ice-cold phosphate buffered saline (1 ml g−1 body weight, pH 7.4) via a steel cannulae inserted into the left ventricle of the heart, and the right atria was cut to drain the blood. The brains were carefully removed over ice and immediately frozen in isopentane on dry ice (−40 ◦ C). The tissue was then stored at −70 ◦ C until sectioning. Sections (25 ␮m) were cut at regular intervals throughout the brain using a cryostat at −15 ◦ C and thaw-mounted onto gelatin (0.5% gelatin, 0.05% chromic potassium sulphate in distilled H2 O) subbed glass slides. Sections were fixed in 5% paraformaldehyde (5 min), then dehydrated in alcohol (70–100%) before being mounted in vectashield mounting medium for fluorescent imaging (Vector, UK). The coverslips were fixed in place using clear nail varnish and the sections imaged using a fluorescence microscopy. The position of the cannulae was confirmed during sectioning.

exposure 2.52 s, black level 25, white level 85 and control box set to 1.15. The images were saved directly onto disk in the format of pict files. Results from the in vitro studies were collected from four separate cultures treated with the fluorescent oligonucleotides. Each separate culture consisted of one, six-well plate of culture neurones and each well was examined and the density and number of stained cells was scored by eye. Representative images of the neuronal and non-neuronal cells were also acquired and saved as pict files. A similar process was used for the in vivo experiments, however, each section was visualised and the individual brain regions were scored as fluorescent or non-fluorescent. The sections were collected at regular intervals through the brain to facilitate estimation of the distance of the fluorescent signal from the site of administration. Representative images were also acquired and saved as pict files.

3. Results 3.1. In vitro uptake into primary cortical neurones Representative pictures for the uptake of the fluorescently labelled phosphorothioate antisense, into primary cortical neurones, are shown in Fig. 2. The results for the uptake study are qualitative and these pictures are examples of typical results for uptake from four separate cultures. The uptake into primary cortical neurones 10 h after administration of the oligonucleotide was poor with only low levels of fluorescence observed in the majority of cells present on the slide. Three slides from each culture were examined and the uptake was consistent throughout. In each case, a small number of cells appeared to show a higher level of uptake (Fig. 2) with the fluorescent signal being particularly strong from the nucleus of the cell. The number of cells showing this level of uptake corresponds with the proportion of glial cells present with this culture protocol. Furthermore, the morphology of the cells was more like a glial population than cortical neurones. The uptake into the nucleus also shows the fluorescence was evident within the cell and not a result of the oligonucleotide sticking to the cell membrane. 3.2. In vivo uptake

2.4. Fluorescence microscopy The fixed, mounted sections from the in vivo uptake and the fixed, mounted coverslips from the in vitro uptake study were both imaged using the Improvision Openlab system provided by the Cell Imaging Facility at the University of Bristol. The microscope used was a Leica DM IRBE inverted epifluorescence microscope with phase contrast (excitation 450–490 nm, emission 515). Images were acquired with a cooled integrating CCD camera (Photonic Sciences Coolview) using software for image acquisition on an Apple Macintosh computer. The settings for the Openlab programme were kept the same for all the images acquired,

Representative pictures from a selection of brain areas are shown in Fig. 3. The fixation and dehydration steps were used to limit redistribution of the oligonucleotide following thaw mounting. Although some fluorescent oligonucleotide may have leaked out when the sections were cut and thawed, the images only show cellular uptake and do not suggest a significant effect of ‘leaked’ oligonucleotide. The fluorescent signal of the phosphorothioate oligonucleotide was detected for 2–3 mm into the brain tissue surrounding the site of the injection. The results show the distribution of the oligonucleotide is pre-dominantly on the left side of the brain, the same side of the infusion with a weaker signal detected on

E.S.J. Robinson et al. / Journal of Neuroscience Methods 147 (2005) 48–54

51

Fig. 2. Representative images showing the uptake of the fluorescently labelled phosphorothioate antisense, probe 1, into primary cortical neurones. The cells were treated with 1 ␮M oligonucleotide and fixed 10 h later. The top figure shows the low level of fluorescence observed in cortical neurones. A small number of cells within the culture appeared to display a higher level of fluorescence as shown in the bottom image, labelled A. The number of cells showing this level of uptake corresponds with the 10% population of glial cells predicted for this culture protocol.

the right side of the brain. The eppendymal cells of both the lateral ventricles had an intense fluorescent signal indicating the uptake of the oligonucleotide into these cells. On the left side of the brain, fluorescence within the eppendymal cells was detected up to 3 mm into the forebrain from the site of injection. Fluorescence was also detected around the lateral ventricle, the septum, striatum, hypothalamus and hippocampus on both sides of the brain. However, the signal from the right side of the brain was in general relatively weak compared to the left side of the brain where these brain areas showed a strong fluorescent signal. The uptake appeared to be relatively evenly distributed around the site of the injection with the signal weakening both medio-laterally as well as rostro-caudally from the injection site. The uptake in the septum, striatum and frontal cortex as well as the corpus callosum on the left side of the brain was granular in appearance and fluorescence was detected within the cells when viewed under higher magnification (data not shown). The cell types taking up the fluorescent oligonucleotide were not determined in this study. Of interest was the similarity in fluorescence seen on the left and the right side of the brain in the hypothalamic area particularly when

all the other brain areas examined showed a higher level of fluorescence on the left than the right side of the brain. The strong fluorescent signal in the hippocampus showed evidence of differential uptake within the hippocampal cell layers (Fig. 3). The aqueduct showed a very weak fluorescent signal and, similarly a weak signal was seen in the third and fourth ventricle but not in the surrounding tissue including the locus coeruleus and the Edinger–Westphal nucleus (Fig. 3). A strong fluorescent signal was also detected in the blood vessels surrounding the brain (Fig. 3).

4. Discussion The results from the in vitro uptake studies showed that the phosphorothioate oligonucleotide antisense to the ␣2A/D -adrenoceptor was poorly taken up by primary cortical neurones in culture. In contrast, glial cells, identified by their morphology and number, present in the primary cortical neurone culture showed a higher level of uptake of the fluorescent oligonucleotide, with a strong signal detected.

52

E.S.J. Robinson et al. / Journal of Neuroscience Methods 147 (2005) 48–54

Fig. 3. These figures are representative images showing the uptake and distribution of the antisense following i.c.v. administration. Panels (A) and (B) represent the fluorescent signal detected on the left side of the brain approximately +0.7 mm and +0.3 mm AP to bregma. Panels (C) (inset represents fluorescence detected in the dentate gyrus) and (D) show corresponding images from the left and right side of the hippocampus, respectively, sections taken from the CA1 region approximately −2.5 mm AP to bregma. Panel (E) represents the fluorescent signal detected in the region of the fourth ventricle approximately −10 mm AP to bregma. Panel (F) shows the strong fluorescent signal detected in the vasculature surrounding the brain. Abbreviations: Sep, septum; LV, lateral ventricle; Str, striatum; FC, frontal cortex; CC, corpus callosum; Hippo, hippocampus; DG, dentate gyrus; BV, blood vessel; Cb, cerebellum; 4V, fourth ventricle.

The poor uptake of the antisense in primary cortical neurones is likely to be a major factor in the lack of effect seen with this antisense sequence in vitro (Robinson et al., 1999b). This result may be as a consequence of the culture condition although the uptake by glial cells did not appear to be affected. The culture conditions, particularly the high concentration of potassium ions, may affect the behaviour of the cell membrane and therefore uptake. However, previous studies using primary cortical neurone cultures have shown successful antisense-mediated knockdown of NPY-Y1 and NMDA receptor binding (Wahlestedt et al., 1993a,b). The presence of AraC in the culture medium, to inhibit the proliferation of non-neuronal cells, has also been proposed to interact with antisense molecules (Manev et al., 1991) and this may affect the uptake of the oligonucleotide. In vivo, the oligonucleotide was taken up in the brain parenchyma within a 2–3 mm radius of the site of the injection. A strong fluorescent signal was seen in brain structures close to the site of the injection and on the same side as the antisense was administered. With the exception of brain regions close to the lateral ventricles there was limited fluorescent signal detected in the parenchyma surrounding the third and fourth ventricle. The uptake of the antisense sequence, in vivo, resulted in a similar pattern of distribution to that previously observed following i.c.v. administration of

a radiolabelled phosphorothioate oligonucleotide (Yee et al., 1994; Yaida and Nowak, 1995; Zhang et al., 1996). The flow of cerebral spinal fluid from the lateral ventricles through the third and fourth ventricle is likely to impact on the uptake of the antisense. Although these data suggest that the CSF distributes the antisense, the lack of uptake into the tissue surrounding the fourth ventricle suggest the antisense is rapidly removed from the CSF. Uptake into the hypothalamic regions, however, may be associated with the movement of the antisense from the lateral ventricle to the third ventricle. The brain areas surrounding the left lateral ventricle, the septum, striatum and hippocampus showed dense fluorescent staining. On the right side of the brain, there was evidence of uptake into the septum but only a small amount of fluorescence was seen in the right striatum. There was also evidence of uptake in the frontal cortex and the corpus callosum particularly on the left side of the brain. This would suggest the corpus callosum was not a barrier to the distribution of the oligonucleotide following i.c.v. administration. A study by Sommer et al. (1993) using local administration of antisense to c-fos into the striatum demonstrated uptake, with a sharp concentration gradient in the area of the neostriatum but the corpus callosum appeared to be a barrier to further diffusion. These differences highlight the importance of investigating the uptake of individual antisense sequences, however, these

E.S.J. Robinson et al. / Journal of Neuroscience Methods 147 (2005) 48–54

studies are limited by the use of single injection vs. a continuous infusion as well as an effect of the dose used. In the present study, the uptake of antisense probe was identified in the brain following a single bolus dose, i.c.v. However, analysis of receptor expression, physiological and behavioural measurement of antisense to the ␣2A/D adrenoceptor used a continuous infusion (Robinson et al., 1999b, 2000). The single bolus dose was used to give a picture of the regional distribution of the oligonucleotide before any degradation had occurred. The effect of route of administration on the distribution has previously been investigated and evidence from uptake studies has shown a continuous infusion may give a more uniform distribution (Whitesell et al., 1993). A continuous infusion may have given a better picture of the distribution of the oligonucleotide but would have been complicated by incomplete uptake as well as the increased possibility of re-distribution of the oligonucleotide post mortem. Preliminary studies also suggested that a significant time after antisense administration was required to provide clear images of the tissue distribution. At 6 h after a single i.c.v. injection, the antisense was localised within the tissue as described in the present study but the signal was diffuse and did not show clear cellular uptake (data not shown). The distribution of this antisense sequence targeting the ␣2A/D -adrenoceptor, corresponds with the findings from receptor expression and functional studies (Robinson et al., 1999b, 2000). For instance, the lack of effect of the antisense to the ␣2A/D -adrenoceptor on binding to the ␣2 adrenoceptors, determined using saturation binding to whole brain membranes, might be due to the limited distribution of the oligonucleotide (Robinson et al., 1996). In this context, although receptor autoradiography, using a single concentration of radioligand, does not give a measure of the maximal binding capacity following treatment, it is more suitable for examining binding to discrete brain areas. The results from autoradiography studies with [3 H]RX821002 following antisense treatment resulted in a specific knockdown in binding in the septum and hypothalamus (Robinson et al., 1999b). Both these brain areas show a strong fluorescent signal, which was greater on the left side of the brain particularly in the septum. These brain regions were the only areas where a significant knockdown in receptor expression was detected; however, other brain areas exhibited knockdown of a similar magnitude but with greater variability. Despite the functional evidence, changes in binding density in the striatum and frontal cortex were not observed and only small reductions in the hippocampus were detected. Although the oligonucleotide has been shown to reach these brain areas and is therefore predicted to inhibit ␣2A/D -adrenoceptor expression as seen in the septum, this study does not show the concentration of oligonucleotide that is required to achieve knockdown. The overall lack of effect observed may be a consequence of submaximal levels of uptake following i.c.v. administration of the antisense. Previous studies have shown that a maximal knockdown of 20–30% can be achieved following antisense administration resulting in a 90% loss of function (Weiss et

53

al., 1997). This is consistent with previous findings using the antisense sequence described in the present investigation (Robinson et al., 1999b, 2000). These discrepant results have been attributed to the presence of two distinct pools of receptors, a functional pool that is rapidly turned over and affected by antisense treatment and, a reserve pool which is non-functional but detected using receptor binding studies (Qin et al., 1995). The combination of the variability seen with the autoradiography technique, a sub-maximal knockdown and labelling of all ␣2 -adrenoceptor sub-types (Hudson et al., 1992; Nicholas et al., 1996) may have contributed to the failure to detect significant reductions in protein expression in all brain areas showing uptake. In summary, these studies demonstrate that significant differences in the uptake of antisense oligonucleotides exist between isolated neurones in culture and neurones in vivo. This uptake study has shown that the antisense probe is not taken up by primary cortical neurones in culture, however, one approach to facilitate uptake into primary neuronal cultures may be to couple the antisense to cationic lipids. The oligonucleotide is distributed within those brain structures close to the site of infusion including the septum and hypothalamus where significant reductions in binding were observed. One advantage of the limited distribution of the antisense probe is that it can be used to target specific brain areas. Therefore, in addition to selective inhibition of the expression of a specific protein, antisense can be targeted to discreet brain regions allowing for functional characterisation. Widespread distribution of the antisense sequence was not observed following i.c.v. administration, however, this study was limited and effects of dose or comparison between a single injection versus a continuous infusion were not carried out.

Acknowledgements The Biotechnology and Biological Science Research Council (BBSRC) and Knoll Pharmaceuticals provided funding for this research.

References Agrawal S, Temsamani J, Jin YT. Pharmacokinetics, biodistribution, and stability of oligodeoxynucleotide phosphorothioates in mice. Proc Natl Acad Sci USA 1991;88:7595–9. Akhtar S, Juliano RL. Cellular uptake and intracellular fate of antisense oligonucleotides. Trends Cell Biol 1992;2:139–44. Hudson AL, Mallard NJ, Tyacke R, Nutt DJ. [3 H]-RX821002: a highly selective ligand for the identification of ␣2 -adrenoceptors in the rat brain. Mol Neuropharmacol 1992;1:219–29. Manev H, Caredda S, Grayson DR. Nonselective inhibition by antisense oligonucleotides of cytosine arabinoside action. Neuroreport 1991;2:589–92. Nicholas AP, Hokfelt T, Pieribone VA. The distribution and significance of CNS adrenoceptors examined with in situ hybridization. Trends Pharmacol Sci 1996;17:245–55.

54

E.S.J. Robinson et al. / Journal of Neuroscience Methods 147 (2005) 48–54

Nunes JP. Central ␣2 -adrenoceptors and blood pressure regulation in the rat. Eur J Pharmacol 1995;278:183–5. Paxinos G, Watson C. The rat brain stereotaxic coordinates. New York: Academic Press; 1983. Qin Z-H, Zhou L-W, Zhang S-P, Wang Y, Weiss B. D2 dopamine receptor antisense oligodeoxynucleotide inhibits the synthesis of a functional pool of D2 dopamine receptors. Mol Pharmacol 1995;48:730–7. Robinson ESJ, Nutt DJ, Hall L, Jackson HC, Hudson AL. In vitro and in vivo effects of antisense on ␣2 -adrenoceptor expression. In: Phillips MI, editor. Methods in enzymology: antisense technology, Part B, applications, vol. 314. San Diego: Academic Press; 1999a. p. 61–76. Robinson ESJ, Nutt DJ, Hall L, Jackson HC, Hudson AL. Autoradiographical and behavioural effects of a chronic infusion of antisense to the ␣2D -adrenoceptor in the rat. Br J Pharmacol 1999b;128:515–22. Robinson ESJ, Nutt DJ, Jackson HC, Hudson AL. ␣2 -Adrenoceptor binding in rat brain following acute i.c.v. antisense to the ␣2D subtype. J Psychopharmacol 1996;10(Suppl.):A16. Robinson ESJ, Nutt DJ, Jackson HC, Hudson AL. Antisense oligonucleotides in psychopharmacology and behaviour: promises and pitfalls. J Psychopharmacol 1997;11:259–69. Robinson ESJ, Nutt DJ, Jackson HC, Hudson AL. Investigation of the behavioural and physiological effects induced by an infusion of antisense to ␣2D -adrenoceptors in the rat. Br J Pharmacol 2000;130:153–9. Schlingensiepen R, Klinger I. Cell culture protocols. In: Schlingensiepen R, Brysch W, Schlingensiepen K-H, editors. Antisense—from technology to therapy. Lab manual and textbook. Berlin: Blackwell Science; 1997. p. 127. Sommer W, Bjelke B, Ganten D, Fuxe K. Antisense oligonucleotide to c-fos induces ipsilateral rotational behaviour to d-amphetamine. Neuroreport 1993;5:277–80. Stein CA, Mori K, Loke SL, Subasinghe C, Shinozuka K, Cohen JS, et al. Phosphorothioate and normal oligodeoxyribonucleotides with 5 linked acridine: characterization and preliminary kinetics of cellular uptake. Gene 1988;72:333–41. Sumners C, Tang W, Zelezna B, Raizada MK. Angiotensin II receptor subtypes are coupled with distinct signal-transduction mechanisms in

neurones and astrocytes from rats brain. Proc Natl Acad Sci USA 1991;88:7567–71. Szklarczyk A, Kaczmarek L. Antisense oligodeoxyribonucleotides: stability and distribution after intracerebral injection into rat brain. J Neurosci Meth 1995;60:181–7. Tischmeyer W, Grimm R, Schichnick H, Brysch W, Schlingensiepen K-H. Sequence-specific impairment of learning by c-jun antisense oligonucleotides. Neuroreport 1994;5:1501–4. Wahlestedt C, Golanov E, Yamamoto S, Yee F, Ericson H, Yoo H, et al. Antisense oligodeoxynucleotides to NMDA-R1 receptor channel protect cortical neurones from excitotoxicity and reduce focal ischaemic infarctions. Nature 1993a;363:260–3. Wahlestedt C, Merlo-Pich E, Koob GF, Yee F, Heilig M. Modulation of anxiety and neuropeptide Y-Y1 receptors by antisense oligodeoxynucleotides. Science 1993b;259:528–31. Weiss B, Davidkova G, Zhang S-P. Antisense strategies in neurobiology. Neurochem Int 1997;31:321–48. Whitesell L, Geselowitz D, Chavany C, Fahmy B, Walbridge S, Alger JR, et al. Stability, clearance and deposition of intraventricularly administered oligodeoxynucleotides: implications for therapeutic application within the central nervous system. Proc Natl Acad Sci USA 1993;90:4665–9. Yaida Y, Nowak TS. Distribution of phosphodiester and phosphorothioate oligonucleotides in rat brain after intraventricular and intrahippocampal administration determined by in situ hybridisation. Regul Pept 1995;59:193–9. Yakubov LA, Deeva EA, Zarytova VF, Ivanova EM, Ryte AS, Yurchenko LV, et al. Mechanism of oligonucleotide uptake by cells: involvement of specific receptors? Proc Natl Acad Sci USA 1989;86:6454–8. Yee F, Ericson H, Reis D, Wahlestedt C. Cellular uptake of intracerebroventricular administered biotin- or digoxigenin-labelled antisense oligodeoxynucleotides in the rat. Cell Mol Neurobiol 1994;14:475– 85. Zhang S-P, Zhou L-W, Morabito M, Lin RCS, Weiss B. Uptake and distribution of fluorescein-labelled D2 dopamine antisense oligodeoxynucleotide in mouse brain. J Mol Neurosci 1996;7:13–28.