In situ hybridization histochemistry and immunocytochemistry were used to examine GABABR1a,b receptor mRNA and protein expression in areas 17 and 18 of the visual cortex of normal macaque monkeys and of monkeys that had been deprived of vision in one eye. In addition, the normal patterns of GABABR1a,b protein expression were immunocytochemically studied in the human visual cortex. Overall levels of GABABR1a,b transcript were higher in area 17 than in area 18. In area 17 GABABR1a,b mRNA levels were highest in layers IVC and VI, moderate in layers II–IVA and low in layers I, IVB and V. In area 18 GABABR1a,b transcript expression was high in layers II and III, moderate in layers IV and VI and low in layers I and V. Immunocytochemistry revealed nearly identical patterns of GABABR1a,b protein expression in areas 17 and 18 in monkey and human. Both pyramidal and non-pyramidal neurons were GABABR1a,b immunoreactive. The majority of intensely immunoreactive neurons in layers II, III, V and VI were pyramidal cells. Numerous non-pyramidal cells were intensely immunoreactive in layer IV of area 17 but layer IV cells were only lightly immunoreactive in area 18. Following 10 day periods of monocular deprivation, induced by intravitreal injections of tetrodotoxin, levels of GABABR1a,b mRNA and protein were decreased in the deprived eye dominance columns of layers IVC and VI.
γ-Aminobuytyric acid (GABA) is the main inhibitory neurotransmitter in the central nervous system. It elicits fast IPSPs by acting on ionotropic GABAA and GABAC receptors (Olsen and Tobin, 1990; Misgeld et al., 1995; Johnston, 1996a,b) and slower and long-lasting inhibitory effects by acting on metabotropic (G protein-coupled) GABAB receptors (Hill and Bowery, 1981; Bowery et al., 1983, 1984; Bowery, 1989). Pre- and post-synaptic GABAB receptor subtypes exist with differences in pharmacological properties (Dutar and Nicoll, 1988; Bonanno and Raiteri, 1992; Waldmeier et al., 1994; Misgeld et al., 1995; Bonanno et al., 1997; Deisz et al., 1997). Pre-synaptic GABAB receptors inhibit neurotransmitter release by inhibiting high voltage gated Ca2+ channels, whereas post-synaptic GABAB receptors inhibit adenylyl cyclase and decrease neuronal excitability by activating inwardly rectifying K+ currents through Kir3 channels, leading to late inhibitory post-synaptic potentials (Gahwiler and Brown, 1985; Andrade et al., 1986; Misgeld et al., 1995; Lüscher et al., 1997; Jones et al., 1998; Kaupmann et al., 1998; Uezono et al., 1998; White et al., 1998; Kuner et al., 1999).
Two subtypes of GABAB receptor have been cloned to date: GABABR1, with four differentially spliced variants (GABABR1a–d) and GABABR2 (Kaupmann et al., 1997, 1998; Bowery, 1997; Bowery and Brown, 1997; Isomoto et al., 1998; White et al., 1998). Heterodimeric assembly of both receptor subtypes, with a 1:1 protein stoichiometry, seems to be critical for maturation and transport of GABABR1 proteins to the plasma membrane, where they form fully functional GABAB receptors able to generate late inhibitory post-synaptic potentials (Jones et al., 1998, Kaupmann et al., 1998; White et al., 1998; Kuner et al., 1999; Bowery and Enna, 2000). Accordingly, GABABR1a and GABABR1b transcripts co-localize with GABABR2 transcripts, showing similar distribution patterns in the rat, and it is likely that the proteins are co-expressed in individual neurons (Jones et al., 1998; Kaupmann et al., 1998; White et al., 1998; Kuner et al., 1999).
In various regions of the cerebral cortex of different mammalian species GABAA- and GABAB-mediated inhibitory effects have been shown to play critical roles in controlling neuronal excitability, in determining stimulus–response properties and in shaping neuronal receptive fields (Sillito, 1975; Sillito et al., 1980; Hicks and Dyckes, 1983; Bolz and Gilbert, 1986; Alloway and Burton, 1986, 1991; Alloway et al., 1989; Crook and Eysel, 1992; McCormick, 1992). The distribution patterns of GABAA receptors and of their constitutive subunits have been extensively studied with a variety of techniques in different cortical regions and mammalian species (Shaw and Cynander, 1986; Rakic et al., 1988; Hendry et al., 1990; Shaw et al., 1991; Meinecke and Rakic, 1992; Huntsman et al., 1994; Huntsman and Jones, 1998). However, knowledge of the distribution of GABAB receptors or their RNAs in the cerebral cortex is fragmentary and limited to general descriptions of the whole rat brain studied by receptor autoradiography (Gehlert et al., 1985; Bowery et al., 1987; Chu et al., 1990; Turgeon and Albin, 1994), in situ hybridization histochemistry (GABABR1a,b) (Kaupmann et al., 1997) or by western blotting and immunocytochemistry (Margeta-Mitrovic et al., 1999). Nothing is known of their areal or cellular distribution in the neocortex of primates.
In the monkey visual cortex expression of glutamic acid decarboxylase (GAD) (Hendry and Jones, 1986, 1988; Hendry et al., 1987; Benson et al., 1991, 1994) and of certain GABAA receptor subunits is regulated in an activity-dependent manner (Huntsman et al., 1994; Hendry et al., 1990, 1994; Jones, 1997; Huntsman and Jones, 1998). Pre- and post-synaptic GABAB receptors affect activity-dependent changes in synaptic efficacy such as LTP (Davies et al., 1990, 1991; Mott and Lewis, 1991; Olpe et al., 1993) and GABABR1a,b gene expression is regulated by retinal activity in the lateral geniculate nucleus (Muñoz et al., 1998). Therefore, activity-dependent regulation of GABABR1a,b gene expression could contribute to cortical plasticity. In the present study we examined areas 17 and 18 of the macaque and human visual cortex for the normal distribution patterns of GABABR1a,b mRNA and protein expression, using in situ hybridization and immunocytochemistry, and studied the effects of monocular deprivation on area 17 of the macaque.
Materials and Methods
This work was carried out on the brains of four macque monkeys (two Macaca mulata and two Macaca fuscata), aged 2 years or older, and on human cortex obtained at autopsy (2–3 h post-mortem delay) from two individuals, aged 35 and 40 years, with no history of neurological or psychiatric diseases. All animal procedures were approved by the Institutional Animal Care and Use Committees. Two macaques had been subjected to monocular deprivation for 10 days prior to death by intravitreal injections of 15 μg tetrodotoxin (TTX) in 10 μl of normal saline at 4 day intervals. TTX, a sodium channel blocker, silences retinal ganglion cell activity and blocks action potentials in the optic nerve, resulting in functional deafferentation of related laminae of the dorsal lateral geniculate nucleus and ocular dominance columns of the visual cortex, identical to the effects of eyelid closure or eye removal (Hendry and Jones, 1986, 1988). All animals were given an overdose of Nembutal (Abbot, Chicago, IL) and perfused through the heart with normal saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. The brains were post-fixed overnight in 4% paraformaldehyde, cryoprotected in 30% sucrose, blocked and frozen in dry ice. Blocks of the visual cortex were cut in the frontal plane at 25 μm on a sliding microtome and serial sections were collected in 0.1 M phosphate buffer for Nissl or immunocytochemical staining or collected and kept for 48 h in sterile 4% paraformaldehyde in phosphate buffer for in situ hybridization histochemistry. The post-mortem human samples were kindly supplied by Dr A. Gómez (Department of Anatomy, Institute of Toxicology, Madrid). After removal of a portion of the occipital lobe the blocks were immediately immersed in a cold solution of 4% paraformaldehyde in phosphate buffer and sectioned coronally in 1.5 cm thick slides. Small blocks of the visual cortex were taken and transferred to a second fixative solution of 4% paraformaldehyde in 0.1 M phosphate buffer. The total time of fixation was ~24 h. Afterwards the blocks were cut serially at 100 μm on a vibratome and the sections were pre-treated for 30 min with a solution of ethanol and H2O2 in phosphate buffer to remove endogenous peroxidase activity, washed in phosphate buffer and then processed for immunocytochemistry as indicated below.
In situ hybridization histochemistry
For in situ hybridization histochemistry sections were pre-washed in 0.1 M glycine in phosphate buffer at room temperature and then incubated in 1 mg/ml proteinase K in phosphate buffer, pH 8, for 20 min at 37°C. All solutions were prepared under sterile conditions. After washing in 0.25% acetic anhydride in 0.1 M triethanolamine in phosphate buffer, pH 8.0, and in 2× SSC, pH 7 (1× SSC consists of 0.88% NaCl and 0.44% Na3C6H5O3•2H2O), sections were incubated in pre-hybridization solution containing 50% formamide, 10% dextran sulfate, 0.7% Ficoll, 0.7% polyvinyl pyrolidone, 0.5 mg/ml yeast tRNA, 0.33 mg/ml denatured herring sperm DNA and 20 mM dithiothreitol (DTT) for 1 h at 60°C. Subsequently, sections were incubated in the same solution, with the addition of 1 × 106 c.p.m./ml of a 33P-labeled antisense or sense cRNA probe specific for the GABABR1a,b mRNA, for at least 20 h at 60°C. Subcloning of the monkey-specific cDNA, the preparation of riboprobes, their specificity and the detailed protocol for in situ hybridization histochemistry have been described elsewhere (Muñoz et al., 1998). Riboprobes (418 bp) complementary to the pan region which identifies alternatively spliced GABABR1a (from 2463 to 2872 bp) and GABABR1b (from 2115 to 2524 bp) were transcribed from monkey cDNA. Following hybridization, sections were washed twice over 10 min in 4× SSC at 60°C, digested with 20 mg/ml ribonuclease A (pH 8.0) for 30 min at 45°C and washed through descending concentrations of SSC with 5 mM DTT to a final stringency of 0.1× SSC at 60°C for 1 h. Sections were mounted on gelatin-coated glass slides, allowed to dry and then exposed to Amersham β Max film for 4 days at 4°C. After development of the film, sections were dipped in Kodak NTB2 emulsion and exposed for 4 weeks at 4°C. The autoradiographs were then developed in Kodak D19, fixed in Kodak Rapid Fixer, washed in water, dried overnight, lipid-extracted in a 1:1 mixture of 100% chloroform and 100% ethanol, rehydrated through descending concentrations of ethanol, stained through the emulsion with cresyl violet, dehydrated, cleared in xylene and coverslipped in DPX. Film autoradiograms of sections were quantified by densitometry, using a microcomputer imaging device (MCID/M4; Imaging Research, St Catharine's, Ontario, Canada). Repeated optical density readings were taken in scans of defined width across cortical layers or across the extent of the tangentially sectioned layer IVC. Areal boundaries and cortical layers were identified in scans taken from adjacent Nissl stained sections. Ocular dominance columns within layer IVC of the visually deprived monkeys were identified by reference to adjacent sections stained for cytochrome oxidase (CO) in which alternating stripes of dense and light CO staining represent undeprived and deprived ocular dominance columns, respectively. Absolute values of radioactivity in nCi/g were obtained from the integrated optical density readings by reference to 14C plastic standards (Amersham) exposed on the same sheet of film. Sense controls revealed low background labeling only. Differences in expression levels were tested for significance by Student's t-test.
Both monkey and human cortex was processed identically. Free floating sections were pre-incubated for 1 h in phosphate buffer containing 3% normal goat serum, 1% bovine serum albumin and 0.25% Triton X-100 and then incubated for 24–72 h at 4°C in the same solution containing guinea pig anti-GABABR1a,b (1:2500; Chemicon, Temecula, CA). The antibodies used are directed against a C-terminus peptide present in both the GABAB1a and GABAB1b splice variants (Magreta-Mitrovic et al., 1999). Sections were then washed three times over 10 min in phosphate buffer and subsequently incubated in a 1:200 solution of secondary goat anti-guinea pig biotinylated antibodies (Vector, Burlingame, CA), for 1–2 h at room temperature. Sections were washed in phosphate buffer three times for 10 min each and then processed by the avidin–biotin peroxidase method, using Vectastain ABC Elite kits. Sections were then reacted histochemically with 0.05% 3,3′-diaminobenzidine 4HCl (DAB) (Sigma, St Louis, MO) and 0.01% H2O2 in phosphate buffer for 10–15 min. After rinsing, sections were mounted on gelatin-coated glass slides, dried overnight, dehydrated through ascending ethanols, cleared in xylene and coverslipped in DPX. Controls consisted of staining selected sections after replacing the primary antibody with pre-immune goat serum (1:2500) or after omission of either the primary or secondary antibody. These procedures revealed no specific staining.
Normal Laminar Distribution of GABABR1a,b mRNA in Monkey Visual Cortex
Overall levels of GABABR1a,b transcript expression were very high in visual cortex but displayed distinctive laminar patterns in areas 17 and 18 that delineated the sharp border between these areas (Figs 1–3). GABABR1a,b transcript expression was higher in area 17 than in area 18 (Figs 1, 2), reflecting the greater cell density in area 17 (Fig. 3). In area 17 GABABR1a,b mRNA levels were highest in two bands corresponding to layers IVCβ and VI, separated by a band of low expression in layer V (Figs 1A–D, 3). There was also a wide band of enhanced GABABR1 expression corresponding to layers II–IVA and a second corresponding to layer IVCα, while trancript expression was low in layer IVB and weakest in layer I and the white matter (Figs 1A–D, 3). In area 18 GABABR1a,b transcripts were expressed at the highest levels in layers II and III (Figs 1A,B,E,F, 3). Layers IV and VI showed moderate levels, while the lowest transcript expression was found in layers I and V (Figs 1A,B,E,F, 3). Microscopic examination of emulsion-dipped autoradiograms revealed silver grains representing hybridization of cRNA probe overlying numerous, large Nissl stained profiles corresponding to the nuclei of neurons (Fig. 4). Both large (10–20 μm) and small (5–10 μm) neurons were labeled, suggesting expression in pyramidal cells and interneurons. There was no labeling of neuroglial cells in the cortex or in the white matter (Fig. 4).
Normal GABABR1a,b Protein Expression in Monkey and Human Visual Cortex
The patterns of GABABR1a,b immunostaining were similar to the mRNA expression patterns (Figs 5, 6) and very similar in monkey and human visual cortex (Figs 5, 6). In both areas 17 and 18 GABABR1a,b immunostaining was found mainly in neuronal somata and proximal dendrites. In both areas GABABR1a,b immunostaining followed the laminar pattern of these areas and the border between them was clear (Fig. 6A,B). In area 17 the intensity of GABABR1a,b immunostaining was very high in layer IVC (especially IVCβ) and layer VI, moderately high in layers III and IVA and relatively lower in layers I, II, IVB and V (Figs 5B, 6B,C). No GABABR1a,b immunoreactive neurons were found in layer I. Layer II contained a moderate number of GABABR1a,b-positive somata, including small round neurons and small and medium sized pyramidal cells with weakly stained apical dendrites. Layer III was characterized by the presence of darkly immunostained medium sized pyramidal cells and lightly stained round cells. Layer IVA showed a sparse population of large and medium sized, intensely immunoreactive cells, many of them identifiable as pyramidal cells by the possession of apical dendrites; others lacking apical dendrites were probably the large stellate cells of this layer (Cajal, 1899). The latter cells were more numerous and more intensely stained in the monkey than in the human area 17 (Figs 5E, 6E). Layer IVB was characterized by a sparse population of medium sized, lightly immunoreactive neurons (Figs 5E, 6E). Layer IVCα showed a dense population of small and medium sized round, intensely stained neurons, which included both pyramidal and non-pyramidal cells. Layer IVCβ showed the highest density of GABABR1a-b immunostaining, mainly in small neurons (Fig. 6E). Layer V showed a large population of small, round and weakly stained neurons (Fig. 6F), a few darkly stained medium sized and large pyramidal cells, including the large and sparsely distributed pyramidal cells located at the border between layers V and VI (Fig. 6F). Layer VI displayed a high density of small and medium sized immunoreactive pyramidal neurons and a lesser number of round and irregular shaped cells (Fig. 6F).
In area 18 the most intense GABABR1a,b immunostaining was found in the deep aspect of layer III (Figs 5D, 6B,D) This was due mainly to the presence of darkly stained large pyramidal neurons (Figs 5F, 6G). Layers II, IV and V showed a low intensity of GABABR1a,b immunostaining (Figs 5D, 6D). No GABABR1a,b immunoreactive cells were found in layer I (Fig. 6D). In layer II and the upper part of layer III relatively few lightly stained cells were present, including pyramidal and round neurons (Figs 5D, 6D). Layer IV showed a high density of faintly GABABR1a,b-positive, small round neurons (Figs 5F, 6G) and a small number of larger and more intensely stained somata, including some pyramidal cells. Layer V exhibited a sparse population of GABAB-R1a,b immunoreactive cells, including small round somata and pyramidal cells of varying sizes with well-stained apical dendrites (Figs 5D, 6D). Layer VI showed a large population of moderately immunostained medium sized cells, including both pyramidal and round neurons (Figs 5D, 6D).
Effects of Monocular Deprivation on GABABR1a,b Expression in the Macaque Visual Cortex
Monocular deprivation for 10 days was clearly associated with reduced levels of GABABR1a,b mRNA in the deprived ocular dominance columns of area 17 (Figs 7–99). No changes were detected in area 18. Alternating light and dark autoradiographic stripes were revealed in sections through layers IVCβ and VI hybridized with the GABABR1a,b antisense riboprobes (Figs 7B, 8A). The effects were most evident in layer IVCβ (Fig. 7B). The alignment of autoradiograms with adjacent CO stained sections revealed that the more lightly labeled autoradiographic stripes corresponded to the weakly CO stained, deprived eye dominance columns, whereas the high density stripes corresponded to the densely CO stained, non-deprived columns (Figs 7A,B, 8A). The stripes of dense GABABR1a,b mRNA expression were slightly wider than the weakly hybridized deprived stripes (Figs 7B, 8A). Optical density measurements through layers IV and VI showed that, compared with the non-deprived columns, mRNA levels in the deprived columns were significantly reduced in layer IV and in layer VI (Fig. 8B). No significant changes in GABABR1a,b mRNA expression levels were found in layers II and III; occasional irregularities in the uniformity of expression in these layers were inconsistent and not coincident with CO stained blobs revealed in these layers in adjacent CO stained sections. Immunoreactivity was also affected by monocular deprivation (Fig. 7C,D). Deprivation-induced reductions in GABABR1a,b immunostaining were most evident in sections through layer IVC cut parallel to the pial surface (Fig. 7D). As with mRNA labeling, lightly immunostained stripes co-extensive with weakly CO stained stripes corresponded to the deprived eye columns (Fig. 7C,D). The reductions in GABABR1a,b immunostaining were due to decreases in immunostaining of both somata and neuropil within the deprived columns. No overt differences in cell density or intensity of Nissl staining could be identified in the corresponding parts of adjacent Nissl stained sections.
The present study is the first description of the GABABR1a,b mRNA and protein distribution patterns in the primate neocortex. Levels of GABABR1a,b receptor mRNA and protein expression are high overall and higher in area 17 than in area 18 and most neurons appear to express the receptor. GABABR1a,b receptors showed areal, laminar and cellular distribution patterns that are virtually identical in macaques and humans. GABABR1a,b mRNA and protein expression is decreased in deprived eye dominance columns of layers IVC and VI, indicating activity-dependent regulation.
The RNA probes and the antiserum used in the present study do not distinguish between the GABABR1a and GABABR1b splice variants. In situ hybridization studies in the adult rat brain showed that GABABR1a mRNA is expressed at higher levels than GABABR1b (Bischoff et al., 1997). GABABR1c and GABABR1d mRNAs are also present at high levels in the rat forebrain (Isomoto et al., 1998) and they also could contribute to the assembly of GABAB receptors in the cerebral cortex.
In the rat cerebral cortex GABABR2 transcripts are co-localized with GABABR1a and GABABR1b transcripts (Kaupmann et al., 1997, 1998), suggesting that they are co-expressed in individual neurons (Jones et al., 1998; Kaupmann et al., 1998; White et al., 1998; Kuner et al., 1999). Further studies using RNA probes and antibodies specific for different splice variants will be necessary to map their distribution and to evaluate their relative contribution to the GABAB receptors in the primate visual cortex. The present study in the primate neocortex, together with previous observations in the rat (Margeta Mitrovic et al., 1999), showing the presence of GABABR1a,b immunostaining in widespread neuronal populations, including pyramidal and non-pyramidal cells in every cortical layer except layer I, supports the physiological evidence for involvement of GABAB receptors in slow inhibition in pyramidal and non-pyramidal neurons of both supragranular and infragranular layers (Connors et al., 1982, 1988; Avoli, 1986; Howe et al., 1987; Deisz and Prince, 1989; Chagnac-Amitai and Connors, 1989a,b; McCormick, 1989; Douglas and Martin, 1991; Hirsch and Gilbert, 1991; Kawaguchi, 1992; Deisz et al., 1993; Hirsch, 1995; van Brederode and Spain, 1995; Shao and Burkhalter 1999).
The ubiquity of expression of GABABR1a,b mRNA and protein in neurons of the visual cortex suggests the involvement of GABAB receptors in many or all intracortical circuits. Slow GABAB-mediated inhibition is involved in the depression of polysynaptic EPSPs in feedforward but not in feedback connections joining supragranular layers of rat visual cortical areas (Shao and Burkhalter, 1999). Supragranular layer pyramidal cells in area LM of the rat, receiving forward inputs from area 17, are affected by slow GABAB-mediated IPSPs. In contrast, feedback from area LM to area 17 rarely evokes slow GABAB-mediated IPSPs in layer II/III pyramidal cells of area 17. However, the latter cells show a high incidence of GABAB responses after stimulation of interlaminar and horizontal inputs at the layer VI–white matter border (Shao and Burkhalter, 1999).
In area 17 the intense GABABR1a,b expression is closely associated with the main termination sites of inputs from the parvocellular layers of the lateral geniculate nucleus in layers IVA, IVCβ and the upper aspect of layer VI (Gilbert, 1983). In layer IVA a moderate expression of GABABR1a,b mRNA, corresponding to a sparse population of large and medium sized intensely GABABR1a,b immunoreactive cells, including some pyramidal cells, was found. This neuronal population was more intensely immunostained in macaques than in humans, which may be associated with the presence in macaques of a more developed P-geniculate afferent-related sublayer within layer IVA (Preuss et al., 1999). In contrast to area 17, layer IV of area 18, which is the main recipient layer of thalamic afferents, showed low levels of GABABR1a,b transcript expression which matched the faint GABABR1a,b immunostaining of granular cells.
In the present study GABAB receptors have been shown to be under activity-dependent control, mRNA and protein levels being down-regulated in deprived eye dominance columns by brief periods of monocular deprivation. A similar deprivation-induced decrease in GABABR1a,b mRNA expression has been previously described in both the magno-and parvocellular layers of the dorsal lateral geniculate nucleus (Muñoz et al., 1998). Deprivation-induced reductions in GABABR1a,b mRNA are prominent in the relevant ocular dominance columns of layer IVC and to a lesser extent in layer VI, but blobs and interblob regions in layers II/III and layer V are unnaffected, at least at levels detectable by the techniques used.
The present results provide further evidence for the involvement of retinal activity in regulating gene transcription in cortical circuits that maintain the balance of excitation and inhibition and which may play an important role in ocular dominance plasticity of the adult macaque cerebral cortex (Jones, 1990, 1993). GAD and GABAA receptors are also down-regulated by monocular deprivation (Hendry and Jones, 1986, 1988; Hendry et al., 1987; Benson et al., 1991, 1994). For the GABAA receptor subunits this occurs selectively. mRNA and protein expression of the α1, α3, β2 and γ2 subunits, which probably form the basis for the majority of functional GABAA receptors in the adult monkey visual cortex, are decreased in the deprived eye dominance columns in layer IVCβ and to a lesser extent in layer VI (Hendry et al., 1990, 1994; Huntsman et al., 1994; Jones, 1997; Huntsman and Jones, 1998). However, the α2, β1, β3 and γ1 subunits remain unaffected (Huntsman et al., 1994; Huntsman and Jones, 1998). The down-regulation of GABA and GAD expression, together with that of GABABR1a,b receptors and important GABAA receptor subunits, should leave the deprived eye dominance columns with reduced levels of both slow and fast GABA-mediated inhibition, probably contributing to the functional adaptation observed in visually deprived monkeys. Since GABAB receptors are located pre- as well as post-synaptically, their loss may represent an effort on the part of the visual cortex to maintain limited activity in pathways leading from the lateral geniculate nucleus to superficial layers of the visual cortex and thence on to extrastriate visual areas.
This work was supported by grant NS21377 from the National Institutes of Health, US Public Health Service and DGCYT grant PM99/1005 from the Spanish Goverment. Post-mortem human samples were kindly supplied by Dr A. Gómez (Department of Anatomy, Institute of Toxicology, Madrid). We thank Mr Phong Nguyen for excellent technical assistance.
Address correspondence to Dr Edward G. Jones, Center for Neuroscience, University of California at Davis, 1544 Newton Court, Davis, CA 95616, USA. Email: firstname.lastname@example.org.