- Email: [email protected]
GABAergic parvalbumin-immunoreactive large calyciform presynaptic complexes in the reticular nucleus of the rat thalamus
Journal of Chemical Neuroanatomy 30 (2005) 17–26 www.elsevier.com/locate/jchemneu
GABAergic parvalbumin-immunoreactive large calyciform presynaptic complexes in the reticular nucleus of the rat thalamus Bertalan Csillik a,*, Andra´s Miha´ly a, Beata Krisztin-Pe´va a, Zolta´n Chadaide c, Mohtasham Samsam b, Elizabeth Knyiha´r-Csillik c, Robert Fenyo a a
Department of Anatomy, Albert Szent-Gyo¨rgyi Medical and Pharmaceutical Center, University of Szeged, 40 Kossuth Lajos sgt, H-6701 Szeged, Hungary b Department of Neuroanatomy, University of Saba, Netherlands Antilles c Department of Neurology, Albert Szent-Gyo¨rgyi Medical and Pharmaceutical Center, University of Szeged, Hungary Received 1 December 2004; received in revised form 3 March 2005; accepted 25 March 2005 Available online 23 May 2005
Abstract In the reticular thalamic nucleus of the rat, nearly all neurons are parvalbumin-immunoreactive. We found that in addition, though superficially similar to large parvalbumin-immunoreactive neurons, also numerous peculiar parvalbumin-immunoreactive complexes are present in the reticular thalamic nucleus which are not identical with parvalbumin-immunoreactive perikarya, as shown by nuclear variation curves. Light and electron microscopic immunocytochemical studies revealed that these parvalbumin-immunoreactive complexes are brought about by parvalbumin-immunoreactive calyciform terminals which establish synapses with large, parvalbumin-immunonegative dendritic profiles. Transection of thalamo-reticular connections did not cause any alteration of calyciform terminals in the reticular thalamic nucleus. Nuclear counterstaining revealed that parvalbumin-immunoreactive calyciform terminals originated from local parvalbumin-immunoreactive interneuronal perikarya, which, depending of the length of the ‘‘neck’’ protruding from the perikaryon, establish somato-dendritic, axodendritic or dendro-dendritic synapses. Light and electron microscopic immunocytochemical investigations prove that the parvalbuminimmunoreactive calyciform complexes contain also GABA, that are likely to be inhibitory. In accordance with literature data, our results suggest that parvalbumin-immunoreactive GABAergic calyciform terminals in the reticular thalamic nucleus may be instrumental in intrinsic cell-to-cell communications and, as such, may be involved in synchronisation of thalamo-cortical oscillations, in the production of sleep spindles and in attentional processes. # 2005 Elsevier B.V. All rights reserved. Keywords: Thalamus; Reticular nucleus; Parvalbumin; GABA; Calyciform synapses; Somato-dendritc synapses; Dendro-dendritic synapses
1. Introduction The reticular thalamic nucleus (RTN), the ‘‘guardian of the gate’’ according to the controversial searchlight theory (Crick, 1984), is a shell-shaped structure, semicircular in coronal sections and hook-like in paramedian sagittal sections of the brain (Koelliker, 1896; Cajal, 1911). The RTN occupies a strategic position between the neocortex and the thalamus (Pinault, 2004). By axon collaterals innervating GABAergic RTN cells (De Biasi et al., 1986; * Corresponding author. Tel.: +36 62 544 918/545 665 (secretary); fax: +36 62 545 707. E-mail address: [email protected] (B. Csillik). 0891-0618/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jchemneu.2005.03.010
Houser et al., 1980), the RTN controls thalamo-cortical and cortico-thalamic functions (Pinault, 2004). Intrinsic cell-to-cell communications are instrumental in the function of RTN (Pinault, 2004); these are supposed to be mediated by dendro-dendritic synapses (Ahlsen and Lindstom, 1982), postulated to be present in the RTN by the Scheibels as early as 1972. Though the cell-to-cell communications are generally thought to be involved in lateral inhibition, some of them may be excitatory (Andersen and Eccles, 1962; Alger and Nicoll, 1979; Spreafico et al., 1988). It is assumed that dendro-dendritic synapses are instrumental in thalamo-cortical oscillations (Llina´s and Jahnsen, 1982; Deschenes et al., 1984; Jahnsen and Llina´s, 1984; Steriade and Deschenes, 1984), the RTN being the
18
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
pacemaker of sleep spindles (Steriade et al., 1987); thus GABAergic neurons may provide a key structure implicated in normal and paroxysmal oscillatory activities of RTN (Steriade, 2001; Jones, 2002). The overwhelming majority of neurons in the RTN display parvalbumin immunoreactivity (PV-IR; Celio, 1990). Several PV-IR structures, however, only seemingly correspond to PV-IR large nerve cells: microstructural analysis revealed that some of the large, PV-IR structures in the RTN are, in reality, presynaptic terminals of somatodendritic synapses (Csillik et al., 2002). In the present paper, we report on light and electron microscopic immunohistochemical studies, aiming to reveal whether or not, in addition to somato-dendritic juunctions, the PV-IR large calyciform complexes establish other types of synapses in the RTN. GABA immunostaining was performed at the light and electron microscopic levels, supplemented by fluorescence immunohistochemical studies in order to determine whether the PV-IR calyciform complexes are identical with similar structures expressing GABA. In order to estimate the relative importance of GABAergic and PV-IR calyciform complexes in the structure and in the function of RTN, the numbers and sizes of calyciform terminals have been determined by means of ocular micrometry and by stereological methods.
2. Materials and methods Investigations were carried out on 48 young adult Wistar rats, obtained from the animal house of the Albert SzentGyo¨ rgyi Medical and Pharmaceutical Center, University of Szeged. Care of the animals was in conformity with the guidelines controlling experiments and procedures in live animals, as described in the Principles of Laboratory Animal Care (NIH Publication no. 85-23, revised 1985), and also complied with the guidelines of the Hungarian Ministry of Welfare. Experiments were carried out in accordance with the European Communities Council Directive (24 November 1986; 86/609/EEC) and the Guidelines for Ethics in Animal Experiments, of the Albert Szent-Gyo¨ rgyi Medical and Pharmaceutical Center of the University of Szeged. 2.1. Immunocytochemistry Animals were subjected to transcardial perfusion with Zamboni’s picroaldehyde fixative in deep chloral hydrate anesthesia. The brain was removed, post-fixed in the same fixative overnight at 4 8C and sectioned with a cryostat either in the coronal plane, 5.0–7.0 mm from the frontal pole ( 1.8 to 3.8 mm from the bregma), or in a paramedian sagittal plane, 1.0–4.0 mm from the midline. Section thickness was 25 mm. For the demonstration of parvalbumin (PV), we used a polyclonal anti-mouse PV antibody raised in rabbits (Sigma, St. Louis, MO). Preliminary experiments showed best staining results with antibody dilutions 1:2000–l:3500.
Endogenous peroxidase activity was blocked by the application of 0.3% hydrogen peroxide diluted in methanol, for 10 min, followed by three successive rinses in 0.1 M phosphate buffer. Free-floating sections were pre-treated with blocking serum (0–1.0 M phosphate buffered saline [PBS]), 10% normal goat serum, l% bovine serum albumin [BSA] and 0.3% Triton X-100) on a shaker plate at room temperature for l h, and then transferred into the primary antibody. Incubation was carried out at 4 8C on a shaker for 36 h, followed by three rinses in 0.1 M phosphate buffer. To detect the bound primary antibody, we used the avidin– biotin peroxidase method. Kits were obtained from Vector Laboratories (Burlinghouse, CA, USA). The secondary antibody, biotinylated anti-mouse immunoglobulin was applied for 90 min at room temperature. Three more rinses in 0.1 M phosphate buffer were followed by incubation in the avidin–biotinylated-peroxidase complex for 60 min at room temperature. After three rinses in 0.1 M phosphate buffer, peroxidase activity was visualized by the histochemical reaction involving diamino-benzidine-tetrahydrochloride and hydrogen peroxide (3 ml of 30% H2O2 to 10 ml 1% DAB). After three rinses in 0.1 M phosphate buffer, the free-floating sections were dehydrated in a graded series of ethanols, cleared in xylene and coverslipped with Permount. The specificity of the immunohistochemical reaction was assessed by means of the following treatments: (1) omission of the first specific antiserum; (2) use of normal rabbit or mouse serum instead of anti-PV antiserum; (3) treatment according to the avidin–biotin complex method, from which one of the steps had been omitted; (4) preabsorbtion of the specific antibody with the blocking peptides sc7447P and sc7448P at 4 8C for 24 h. None of these specimens showed any reaction. GABA-IR was detected at the light microscopic level by means of a GABA antiserum (Chemicon, Temecula, CA, USA), obtained from rabbit, dilution 1:500. Staining of PV and GABA by means of immune fluorescence was performed on 30 mm cryostat sections of paraformaldehyde fixed rat brains, obtained in the paramedian sagittal plane. Sections were treated as free-floating specimens. The simultaneous (‘‘cocktail’’) method was employed. Blocking of endogeneous peroxidase was performed with 2% H2O2. After rinsing in PBS, containing 0.01% H2O2 and 0.1% Triton X-100, a PV antiserum, raised in mouse (Sigma) in a dilution of 1:2000, and GABA antiserum, raised in rabbit (Chemicon) in a dilution of 1:500, were used as primary antibodies. Incubation was performed overnight at room temperature. After three times rinsing in PBS, the sections were incubated in biotinylated anti-mouse IgG made in horse (Vector Laboratories) for 90 min. After rinsing in PBS, the incubation was followed in a mixture of fluorescein iso-thio-cyanate (FITC) labeled anti-rabbit IgG, raised in goat (Vector Laboratories), 1:200, and streptavidin peroxidase-labeled CY3 (Jackson, West Grove, PA, USA), 1:100 for 2 h at room temperature. After three rinses, 15 min
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
each in PBS, the sections were mounted in polyvinyl alcohol mounting medium (DABCO antifading, Sigma RBI). To prove the specificity of the GABA antibody, pure GABA (Tocris, Avonmouth, UK) was included in the cocktail containing primary antibodies. In such specimens, GABAIR did not show up while that of PV persisted. Investigation and photography of the sections was performed with a Nikon Eclipse E600 microscope, equipped with B2A and G.2A filters and a digital camera (SPOT RT Slider). The intensity of the immunoreaction was measured by densitometry, performed by digitalizing immunostained sections with a SPOT RT Slider CCD camera (1600 1200 pixels, 8 bits) attached to a Nikon Eclipse E600 microscope using 40 objective and 10 eyepiece. The captured images were analyzed by ImageProPlus v4.5 morphometric software (Media Cybernetics, Silver Sring, MD, USA). Areae of interest were rectangular areae measuring 0.008 mm2. Five rectangular fields were analyzed per animal, each field chosen randomly in different sections in a blinded manner. Gray values of the selected areas were obtained using a ScionNIH image analysis software. The images were captured directly from the microscopic slides, using a 5 objective lens by means of a black and white camera Cohu CCD and displayed on a computer monitor. The software used was the ImageProPlus v.4.5 program which automatically assigned the average gray value of the screen pixels coined to the outlined area; a value of 0 indicated a white pixel and 225 indicated a black pixel. Diameters of nuclei of PV-IR cells, and the diameters of the dendrites surrounded by the calyciform endings were determined by means of an ocular micrometer (Zeiss, Jena, Germany), using a 100 oil immersion front lense. Nuclear counterstaining of PV immunostained sections was performed with a 0.001% cresyl violet solution. Stereological determination of the numbers of PV-IR calyciform terminals in the RTN was performed according to the fractionator method described by West (1999) and by Buckmaster and Dudek (1997). On the basis of PV-stained serial sections, also the total PV-IR cell number in the RTN was determined. The number of sections analyzed per RTN, which extends 3.2 mm in the sagittal plane, was 13; total section thickness for disector height was 30 mm. For electron microscopic immunohistochemistry of PVIR, 50 mm thick free-floating sections were obtained on a vibratome and processed like the samples for light microscopy, with the only exception that Triton X-l00 was omitted from all of the solutions. Having visualized the peroxidase activity, free-floating sections were post-fixed in an osmic acid solution (2% OsO4 in phosphate buffer) for 1 h and dehydrated in a graded series of ethanol. Sections were flat embedded in Durcupan on Liquid Release-treated slides, and polymerized for 36 h at 56 8C. Selected regions of the brain were excised and remounted. Ultrathin serial sections were obtained on a Reichert Ultrotome, using a diamond knife. Sections were stained with lead citrate according to Reynolds, and analyzed and photographed
19
on a Zeiss Opton 902 electron microscope (Oberkochen, Germany). For electron microscopic immunohistochemistry of GABA-IR, 50 mm thick free-floating sections were obtained on a vibratome and processed like the samples for light microscopy, with the only exception that Triton X-l00 was omitted from all of the solutions. Having visualized the peroxidase activity, free-floating sections were post-fixed in an osmic acid solution (2% OsO4 in phosphate buffer) for 1 h and dehydrated in a graded series of ethanol. Sections were flat embedded in Durcupan. Selected regions of the brain were excised and remounted. Further treatment and fine sectioning of the samples was identical with those of PV-IR. In several cases, visualization of GABA-IR on PV immunostained sections was attempted with l0 nm goldlabeled goat anti-mouse IgG (Amersham, UK), in a dilution of 1:10, by means of the post-embedding technique. Experimental surgery was performed in nembutal anesthesia on six animals. RTN was isolated from the rest of the thalamus by means of a 8 mm deep coronal incision, 6 mm anterior from the bregma, with a 3 mm-wide ‘glass knife’ cut out from a histological coverslip (Palkovits et al., 1982) through a 3-mm-wide line-shaped hole drilled on the skull. For the statistical evaluation of the results, the mean values and the standard error (S.E.M.) were determined. Student’s independent samples t-test was used; values obtained from control and experimental animals were compared according to the ‘‘one-way ANOVA’’ technique (Bonferroni method or ‘‘post-hoc’’ test).
3. Results In conventional coronal sections of the rat brain, RTN appears as a PV-IR cap encircling the anterior portion of the thalamus (Fig. 1). Most of the PV-IR elements are nerve cell perikarya; some of the PV-immunopositive profiles in RTN, however, while deceptively similar to PV-immunopositive nerve cells equipped with large, immunonegative nuclei were found to correspond to cytoplasmic elements which embraced cross sections of large, immuno-negative structures (Fig. 2–6). More thorough scrutiny of the light microscopic immunocytochemical specimens (Fig. 7) revealed that the PV-IR profiles were not perikarya, but identical with PV-IR calyciform nerve endings, surrounding PV immunonegative large dendritic profiles. Nuclear counterstaining (Fig. 6) excluded the possibility that the PV immunonegative structures surrounded by the PV-IR terminals could have been cell nuclei. In the calyciform terminals, similar to those which exhibited PV-IR, also GABA immunoreactivity could be demonstrated (Figs. 8–10). Diameters of nuclei and those of the dendritic bulbs surrounded by PV-IR calyciform terminals were determined at the immersion optical level by means of a Zeiss ocular
20
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 1–6. Fig. 1. Parvalbumin immunoreactivity (PV-IR) of the RTN (ret) at low power in a coronal section of the brain. Note strong PV-IR of RTN as contrasted to the rest of the thalamus (asterisk). For orientation, see the symbol: S, superior; M, medial; L, lateral. Scale bar, 90 mm. Fig. 2. PV-IR of a terminal (arrowhead), surrounding the cross section of a dendrite (arrow) in a paramedian sagittal section of the brain. Scale bar, 10 mm. Fig. 3. The calyciform PV immunopositive complex (arrowhead) surrounds a dendrite (arrow). The pattern is deceptively similar to a PV-IR nerve cell equipped with a large immunonegative nucleus. Scale bar, 10 mm. Fig. 4. The calyciform PV immunopositive complex (arrowhead) embraces the grazing section of a large dendrite (arrow). Scale bar, 10 mm. Fig. 5. Three (partly incomplete) PV-IR axonal complexes (arrowheads), surrounding large, PV-immunonegative dendrites (arrows). Arrowhead with asterisk denotes cross section of a nucleus. Scale bar, 10 mm. Fig. 6. Three calyciform PV-IR complexes (arrowheads) embrace three immunonegative dendritic bulbs (arrows). Nuclear counterstaining with cresyl violet. Scale bar, 10 mm.
micrometer (Fig. 11). Accordingly, the spherical nuclei of PV-IR neurons fall into four categories: (1) nuclei of nondescript neurons, amounting to 20% of the whole nuclear population; these nuclei are were very small, measuring only 2–3 mm in diameter; (2) small fusiform cells had nuclei with a diameter of 5–7 mm; this category
amounted to 24% of the whole PV-IR nuclear population; (3) nuclei of large fusiform cells measured 6–8 mm; this group made up 29% of the PV-IR nuclear population, while (4) nuclei of large spherical cells had diameters of 8– 9 mm; this group amounted to 26% of the population of PVIR neurons in RTN. The diameters of the slightly ovoid
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
21
Fig. 7–10. Fig. 7. The PV-IR calyciform endings (arrowheads) surround a large dendritic profile (arrow). Scale bar, 10 mm. Fig. 8. A GABA-IR calyciform ending (arrowhead) surrounds a GABA immunonegative dendritic profile (arrow). The whole complex is deceptively similar to a GABA-IR neuron. Scale bar, 10 mm. Fig. 9. Arrowhead indicates a GABA-IR terminal, surrounding a GABA-immunonegative dendritic profile. Scale bar, 10 mm. Fig. 10. Arrowhead indicates a GABA-IR terminal, surrounding a GABA-immunonegative dendritic profile. Scale bar, 10 mm.
Fig. 11. Percentage values of variation curves of profiles of nuclei of PV-IR neurons and of those of dendritic profiles surrounded by PV-IR calyciform terminals in RTN of the normal rat. S, small nuclei of nondescript neurons; SF, nuclei of small fusiform neurons; LF, nuclei of large fusiform neurons; LS, nuclei of large spheroid neurons; D, profiles of PV immunonegative dendritic bulbs.
profiles of dendritic bulbs, surrounded by PV-IR calyciform elements, measured, in average, 10 mm 12 mm. The dendritic profiles amounted to 12% of the total population of PV-IR elements (nuclei and dendritic bulbs). Even more characteristic is the comparison of the surface areas occupied by the profiles of PV-immunonegative structures (nuclei + dendritic bulbs): that of neuronal nuclei varied between 5 mm2 and 84 mm2, while those of the dendritic bulbs varied between 68 mm2 and 101 mm2. According to stereological determinations, the number of PV-IR neurons in the RTN in a young adult rat was 14,700, while that of PVIR calyciform endings was 2040. In several cases it could be observed that perikarya of PVIR nerve cells were directly continuous with calyciform terminals. In such cases, the perikarya of the PV-IR neurons were equipped with neck-like protrusions, at the end of which the calyciform terminal emerged, surrounding a large dendritic profile. (Fig. 12a and b). According to low power fluorescence microscopy (Fig. 12c–e), localization of PV-IR structures in the RTN coincided with those which exerted GABA-IR. Nuclear counterstaining of GABA-IR specimens
22
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 12. (a and b) PV-IR neuron equipped with a neck-like outgrowth (arrow), surrounding a dendritic bulb (D). (a) PV-IR without counterstaining; (b) PV-IR after counterstaining with cresyl violet. Asterisk indicates nucleus of the PV-IR neuron; D, dendritic bulb. Scale bar, 10 mm. (c and d) Fluorescence microcopic demonstration of PV (c), and GABA (d) of RTN at low power. Scale bar, 0.1 mm. (e–g) Fluorescence microcopic demonstration of GABA (e and g), and superposition of GABA + PV (f) in the RTN at medium power. Scale bars, 40 mm. Arrows point at immuno-negative dendrites surrounded by GABA-IR calyciform terminals. Note that the GABA-IR calyciform complexes (lower arrows) in e and f are strikingly similar to that depicted in h with the DAB method. (f and g) GABA-IR in RTN after nuclear counterstaining with cresyl violet. (f) GABA-IR calyciform ending (arrowhead) surrounds dendritic bulb (D). (g) GABAIR nondescript cell in RTN with a very small nucleus (arrow with asterisk). Scale bar, 10 mm.
revealed identity of GABA-IR calyciform terminals with the PV-IR ones (Fig. 12f). The small nuclei, which yielded the first peak in the nuclear variation curves obtained from PVIR specimens, were found to belong to GABA-IR and PV-IR nondescript cellular elements (Figs. 12g). According to electron microscopic immunohistochemical investigations, the immunonegative profiles embraced by the calyciform PV-IR structures showed every characteristics of large dendrites (Figs. 13 and 14); some of them were even equipped with dendritic spines (Fig. 14). Within the PV-IR calyciform presynaptic elements, synaptic vesicles and mitochondria were found. GABA-IR could be detected in the calyciform terminals also at the level of electron microscopy (Figs. 15 and 16). In several cases,
dendro-dendritic attachment placques were observed between vicinal PV- and GABA-immunonegative dendrites (Fig. 17). Transection of thalamo-reticular pathways did not induce any alteration in the PV-IR and GABA-IR of the calyciform terminals in RTN.
4. Discussion The GABAergic RTN is, from the architectural, immunocytochemical, electrophysiological and pharmacological viewpoints, a heterogenous structure, preferential target of corticothalamic projections (Steriade, 2001).
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 13–14. Fig. 13. Electron microscopic organization of a PV-immunopositive calyciform complex in the reticular nucleus of the rat thalamus (C), full of synaptic vesicles. D, PV immunonegative dendritic profile. Fig. 14. Dendritic profile in a virtually longitudinal section, equipped with two dendritic spines (sp), surrounded by a calyciform PV-IR element (C) containing numerous synaptic vesicles. The whole complex is in the immediate vicinity of a blood vessel (BV), lined by a thin layer of endothelial cytoplasm (asterisk).
Located at the intersection of thalamo-cortical and corticothalamic axons, the RTN occupies a strategic position between the neocortex and the thalamus (Pinault, 2004). The RTN is concerned with almost all functional modalities represented by its auditory, gustatory, somatosensory, visual, visceral, motor and limbic sectors. The body surface is
23
Fig. 15–16. Fig. 15. Dendritic profile (D) surrounded by three GABA-IR elements of a calyiciform nerve ending (C1, C2, C3), loaded with synaptic vesicles. Arrows indicate sites corresponding to presynaptic densities. Fig. 16. Two dendritic profiles (D1 and D2) surrounded by two GABA-IR elements of a calyx (C), loaded with synaptic vesicles. Arrow points at presynaptic density; er, endoplasmic reticulum in one of the dendrites.
somatotopically represented in the RTN (Shosaku et al., 1989). Spreafico et al. (1988, 1991) found three morphological types of PV-IR neurons in the RTN of the adult rat, i.e. small fusiform ‘‘f-type’’ cells, large fusiform ‘‘F-neurons’’ and ‘‘S-type cells’’ with round perikarya and multipolar dendrites. In addition, we found several very small nuclei that belonged to nondescript large PV-IR and/or GABA-IR cellular elements. Dendro-dendritic synapses were described in the RTN by Deschenes et al. (1985), and by Pinault (2004) which were
24
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 17. Serial attachment plaqes (arrows) between two immunonegative dendritic profiles (D1 and D2). Note part of a GABA-IR calyx (C) and a GABA immunonegative axonal profile (A0) in the vicinity.
regarded to be relatively sparse as compared to the overwhelming majority of axo-somatic and axo-dendritic synapses. We found that in the rat RTN, some of the synapses are brought about by large, PV-IR and GABA-IR calyciform structures, which establish synapses with large PV-immunonegative dendritic profiles. According to Cajal (1911), RTN cells near the border of the anterior semilunar and somatosensory nuclei are equipped with thick dendrites that are oriented transversely. This peculiar anatomy of the dendrites might be the reason why calyciform terminals can most often be observed in paramedian sagittal sections of the brain while their occurrence in coronal sections is limited. According to Pinault et al. (1997,1998) dendro-dendritic junctions might be a prominent anatomical substrate underlying interneuronal communications in the RTN. In addition, there occur close appositions between cell bodies and dendrites in the RTN which are similar to gap junctions (Ohara and Lieberman, 1985). Dendrites of RTN often display signs of attachment placques and gap junctions, observed also by us (Fig. 18), are characteristic of electrical transmission which has been supposed to be involved in the function of the RTN (Landisman et al., 2002). The deceptive possibility that all the PV-IR and/or GABA-IR structures in the RTN corresponded to nerve cells equipped with large, immunonegative nuclei has been ruled out on the basis of nuclear counterstaining and electron microscopic immunocytochemical investigations. Electron microscopic immunocytochemistry proved that these structures are either parts of the perykarial cytoplasm
Fig. 18. Electron microscopy of a dendritic profile (D) surrounded by a calyciform PV-IR terminal (asterisk) containing numerous synaptic vesicles. Arrows indicate postsynaptic densities. BV, blood vessel.
(Csillik et al., 2002) or calyciform terminals, as shown in this paper. Calcium-binding proteins are important molecules acting like buffers to modulate dynamics of cytosolic Ca2+ transients (Baimbridge et al., 1992; Pinault, 2004). Presence of calcium-binding proteins such as calbindin, parvalbumin and calretinin, have been described in the RTN by Frassoni et al. (1991), Spreafico et al. (1991), Clemence and Mitrofanis (1992), Resibois and Rogers (1992), Winsky et al. (1992), Lizier et al. (1997), Amadeo et al. (1998, 2001), FitzGibbon et al. (2000) and Kakei et al. (2001). Calcium binding proteins may exert significant action on membrane potential, synaptic transmission and even on the firing pattern of the related neurons (Baimbridge et al., 1992; Caillard et al., 2000). The fact that PV coexists with GABA in the central nervous system is well documented (Lewis and Akil, 1998). According to De Biasi et al. (1986, 1997), most of the PV-IR structures in the RTN express also GABA. De Biasi et al. (1988) reported that GABAergic terminals made synaptic contact mainly with somata and proximal dendrites of nerve cells. In our present experiments, we have proved by simultaneous immunohistochemical reaction at the fluorescence microscopic level, that PV-IR presynaptic complexes are identical with GABA-IR ones. Some of the GABA immunoreactive structures described in the RTN and believed to be neuronal perikarya (Houser et al., 1980) are conspicuously similar to the presynaptic calyciform complexes observed by us. Unfortunately, the fine structure of these GABA-IR structures was not studied electron microscopically; thus their cytological identity could not be established.
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
On the other hand, according to Amadeo et al. (2001) neurons of RTN are PV- and GABA-IR at all developmental ages of the rat. Furthermore, ‘‘PV-positive cell bodies often showed several somatic protrusions, sometimes surrounding unlabelled vesicle-containing profiles or receiving presynaptically convex synaptic contacts from small terminals’’ (Amadeo et al., 2001). In our opinion, the ‘‘somatic protrusions’’ may be identical with the cellular ‘‘necks’’ from which there emerged the presynaptic calyciform complexes exhibiting PV-IR and GABA-IR, as described in this paper. On the other hand, some of the electron micrographs obtained from glutamic acid decarboxylase stained feline RTN, published by Yen et al. (1985) are very similar to parts of the electron microscopic pictures of our PV-IR and GABA-IR calyciform terminals (see especially Fig. 9D in Yen’s paper). Analogous patterns were found in the monkey RTN (Williamson et al., 1994). What can be the position of the peculiar calyciform presynaptic complexes in the intrinsic circuitry of the RTN? According to Carpenter and Sutin (1983), RTN does not obtain afferents from the periphery; this has been proved also by our earlier immunohistochemical studies (Csillik et al., 1996). Furthermore, since transection of thalamoreticular pathways did not induce alterations in structure and cytochemistry of calyciform complexes, it stands to reason to assume the local origin of the PV-IR and GABA-IR calyciform complexes. Indeed, though in small numbers, we found PV-IR and GABA-IR interneurons in RTN, equipped with a neck-like outgrowth from which the calyciform complex originated. Thus, the calyciform complexes and the neurons from which they derive, seem to be involved in intrinsic cell-to-cell communication (Pinault, 2004) and thus probably instrumental in synchronization of thalamocortical oscillations (Deschenes et al., 1984, Jahnsen and Llina´ s, 1984; Steriade and Deschenes, 1984), in the production of sleep spindles (Steriade et al., 1987) and in paroxysmal oscillatory activities (Steriade, 2001; Jones, 2002). Electrical stimulation of RTN (Knyiha´ r-Csillik et al., in press) results in activation of c-fos IR in the ipsilateral retrosplenial cortex after bicuculline administration, while stimulation of the retrosplenial cortex induces activation of c-fos IR in the ipsilateral RTN. The results of electrical stimulation support the well-known fact that RTN is in synaptic contact with the cortex (Guillery et al., 1998; Guillery and Harting, 2003; McAlonan and Brown, 2002) which seem to be mediated by GABAergic structures. According to Miha´ ly et al. (1998) inhibitory neurons in the RTN are primarily activated during focal neocortical seizures. RTN is supposed to be involved in attentional processes (MacAlonen et al., 2000); recent studies of Llina´ s et al. (2002) indicate a coincidence detector function of RTN. Sensory systems are known to be subjected to modulation by attention or distraction; the RTN is a key region in selective attention (Stehberg et al., 2001), and in executive attention
25
(Kilmer, 2001). This seems to be brought about by activation of RTN reflecting to attentional gating (MacAlonen et al., 2000). It is up to further studies whether or not, the large calyciform synapses play a part in this system.
Acknowledgements These studies were supported by the Hungarian National Research Foundation (OTKA), Grants No T-025033, T032039 and T-016046, and by Grants 40/2000 and 7/2003 from the Hungarian Medical Research Council (ETT). The text has been kindly edited by Dr. Ka´ roly Balogh, Boston, MA, USA.
References Ahlsen, G., Lindstom, S., 1982. Mutual inhibition between perigeniculate neurones. Brain Res. 236, 477–481. Alger, B.E., Nicoll, R.A., 1979. GABA-mediated biphasic inhibitory responses in hippocampus. Nature 281, 315–317. Andersen, P., Eccles, J.C., 1962. Inhibitory phasing of neuronal discharge. Nature 196, 645–647. Amadeo, A., De Biasi, S., Frassoni, C., Ortino, B., Spreafico, R., 1998. Immunocytochemical and ultrastructural study of the rat perireticular thalamic nucleus during postnatal development. J. Comp. Neurol. 392, 390–401. Amadeo, A., Ortino, B., Frassoni, C., 2001. Parvalbumin and GABA in the developing somatosensory thalamus of the rat: an immunocytochemical ultrastructural correlation. Anat. Embryol. 203, 109–119. Baimbridge, K.G., Celio, M.R., Rogers, J.H., 1992. Calcium binding proteins in the nervous system. Trends Neurosci. 15, 303–308. Buckmaster, P.S., Dudek, F.E., 1997. Neuron loss, granule cell axon reorganization, and functional changes in the dentate gyrus of epileptic kainate-treated rats. J. Comp. Neurol. 385, 385–404. Caillard, O., Moreno, H., Schwaller, B., Llano, I., Celio, M.R., Marty, A., 2000. Role of the calcium-binding protein parvalbumin in short-term synaptic plasticity. Proc. Natl. Acad. Sci. U.S.A. 97, 13372–13377. Cajal, S.R., 1911. Histologie du systeme nerveux de l’homme et des vertebras, vol. 2 (L. Azoulay, Trans. (French)) (N. Swanson and L.W. Swanson, Trans. (English, 1995)). Oxford University Press, p. 335. Carpenter, M.B., Sutin, J., 1983. Human Neuroanatomy. Williams and Wilkins, Baltimore, London, Los Angeles, Sidney, pp. 535–536. Celio, M.R., 1990. Calbindin D-28k and parvalbumin in the rat nervous system. Neuroscience 35, 375–475. Clemence, A.E., Mitrofanis, J., 1992. Cytoarchitectonic heterogeneities in the thalamic reticular nucleus of cats and ferrets. J. Comp. Neurol. 322, 167–180. Crick, F.H.C., 1984. The function of the thalamic reticular complex: the searchlight hypothesis. Proc. Natl. Acad. Sci. U.S.A. 81, 4586–4590. Csillik, B., Rakic, P., Tajti, J., Samsam, M., Ve´ csei, L., Knyiha´ r-Csillik, E., 1996. Alterations of parvalbumin immunoreactivity in Goll’s nucleus and in thalamic nuclei after transection of the dorsal column: transganglionic regulation of a calcium-binding neuroprotein. Ann. Anat. Suppl. 178, 25. ´ ., Gulya, K., Miha´ ly, A., Knyiha´ r-Csillik, E., 2002. Csillik, B., Pa´ lfi, A Somato-dendritic synapses in the nucleus reticularis thalami of the rat. Acta Biol. Hung. 53, 33–41. De Biasi, S., Frassoni, C., Spreafico, R., 1986. GABA immunoreactivity in the thalamic reticular nucleus of the rat. A light and electron microscopical study. Brain Res. 399, 143–147. De Biasi, S., Frassoni, C., Spreafico, R., 1988. The intrinsic organization of the ventroposterolateral nucleus and related reticular thalamic nucleus
26
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
of the rat: a double-labelling ultratsructural investigation with gammaaminobutyric acid immunogold staining and lectin-conjugated horseradish peroxidase. Somatosens. Res. 5, 187–203. De Biasi, S., Amadeo, A., Arcelli, P., Frassoni, C., Spreafico, R., 1997. Postnatal development of GABA immunoreactive terminals in tjhe reticular and ventrobasal nuclei of the rat thalamus: a light and electron microscopical study. Neuroscience 76, 503–516. Deschenes, M., Paradis, M., Roy, J.P., Steriade, M., 1984. Electrophysiology of neurons of lateral thalamic nuclei in cat: resting properties and burst discharges. J. Neurophysiol. 51, 1196–1219. Deschenes, M., Madariaga-Domich, A., Steriade, M., 1985. Dendrodendritic synapses in the cat reticularis thalami nucleus: a structural basis for thalamic spindle synchronization. Brain Res. 334, 165– 168. FitzGibbon, T., Solomon, S.G., Goodchild, A.K., 2000. Distribution of calbindin, parvalbumin, and calretinin immunoreactivity in the reticular thalamic nucleus of the marmoset: evidence for a medial leaflet of incertal neurons. Exp. Neurol. 164, 371–383. Frassoni, C., Bentivoglio, M., Spreafico, R., Sanchez, M.P., Puelles, L., Fairen, A., 1991. Postnatal development of calbindin and parvalbumin immunoreactivity in the thalamus of the rat. Brain Res. Dev. Brain Res. 58, 243–249. Guillery, R.W., Feigl, S.L., Lozsa´ di, D., 1998. Paying attention to the thalamic reticular nucleus. Trends Neurosci. 21, 28–32. Guillery, R.W., Harting, J.K., 2003. Structure and connections of the thalamic reticular nucleus: advancing views over half a century. J. Comp. Neurol. 463, 360–371. Houser, C.R., Vaughn, J.E., Barber, R.P., Roberts, E., 1980. GABA neurons are the major cell type of the nucleus reticularis thalami. Brain Res. 200, 341–354. Jahnsen, H., Llina´ s, R., 1984. Ionic basis for electroresponsiveness and oscillatory properties of guinea-pig thalamic neurones in vitro. J. Physiol. 349, 227–247. Jones, E.G., 2002. Thalamic circuitry and thalamocortical synchrony. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 357, 1659–1673. Kakei, S., Jie, N.A., Shinoda, Y., 2001. Thalamic terminal morphology and distribution of single corticothalamic axons originating from layers 5 and 6 of the cat motor cortex. J. Comp. Neurol. 437, 170–185. Kilmer, W., 2001. A thalamo-cortical model of the executive attention system. Biol. Cybern. 84, 279–289. Knyiha´ r-Csillik, E., Chadaide, Z., Csillik, B. Activation of c-fos in the reticular thalamic nucleus after electrical stimulation of the retrosplenial cortex. Ann. Anat. 186, in press. Koelliker, A., 1896. Handbuch der Gewebekehre des Menschen. In: Engelmann, W. (Ed.), Nervensystem des Menschen und der Thiere, sixth ed., vol. 2. Leipzig. Landisman, C.E., Long, M.A., Beierlein, M., Deans, D.L., Connors, B.W., 2002. Electrical synapses in the thalamic reticular nucleus. J. Neurosci. 22, 1002–1009. Lewis, D.A., Akil, M., 1998. Axon terminals of the chandelier class of cortical GABA neurons. Mol. Psychiatry 3, 466–467. Lizier, C., Spreafico, R., Battagia, G., 1997. Calretinin in the thalamic reticular nucleus of the rat: distribution and relationship with ipsilateral and contralateral efferents. J. Comp. Neurol. 377, 217–233. Llina´ s, R., Jahnsen, H., 1982. Electrophysiology of mammalian thalamic neurones in vitro. Nature 297, 406–408. Llina´ s, R.R., Leznik, E., Urbano, F.J., 2002. Temporal binding via cortical coincidence detection of specific and non-specific thalamocortical
inputs: a voltage-dependent dye-imaging study in mouse brain slices. Proc. Nat. Acad. Sci. U.S.A. 99, 449–454. MacAlonen, K., Brown, V.J., Bowman, E.M., 2000. Thalamic reticular nucleus activation reflects attentional gating during classical conditioning. J. Neurosci. 20, 8897–8901. McAlonan, K., Brown, V.J., 2002. The thalamic reticular nucleus: more than a sensory nucleus? Neuroscientist 8, 302–305. Miha´ ly, A., Szente, M., Dobo´ , E., Po´ r, I., 1998. Early activation of inhibitory neurons in the thalamic reticular nucleus during focal neocortical seizures. Acta Histochem. 100, 383–393. Ohara, P.T., Lieberman, A.R., 1985. The thalamic reticular nucleus of the adult rat: experimental anatomical studies. J. Neurocytol. 14, 365–411. Palkovits, M., Tapia-Arancibia, L., Kordon, C., Epalbaum, J., 1982. Somatostatin connections between the hypothalamus and the limbic system of the rat. Brain Res. 250, 223–228. Pinault, D., 2004. The thalamic reticular nucleus: structure, function and concept. Brain Res. Brain. Res. Rev. 46, 1–31. Pinault, D., Leresche, N., Charpier, S., Deniau, J.M., Marescaux, C., Vergnes, M., Crunelli, V., 1998. J. Physiol. (London) 509 (Pt 2) 449–456. Pinault, D., Smith, Y., Deschenes, M., 1997. Dendrodendritic and axoaxonal synapses in the thalamic reticular nucleus of the adult rat. J. Neurosci. 17, 3215–3233. Resibois, A., Rogers, J.H., 1992. Calretinin in the brain: an immunohistochemical study. Neuroscience 46, 101–134. Scheibel, M.E., Scheibel, A.B., 1972. Specialized organizational patterns within the nucleus reticularis thalami of the cat. Exp. Neurol. 34, 316– 322. Shosaku, A., Kayama, Y., Sumimoto, I., Sugitani, M., Iwama, K., 1989. Prog. Neurobiol. 32, 77–102. Spreafico, R., de Curtis, M., Frassoni, C., Avanzani, G., 1988. Electrophysiological characteristics of morphologically identified reticular thalamic neurons from rat slices. Neuroscience 27, 629–638. Spreafico, R., Battaglia, G., Frassoni, C., 1991. Reticular thalamic nucleus (RTN) of the rat: cytoarchitectural, Golgi, immunocytochemical, and horseradish peroxidase study. J. Comp. Neurol. 304, 478–490. Stehberg, J., Acuna-Goycolea, C., Ceric, F., Torrealba, F., 2001. The visceral sector of the thalamic reticular nucleus in the rat. Neuroscience 106, 745–755. Steriade, M., 2001. The GABAergic reticular nucleus: a preferential target of corticothalamic projections. PNAS 98, 3625–3627. Steriade, M., Deschenes, M., 1984. The thalamus as a neuronal oscillator. Brain Res. 320, 1–63. Steriade, M., Domich, L., Oakson, G., Deschenes, M., 1987. J. Neurophysiol. 57, 260–273. West, M.J., 1999. Stereological methods for estimating the total number of neurons and synapses: issues of precision and bias. TINS 22, 51–61. Williamson, A.M., Ohara, P.T., Ralston, D.D., Milroy, A.M., Ralston, H.R., 1994. Analysis of gamma-aminobutyric acidergic synaptic contacts int he thae´ lamic reticular nucleus of the monkey. J. Comp. Neurol. 349, 182–193. Winsky, L., Montpied, R., Arai, R., Martin, B.M., Jacobowitz, D.M., 1992. Calretinin distribution in the thalamus of the rat: immunohistochemican and in situ hybridization histochemical analyses. Neuroscince 50, 181– 196. Yen, C.T., Conley, M., Hendry, S.H.C., Jones, E.G., 1985. The morphology of physiologically identified GABAergic neurons in the somatic sensory part of the thalamic reticular nucleus in the cat. J. Neurosci. 5, 2254– 2268.
GABAergic parvalbumin-immunoreactive large calyciform presynaptic complexes in the reticular nucleus of the rat thalamus Bertalan Csillik a,*, Andra´s Miha´ly a, Beata Krisztin-Pe´va a, Zolta´n Chadaide c, Mohtasham Samsam b, Elizabeth Knyiha´r-Csillik c, Robert Fenyo a a
Department of Anatomy, Albert Szent-Gyo¨rgyi Medical and Pharmaceutical Center, University of Szeged, 40 Kossuth Lajos sgt, H-6701 Szeged, Hungary b Department of Neuroanatomy, University of Saba, Netherlands Antilles c Department of Neurology, Albert Szent-Gyo¨rgyi Medical and Pharmaceutical Center, University of Szeged, Hungary Received 1 December 2004; received in revised form 3 March 2005; accepted 25 March 2005 Available online 23 May 2005
Abstract In the reticular thalamic nucleus of the rat, nearly all neurons are parvalbumin-immunoreactive. We found that in addition, though superficially similar to large parvalbumin-immunoreactive neurons, also numerous peculiar parvalbumin-immunoreactive complexes are present in the reticular thalamic nucleus which are not identical with parvalbumin-immunoreactive perikarya, as shown by nuclear variation curves. Light and electron microscopic immunocytochemical studies revealed that these parvalbumin-immunoreactive complexes are brought about by parvalbumin-immunoreactive calyciform terminals which establish synapses with large, parvalbumin-immunonegative dendritic profiles. Transection of thalamo-reticular connections did not cause any alteration of calyciform terminals in the reticular thalamic nucleus. Nuclear counterstaining revealed that parvalbumin-immunoreactive calyciform terminals originated from local parvalbumin-immunoreactive interneuronal perikarya, which, depending of the length of the ‘‘neck’’ protruding from the perikaryon, establish somato-dendritic, axodendritic or dendro-dendritic synapses. Light and electron microscopic immunocytochemical investigations prove that the parvalbuminimmunoreactive calyciform complexes contain also GABA, that are likely to be inhibitory. In accordance with literature data, our results suggest that parvalbumin-immunoreactive GABAergic calyciform terminals in the reticular thalamic nucleus may be instrumental in intrinsic cell-to-cell communications and, as such, may be involved in synchronisation of thalamo-cortical oscillations, in the production of sleep spindles and in attentional processes. # 2005 Elsevier B.V. All rights reserved. Keywords: Thalamus; Reticular nucleus; Parvalbumin; GABA; Calyciform synapses; Somato-dendritc synapses; Dendro-dendritic synapses
1. Introduction The reticular thalamic nucleus (RTN), the ‘‘guardian of the gate’’ according to the controversial searchlight theory (Crick, 1984), is a shell-shaped structure, semicircular in coronal sections and hook-like in paramedian sagittal sections of the brain (Koelliker, 1896; Cajal, 1911). The RTN occupies a strategic position between the neocortex and the thalamus (Pinault, 2004). By axon collaterals innervating GABAergic RTN cells (De Biasi et al., 1986; * Corresponding author. Tel.: +36 62 544 918/545 665 (secretary); fax: +36 62 545 707. E-mail address: [email protected] (B. Csillik). 0891-0618/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jchemneu.2005.03.010
Houser et al., 1980), the RTN controls thalamo-cortical and cortico-thalamic functions (Pinault, 2004). Intrinsic cell-to-cell communications are instrumental in the function of RTN (Pinault, 2004); these are supposed to be mediated by dendro-dendritic synapses (Ahlsen and Lindstom, 1982), postulated to be present in the RTN by the Scheibels as early as 1972. Though the cell-to-cell communications are generally thought to be involved in lateral inhibition, some of them may be excitatory (Andersen and Eccles, 1962; Alger and Nicoll, 1979; Spreafico et al., 1988). It is assumed that dendro-dendritic synapses are instrumental in thalamo-cortical oscillations (Llina´s and Jahnsen, 1982; Deschenes et al., 1984; Jahnsen and Llina´s, 1984; Steriade and Deschenes, 1984), the RTN being the
18
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
pacemaker of sleep spindles (Steriade et al., 1987); thus GABAergic neurons may provide a key structure implicated in normal and paroxysmal oscillatory activities of RTN (Steriade, 2001; Jones, 2002). The overwhelming majority of neurons in the RTN display parvalbumin immunoreactivity (PV-IR; Celio, 1990). Several PV-IR structures, however, only seemingly correspond to PV-IR large nerve cells: microstructural analysis revealed that some of the large, PV-IR structures in the RTN are, in reality, presynaptic terminals of somatodendritic synapses (Csillik et al., 2002). In the present paper, we report on light and electron microscopic immunohistochemical studies, aiming to reveal whether or not, in addition to somato-dendritic juunctions, the PV-IR large calyciform complexes establish other types of synapses in the RTN. GABA immunostaining was performed at the light and electron microscopic levels, supplemented by fluorescence immunohistochemical studies in order to determine whether the PV-IR calyciform complexes are identical with similar structures expressing GABA. In order to estimate the relative importance of GABAergic and PV-IR calyciform complexes in the structure and in the function of RTN, the numbers and sizes of calyciform terminals have been determined by means of ocular micrometry and by stereological methods.
2. Materials and methods Investigations were carried out on 48 young adult Wistar rats, obtained from the animal house of the Albert SzentGyo¨ rgyi Medical and Pharmaceutical Center, University of Szeged. Care of the animals was in conformity with the guidelines controlling experiments and procedures in live animals, as described in the Principles of Laboratory Animal Care (NIH Publication no. 85-23, revised 1985), and also complied with the guidelines of the Hungarian Ministry of Welfare. Experiments were carried out in accordance with the European Communities Council Directive (24 November 1986; 86/609/EEC) and the Guidelines for Ethics in Animal Experiments, of the Albert Szent-Gyo¨ rgyi Medical and Pharmaceutical Center of the University of Szeged. 2.1. Immunocytochemistry Animals were subjected to transcardial perfusion with Zamboni’s picroaldehyde fixative in deep chloral hydrate anesthesia. The brain was removed, post-fixed in the same fixative overnight at 4 8C and sectioned with a cryostat either in the coronal plane, 5.0–7.0 mm from the frontal pole ( 1.8 to 3.8 mm from the bregma), or in a paramedian sagittal plane, 1.0–4.0 mm from the midline. Section thickness was 25 mm. For the demonstration of parvalbumin (PV), we used a polyclonal anti-mouse PV antibody raised in rabbits (Sigma, St. Louis, MO). Preliminary experiments showed best staining results with antibody dilutions 1:2000–l:3500.
Endogenous peroxidase activity was blocked by the application of 0.3% hydrogen peroxide diluted in methanol, for 10 min, followed by three successive rinses in 0.1 M phosphate buffer. Free-floating sections were pre-treated with blocking serum (0–1.0 M phosphate buffered saline [PBS]), 10% normal goat serum, l% bovine serum albumin [BSA] and 0.3% Triton X-100) on a shaker plate at room temperature for l h, and then transferred into the primary antibody. Incubation was carried out at 4 8C on a shaker for 36 h, followed by three rinses in 0.1 M phosphate buffer. To detect the bound primary antibody, we used the avidin– biotin peroxidase method. Kits were obtained from Vector Laboratories (Burlinghouse, CA, USA). The secondary antibody, biotinylated anti-mouse immunoglobulin was applied for 90 min at room temperature. Three more rinses in 0.1 M phosphate buffer were followed by incubation in the avidin–biotinylated-peroxidase complex for 60 min at room temperature. After three rinses in 0.1 M phosphate buffer, peroxidase activity was visualized by the histochemical reaction involving diamino-benzidine-tetrahydrochloride and hydrogen peroxide (3 ml of 30% H2O2 to 10 ml 1% DAB). After three rinses in 0.1 M phosphate buffer, the free-floating sections were dehydrated in a graded series of ethanols, cleared in xylene and coverslipped with Permount. The specificity of the immunohistochemical reaction was assessed by means of the following treatments: (1) omission of the first specific antiserum; (2) use of normal rabbit or mouse serum instead of anti-PV antiserum; (3) treatment according to the avidin–biotin complex method, from which one of the steps had been omitted; (4) preabsorbtion of the specific antibody with the blocking peptides sc7447P and sc7448P at 4 8C for 24 h. None of these specimens showed any reaction. GABA-IR was detected at the light microscopic level by means of a GABA antiserum (Chemicon, Temecula, CA, USA), obtained from rabbit, dilution 1:500. Staining of PV and GABA by means of immune fluorescence was performed on 30 mm cryostat sections of paraformaldehyde fixed rat brains, obtained in the paramedian sagittal plane. Sections were treated as free-floating specimens. The simultaneous (‘‘cocktail’’) method was employed. Blocking of endogeneous peroxidase was performed with 2% H2O2. After rinsing in PBS, containing 0.01% H2O2 and 0.1% Triton X-100, a PV antiserum, raised in mouse (Sigma) in a dilution of 1:2000, and GABA antiserum, raised in rabbit (Chemicon) in a dilution of 1:500, were used as primary antibodies. Incubation was performed overnight at room temperature. After three times rinsing in PBS, the sections were incubated in biotinylated anti-mouse IgG made in horse (Vector Laboratories) for 90 min. After rinsing in PBS, the incubation was followed in a mixture of fluorescein iso-thio-cyanate (FITC) labeled anti-rabbit IgG, raised in goat (Vector Laboratories), 1:200, and streptavidin peroxidase-labeled CY3 (Jackson, West Grove, PA, USA), 1:100 for 2 h at room temperature. After three rinses, 15 min
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
each in PBS, the sections were mounted in polyvinyl alcohol mounting medium (DABCO antifading, Sigma RBI). To prove the specificity of the GABA antibody, pure GABA (Tocris, Avonmouth, UK) was included in the cocktail containing primary antibodies. In such specimens, GABAIR did not show up while that of PV persisted. Investigation and photography of the sections was performed with a Nikon Eclipse E600 microscope, equipped with B2A and G.2A filters and a digital camera (SPOT RT Slider). The intensity of the immunoreaction was measured by densitometry, performed by digitalizing immunostained sections with a SPOT RT Slider CCD camera (1600 1200 pixels, 8 bits) attached to a Nikon Eclipse E600 microscope using 40 objective and 10 eyepiece. The captured images were analyzed by ImageProPlus v4.5 morphometric software (Media Cybernetics, Silver Sring, MD, USA). Areae of interest were rectangular areae measuring 0.008 mm2. Five rectangular fields were analyzed per animal, each field chosen randomly in different sections in a blinded manner. Gray values of the selected areas were obtained using a ScionNIH image analysis software. The images were captured directly from the microscopic slides, using a 5 objective lens by means of a black and white camera Cohu CCD and displayed on a computer monitor. The software used was the ImageProPlus v.4.5 program which automatically assigned the average gray value of the screen pixels coined to the outlined area; a value of 0 indicated a white pixel and 225 indicated a black pixel. Diameters of nuclei of PV-IR cells, and the diameters of the dendrites surrounded by the calyciform endings were determined by means of an ocular micrometer (Zeiss, Jena, Germany), using a 100 oil immersion front lense. Nuclear counterstaining of PV immunostained sections was performed with a 0.001% cresyl violet solution. Stereological determination of the numbers of PV-IR calyciform terminals in the RTN was performed according to the fractionator method described by West (1999) and by Buckmaster and Dudek (1997). On the basis of PV-stained serial sections, also the total PV-IR cell number in the RTN was determined. The number of sections analyzed per RTN, which extends 3.2 mm in the sagittal plane, was 13; total section thickness for disector height was 30 mm. For electron microscopic immunohistochemistry of PVIR, 50 mm thick free-floating sections were obtained on a vibratome and processed like the samples for light microscopy, with the only exception that Triton X-l00 was omitted from all of the solutions. Having visualized the peroxidase activity, free-floating sections were post-fixed in an osmic acid solution (2% OsO4 in phosphate buffer) for 1 h and dehydrated in a graded series of ethanol. Sections were flat embedded in Durcupan on Liquid Release-treated slides, and polymerized for 36 h at 56 8C. Selected regions of the brain were excised and remounted. Ultrathin serial sections were obtained on a Reichert Ultrotome, using a diamond knife. Sections were stained with lead citrate according to Reynolds, and analyzed and photographed
19
on a Zeiss Opton 902 electron microscope (Oberkochen, Germany). For electron microscopic immunohistochemistry of GABA-IR, 50 mm thick free-floating sections were obtained on a vibratome and processed like the samples for light microscopy, with the only exception that Triton X-l00 was omitted from all of the solutions. Having visualized the peroxidase activity, free-floating sections were post-fixed in an osmic acid solution (2% OsO4 in phosphate buffer) for 1 h and dehydrated in a graded series of ethanol. Sections were flat embedded in Durcupan. Selected regions of the brain were excised and remounted. Further treatment and fine sectioning of the samples was identical with those of PV-IR. In several cases, visualization of GABA-IR on PV immunostained sections was attempted with l0 nm goldlabeled goat anti-mouse IgG (Amersham, UK), in a dilution of 1:10, by means of the post-embedding technique. Experimental surgery was performed in nembutal anesthesia on six animals. RTN was isolated from the rest of the thalamus by means of a 8 mm deep coronal incision, 6 mm anterior from the bregma, with a 3 mm-wide ‘glass knife’ cut out from a histological coverslip (Palkovits et al., 1982) through a 3-mm-wide line-shaped hole drilled on the skull. For the statistical evaluation of the results, the mean values and the standard error (S.E.M.) were determined. Student’s independent samples t-test was used; values obtained from control and experimental animals were compared according to the ‘‘one-way ANOVA’’ technique (Bonferroni method or ‘‘post-hoc’’ test).
3. Results In conventional coronal sections of the rat brain, RTN appears as a PV-IR cap encircling the anterior portion of the thalamus (Fig. 1). Most of the PV-IR elements are nerve cell perikarya; some of the PV-immunopositive profiles in RTN, however, while deceptively similar to PV-immunopositive nerve cells equipped with large, immunonegative nuclei were found to correspond to cytoplasmic elements which embraced cross sections of large, immuno-negative structures (Fig. 2–6). More thorough scrutiny of the light microscopic immunocytochemical specimens (Fig. 7) revealed that the PV-IR profiles were not perikarya, but identical with PV-IR calyciform nerve endings, surrounding PV immunonegative large dendritic profiles. Nuclear counterstaining (Fig. 6) excluded the possibility that the PV immunonegative structures surrounded by the PV-IR terminals could have been cell nuclei. In the calyciform terminals, similar to those which exhibited PV-IR, also GABA immunoreactivity could be demonstrated (Figs. 8–10). Diameters of nuclei and those of the dendritic bulbs surrounded by PV-IR calyciform terminals were determined at the immersion optical level by means of a Zeiss ocular
20
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 1–6. Fig. 1. Parvalbumin immunoreactivity (PV-IR) of the RTN (ret) at low power in a coronal section of the brain. Note strong PV-IR of RTN as contrasted to the rest of the thalamus (asterisk). For orientation, see the symbol: S, superior; M, medial; L, lateral. Scale bar, 90 mm. Fig. 2. PV-IR of a terminal (arrowhead), surrounding the cross section of a dendrite (arrow) in a paramedian sagittal section of the brain. Scale bar, 10 mm. Fig. 3. The calyciform PV immunopositive complex (arrowhead) surrounds a dendrite (arrow). The pattern is deceptively similar to a PV-IR nerve cell equipped with a large immunonegative nucleus. Scale bar, 10 mm. Fig. 4. The calyciform PV immunopositive complex (arrowhead) embraces the grazing section of a large dendrite (arrow). Scale bar, 10 mm. Fig. 5. Three (partly incomplete) PV-IR axonal complexes (arrowheads), surrounding large, PV-immunonegative dendrites (arrows). Arrowhead with asterisk denotes cross section of a nucleus. Scale bar, 10 mm. Fig. 6. Three calyciform PV-IR complexes (arrowheads) embrace three immunonegative dendritic bulbs (arrows). Nuclear counterstaining with cresyl violet. Scale bar, 10 mm.
micrometer (Fig. 11). Accordingly, the spherical nuclei of PV-IR neurons fall into four categories: (1) nuclei of nondescript neurons, amounting to 20% of the whole nuclear population; these nuclei are were very small, measuring only 2–3 mm in diameter; (2) small fusiform cells had nuclei with a diameter of 5–7 mm; this category
amounted to 24% of the whole PV-IR nuclear population; (3) nuclei of large fusiform cells measured 6–8 mm; this group made up 29% of the PV-IR nuclear population, while (4) nuclei of large spherical cells had diameters of 8– 9 mm; this group amounted to 26% of the population of PVIR neurons in RTN. The diameters of the slightly ovoid
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
21
Fig. 7–10. Fig. 7. The PV-IR calyciform endings (arrowheads) surround a large dendritic profile (arrow). Scale bar, 10 mm. Fig. 8. A GABA-IR calyciform ending (arrowhead) surrounds a GABA immunonegative dendritic profile (arrow). The whole complex is deceptively similar to a GABA-IR neuron. Scale bar, 10 mm. Fig. 9. Arrowhead indicates a GABA-IR terminal, surrounding a GABA-immunonegative dendritic profile. Scale bar, 10 mm. Fig. 10. Arrowhead indicates a GABA-IR terminal, surrounding a GABA-immunonegative dendritic profile. Scale bar, 10 mm.
Fig. 11. Percentage values of variation curves of profiles of nuclei of PV-IR neurons and of those of dendritic profiles surrounded by PV-IR calyciform terminals in RTN of the normal rat. S, small nuclei of nondescript neurons; SF, nuclei of small fusiform neurons; LF, nuclei of large fusiform neurons; LS, nuclei of large spheroid neurons; D, profiles of PV immunonegative dendritic bulbs.
profiles of dendritic bulbs, surrounded by PV-IR calyciform elements, measured, in average, 10 mm 12 mm. The dendritic profiles amounted to 12% of the total population of PV-IR elements (nuclei and dendritic bulbs). Even more characteristic is the comparison of the surface areas occupied by the profiles of PV-immunonegative structures (nuclei + dendritic bulbs): that of neuronal nuclei varied between 5 mm2 and 84 mm2, while those of the dendritic bulbs varied between 68 mm2 and 101 mm2. According to stereological determinations, the number of PV-IR neurons in the RTN in a young adult rat was 14,700, while that of PVIR calyciform endings was 2040. In several cases it could be observed that perikarya of PVIR nerve cells were directly continuous with calyciform terminals. In such cases, the perikarya of the PV-IR neurons were equipped with neck-like protrusions, at the end of which the calyciform terminal emerged, surrounding a large dendritic profile. (Fig. 12a and b). According to low power fluorescence microscopy (Fig. 12c–e), localization of PV-IR structures in the RTN coincided with those which exerted GABA-IR. Nuclear counterstaining of GABA-IR specimens
22
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 12. (a and b) PV-IR neuron equipped with a neck-like outgrowth (arrow), surrounding a dendritic bulb (D). (a) PV-IR without counterstaining; (b) PV-IR after counterstaining with cresyl violet. Asterisk indicates nucleus of the PV-IR neuron; D, dendritic bulb. Scale bar, 10 mm. (c and d) Fluorescence microcopic demonstration of PV (c), and GABA (d) of RTN at low power. Scale bar, 0.1 mm. (e–g) Fluorescence microcopic demonstration of GABA (e and g), and superposition of GABA + PV (f) in the RTN at medium power. Scale bars, 40 mm. Arrows point at immuno-negative dendrites surrounded by GABA-IR calyciform terminals. Note that the GABA-IR calyciform complexes (lower arrows) in e and f are strikingly similar to that depicted in h with the DAB method. (f and g) GABA-IR in RTN after nuclear counterstaining with cresyl violet. (f) GABA-IR calyciform ending (arrowhead) surrounds dendritic bulb (D). (g) GABAIR nondescript cell in RTN with a very small nucleus (arrow with asterisk). Scale bar, 10 mm.
revealed identity of GABA-IR calyciform terminals with the PV-IR ones (Fig. 12f). The small nuclei, which yielded the first peak in the nuclear variation curves obtained from PVIR specimens, were found to belong to GABA-IR and PV-IR nondescript cellular elements (Figs. 12g). According to electron microscopic immunohistochemical investigations, the immunonegative profiles embraced by the calyciform PV-IR structures showed every characteristics of large dendrites (Figs. 13 and 14); some of them were even equipped with dendritic spines (Fig. 14). Within the PV-IR calyciform presynaptic elements, synaptic vesicles and mitochondria were found. GABA-IR could be detected in the calyciform terminals also at the level of electron microscopy (Figs. 15 and 16). In several cases,
dendro-dendritic attachment placques were observed between vicinal PV- and GABA-immunonegative dendrites (Fig. 17). Transection of thalamo-reticular pathways did not induce any alteration in the PV-IR and GABA-IR of the calyciform terminals in RTN.
4. Discussion The GABAergic RTN is, from the architectural, immunocytochemical, electrophysiological and pharmacological viewpoints, a heterogenous structure, preferential target of corticothalamic projections (Steriade, 2001).
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 13–14. Fig. 13. Electron microscopic organization of a PV-immunopositive calyciform complex in the reticular nucleus of the rat thalamus (C), full of synaptic vesicles. D, PV immunonegative dendritic profile. Fig. 14. Dendritic profile in a virtually longitudinal section, equipped with two dendritic spines (sp), surrounded by a calyciform PV-IR element (C) containing numerous synaptic vesicles. The whole complex is in the immediate vicinity of a blood vessel (BV), lined by a thin layer of endothelial cytoplasm (asterisk).
Located at the intersection of thalamo-cortical and corticothalamic axons, the RTN occupies a strategic position between the neocortex and the thalamus (Pinault, 2004). The RTN is concerned with almost all functional modalities represented by its auditory, gustatory, somatosensory, visual, visceral, motor and limbic sectors. The body surface is
23
Fig. 15–16. Fig. 15. Dendritic profile (D) surrounded by three GABA-IR elements of a calyiciform nerve ending (C1, C2, C3), loaded with synaptic vesicles. Arrows indicate sites corresponding to presynaptic densities. Fig. 16. Two dendritic profiles (D1 and D2) surrounded by two GABA-IR elements of a calyx (C), loaded with synaptic vesicles. Arrow points at presynaptic density; er, endoplasmic reticulum in one of the dendrites.
somatotopically represented in the RTN (Shosaku et al., 1989). Spreafico et al. (1988, 1991) found three morphological types of PV-IR neurons in the RTN of the adult rat, i.e. small fusiform ‘‘f-type’’ cells, large fusiform ‘‘F-neurons’’ and ‘‘S-type cells’’ with round perikarya and multipolar dendrites. In addition, we found several very small nuclei that belonged to nondescript large PV-IR and/or GABA-IR cellular elements. Dendro-dendritic synapses were described in the RTN by Deschenes et al. (1985), and by Pinault (2004) which were
24
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
Fig. 17. Serial attachment plaqes (arrows) between two immunonegative dendritic profiles (D1 and D2). Note part of a GABA-IR calyx (C) and a GABA immunonegative axonal profile (A0) in the vicinity.
regarded to be relatively sparse as compared to the overwhelming majority of axo-somatic and axo-dendritic synapses. We found that in the rat RTN, some of the synapses are brought about by large, PV-IR and GABA-IR calyciform structures, which establish synapses with large PV-immunonegative dendritic profiles. According to Cajal (1911), RTN cells near the border of the anterior semilunar and somatosensory nuclei are equipped with thick dendrites that are oriented transversely. This peculiar anatomy of the dendrites might be the reason why calyciform terminals can most often be observed in paramedian sagittal sections of the brain while their occurrence in coronal sections is limited. According to Pinault et al. (1997,1998) dendro-dendritic junctions might be a prominent anatomical substrate underlying interneuronal communications in the RTN. In addition, there occur close appositions between cell bodies and dendrites in the RTN which are similar to gap junctions (Ohara and Lieberman, 1985). Dendrites of RTN often display signs of attachment placques and gap junctions, observed also by us (Fig. 18), are characteristic of electrical transmission which has been supposed to be involved in the function of the RTN (Landisman et al., 2002). The deceptive possibility that all the PV-IR and/or GABA-IR structures in the RTN corresponded to nerve cells equipped with large, immunonegative nuclei has been ruled out on the basis of nuclear counterstaining and electron microscopic immunocytochemical investigations. Electron microscopic immunocytochemistry proved that these structures are either parts of the perykarial cytoplasm
Fig. 18. Electron microscopy of a dendritic profile (D) surrounded by a calyciform PV-IR terminal (asterisk) containing numerous synaptic vesicles. Arrows indicate postsynaptic densities. BV, blood vessel.
(Csillik et al., 2002) or calyciform terminals, as shown in this paper. Calcium-binding proteins are important molecules acting like buffers to modulate dynamics of cytosolic Ca2+ transients (Baimbridge et al., 1992; Pinault, 2004). Presence of calcium-binding proteins such as calbindin, parvalbumin and calretinin, have been described in the RTN by Frassoni et al. (1991), Spreafico et al. (1991), Clemence and Mitrofanis (1992), Resibois and Rogers (1992), Winsky et al. (1992), Lizier et al. (1997), Amadeo et al. (1998, 2001), FitzGibbon et al. (2000) and Kakei et al. (2001). Calcium binding proteins may exert significant action on membrane potential, synaptic transmission and even on the firing pattern of the related neurons (Baimbridge et al., 1992; Caillard et al., 2000). The fact that PV coexists with GABA in the central nervous system is well documented (Lewis and Akil, 1998). According to De Biasi et al. (1986, 1997), most of the PV-IR structures in the RTN express also GABA. De Biasi et al. (1988) reported that GABAergic terminals made synaptic contact mainly with somata and proximal dendrites of nerve cells. In our present experiments, we have proved by simultaneous immunohistochemical reaction at the fluorescence microscopic level, that PV-IR presynaptic complexes are identical with GABA-IR ones. Some of the GABA immunoreactive structures described in the RTN and believed to be neuronal perikarya (Houser et al., 1980) are conspicuously similar to the presynaptic calyciform complexes observed by us. Unfortunately, the fine structure of these GABA-IR structures was not studied electron microscopically; thus their cytological identity could not be established.
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
On the other hand, according to Amadeo et al. (2001) neurons of RTN are PV- and GABA-IR at all developmental ages of the rat. Furthermore, ‘‘PV-positive cell bodies often showed several somatic protrusions, sometimes surrounding unlabelled vesicle-containing profiles or receiving presynaptically convex synaptic contacts from small terminals’’ (Amadeo et al., 2001). In our opinion, the ‘‘somatic protrusions’’ may be identical with the cellular ‘‘necks’’ from which there emerged the presynaptic calyciform complexes exhibiting PV-IR and GABA-IR, as described in this paper. On the other hand, some of the electron micrographs obtained from glutamic acid decarboxylase stained feline RTN, published by Yen et al. (1985) are very similar to parts of the electron microscopic pictures of our PV-IR and GABA-IR calyciform terminals (see especially Fig. 9D in Yen’s paper). Analogous patterns were found in the monkey RTN (Williamson et al., 1994). What can be the position of the peculiar calyciform presynaptic complexes in the intrinsic circuitry of the RTN? According to Carpenter and Sutin (1983), RTN does not obtain afferents from the periphery; this has been proved also by our earlier immunohistochemical studies (Csillik et al., 1996). Furthermore, since transection of thalamoreticular pathways did not induce alterations in structure and cytochemistry of calyciform complexes, it stands to reason to assume the local origin of the PV-IR and GABA-IR calyciform complexes. Indeed, though in small numbers, we found PV-IR and GABA-IR interneurons in RTN, equipped with a neck-like outgrowth from which the calyciform complex originated. Thus, the calyciform complexes and the neurons from which they derive, seem to be involved in intrinsic cell-to-cell communication (Pinault, 2004) and thus probably instrumental in synchronization of thalamocortical oscillations (Deschenes et al., 1984, Jahnsen and Llina´ s, 1984; Steriade and Deschenes, 1984), in the production of sleep spindles (Steriade et al., 1987) and in paroxysmal oscillatory activities (Steriade, 2001; Jones, 2002). Electrical stimulation of RTN (Knyiha´ r-Csillik et al., in press) results in activation of c-fos IR in the ipsilateral retrosplenial cortex after bicuculline administration, while stimulation of the retrosplenial cortex induces activation of c-fos IR in the ipsilateral RTN. The results of electrical stimulation support the well-known fact that RTN is in synaptic contact with the cortex (Guillery et al., 1998; Guillery and Harting, 2003; McAlonan and Brown, 2002) which seem to be mediated by GABAergic structures. According to Miha´ ly et al. (1998) inhibitory neurons in the RTN are primarily activated during focal neocortical seizures. RTN is supposed to be involved in attentional processes (MacAlonen et al., 2000); recent studies of Llina´ s et al. (2002) indicate a coincidence detector function of RTN. Sensory systems are known to be subjected to modulation by attention or distraction; the RTN is a key region in selective attention (Stehberg et al., 2001), and in executive attention
25
(Kilmer, 2001). This seems to be brought about by activation of RTN reflecting to attentional gating (MacAlonen et al., 2000). It is up to further studies whether or not, the large calyciform synapses play a part in this system.
Acknowledgements These studies were supported by the Hungarian National Research Foundation (OTKA), Grants No T-025033, T032039 and T-016046, and by Grants 40/2000 and 7/2003 from the Hungarian Medical Research Council (ETT). The text has been kindly edited by Dr. Ka´ roly Balogh, Boston, MA, USA.
References Ahlsen, G., Lindstom, S., 1982. Mutual inhibition between perigeniculate neurones. Brain Res. 236, 477–481. Alger, B.E., Nicoll, R.A., 1979. GABA-mediated biphasic inhibitory responses in hippocampus. Nature 281, 315–317. Andersen, P., Eccles, J.C., 1962. Inhibitory phasing of neuronal discharge. Nature 196, 645–647. Amadeo, A., De Biasi, S., Frassoni, C., Ortino, B., Spreafico, R., 1998. Immunocytochemical and ultrastructural study of the rat perireticular thalamic nucleus during postnatal development. J. Comp. Neurol. 392, 390–401. Amadeo, A., Ortino, B., Frassoni, C., 2001. Parvalbumin and GABA in the developing somatosensory thalamus of the rat: an immunocytochemical ultrastructural correlation. Anat. Embryol. 203, 109–119. Baimbridge, K.G., Celio, M.R., Rogers, J.H., 1992. Calcium binding proteins in the nervous system. Trends Neurosci. 15, 303–308. Buckmaster, P.S., Dudek, F.E., 1997. Neuron loss, granule cell axon reorganization, and functional changes in the dentate gyrus of epileptic kainate-treated rats. J. Comp. Neurol. 385, 385–404. Caillard, O., Moreno, H., Schwaller, B., Llano, I., Celio, M.R., Marty, A., 2000. Role of the calcium-binding protein parvalbumin in short-term synaptic plasticity. Proc. Natl. Acad. Sci. U.S.A. 97, 13372–13377. Cajal, S.R., 1911. Histologie du systeme nerveux de l’homme et des vertebras, vol. 2 (L. Azoulay, Trans. (French)) (N. Swanson and L.W. Swanson, Trans. (English, 1995)). Oxford University Press, p. 335. Carpenter, M.B., Sutin, J., 1983. Human Neuroanatomy. Williams and Wilkins, Baltimore, London, Los Angeles, Sidney, pp. 535–536. Celio, M.R., 1990. Calbindin D-28k and parvalbumin in the rat nervous system. Neuroscience 35, 375–475. Clemence, A.E., Mitrofanis, J., 1992. Cytoarchitectonic heterogeneities in the thalamic reticular nucleus of cats and ferrets. J. Comp. Neurol. 322, 167–180. Crick, F.H.C., 1984. The function of the thalamic reticular complex: the searchlight hypothesis. Proc. Natl. Acad. Sci. U.S.A. 81, 4586–4590. Csillik, B., Rakic, P., Tajti, J., Samsam, M., Ve´ csei, L., Knyiha´ r-Csillik, E., 1996. Alterations of parvalbumin immunoreactivity in Goll’s nucleus and in thalamic nuclei after transection of the dorsal column: transganglionic regulation of a calcium-binding neuroprotein. Ann. Anat. Suppl. 178, 25. ´ ., Gulya, K., Miha´ ly, A., Knyiha´ r-Csillik, E., 2002. Csillik, B., Pa´ lfi, A Somato-dendritic synapses in the nucleus reticularis thalami of the rat. Acta Biol. Hung. 53, 33–41. De Biasi, S., Frassoni, C., Spreafico, R., 1986. GABA immunoreactivity in the thalamic reticular nucleus of the rat. A light and electron microscopical study. Brain Res. 399, 143–147. De Biasi, S., Frassoni, C., Spreafico, R., 1988. The intrinsic organization of the ventroposterolateral nucleus and related reticular thalamic nucleus
26
B. Csillik et al. / Journal of Chemical Neuroanatomy 30 (2005) 17–26
of the rat: a double-labelling ultratsructural investigation with gammaaminobutyric acid immunogold staining and lectin-conjugated horseradish peroxidase. Somatosens. Res. 5, 187–203. De Biasi, S., Amadeo, A., Arcelli, P., Frassoni, C., Spreafico, R., 1997. Postnatal development of GABA immunoreactive terminals in tjhe reticular and ventrobasal nuclei of the rat thalamus: a light and electron microscopical study. Neuroscience 76, 503–516. Deschenes, M., Paradis, M., Roy, J.P., Steriade, M., 1984. Electrophysiology of neurons of lateral thalamic nuclei in cat: resting properties and burst discharges. J. Neurophysiol. 51, 1196–1219. Deschenes, M., Madariaga-Domich, A., Steriade, M., 1985. Dendrodendritic synapses in the cat reticularis thalami nucleus: a structural basis for thalamic spindle synchronization. Brain Res. 334, 165– 168. FitzGibbon, T., Solomon, S.G., Goodchild, A.K., 2000. Distribution of calbindin, parvalbumin, and calretinin immunoreactivity in the reticular thalamic nucleus of the marmoset: evidence for a medial leaflet of incertal neurons. Exp. Neurol. 164, 371–383. Frassoni, C., Bentivoglio, M., Spreafico, R., Sanchez, M.P., Puelles, L., Fairen, A., 1991. Postnatal development of calbindin and parvalbumin immunoreactivity in the thalamus of the rat. Brain Res. Dev. Brain Res. 58, 243–249. Guillery, R.W., Feigl, S.L., Lozsa´ di, D., 1998. Paying attention to the thalamic reticular nucleus. Trends Neurosci. 21, 28–32. Guillery, R.W., Harting, J.K., 2003. Structure and connections of the thalamic reticular nucleus: advancing views over half a century. J. Comp. Neurol. 463, 360–371. Houser, C.R., Vaughn, J.E., Barber, R.P., Roberts, E., 1980. GABA neurons are the major cell type of the nucleus reticularis thalami. Brain Res. 200, 341–354. Jahnsen, H., Llina´ s, R., 1984. Ionic basis for electroresponsiveness and oscillatory properties of guinea-pig thalamic neurones in vitro. J. Physiol. 349, 227–247. Jones, E.G., 2002. Thalamic circuitry and thalamocortical synchrony. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 357, 1659–1673. Kakei, S., Jie, N.A., Shinoda, Y., 2001. Thalamic terminal morphology and distribution of single corticothalamic axons originating from layers 5 and 6 of the cat motor cortex. J. Comp. Neurol. 437, 170–185. Kilmer, W., 2001. A thalamo-cortical model of the executive attention system. Biol. Cybern. 84, 279–289. Knyiha´ r-Csillik, E., Chadaide, Z., Csillik, B. Activation of c-fos in the reticular thalamic nucleus after electrical stimulation of the retrosplenial cortex. Ann. Anat. 186, in press. Koelliker, A., 1896. Handbuch der Gewebekehre des Menschen. In: Engelmann, W. (Ed.), Nervensystem des Menschen und der Thiere, sixth ed., vol. 2. Leipzig. Landisman, C.E., Long, M.A., Beierlein, M., Deans, D.L., Connors, B.W., 2002. Electrical synapses in the thalamic reticular nucleus. J. Neurosci. 22, 1002–1009. Lewis, D.A., Akil, M., 1998. Axon terminals of the chandelier class of cortical GABA neurons. Mol. Psychiatry 3, 466–467. Lizier, C., Spreafico, R., Battagia, G., 1997. Calretinin in the thalamic reticular nucleus of the rat: distribution and relationship with ipsilateral and contralateral efferents. J. Comp. Neurol. 377, 217–233. Llina´ s, R., Jahnsen, H., 1982. Electrophysiology of mammalian thalamic neurones in vitro. Nature 297, 406–408. Llina´ s, R.R., Leznik, E., Urbano, F.J., 2002. Temporal binding via cortical coincidence detection of specific and non-specific thalamocortical
inputs: a voltage-dependent dye-imaging study in mouse brain slices. Proc. Nat. Acad. Sci. U.S.A. 99, 449–454. MacAlonen, K., Brown, V.J., Bowman, E.M., 2000. Thalamic reticular nucleus activation reflects attentional gating during classical conditioning. J. Neurosci. 20, 8897–8901. McAlonan, K., Brown, V.J., 2002. The thalamic reticular nucleus: more than a sensory nucleus? Neuroscientist 8, 302–305. Miha´ ly, A., Szente, M., Dobo´ , E., Po´ r, I., 1998. Early activation of inhibitory neurons in the thalamic reticular nucleus during focal neocortical seizures. Acta Histochem. 100, 383–393. Ohara, P.T., Lieberman, A.R., 1985. The thalamic reticular nucleus of the adult rat: experimental anatomical studies. J. Neurocytol. 14, 365–411. Palkovits, M., Tapia-Arancibia, L., Kordon, C., Epalbaum, J., 1982. Somatostatin connections between the hypothalamus and the limbic system of the rat. Brain Res. 250, 223–228. Pinault, D., 2004. The thalamic reticular nucleus: structure, function and concept. Brain Res. Brain. Res. Rev. 46, 1–31. Pinault, D., Leresche, N., Charpier, S., Deniau, J.M., Marescaux, C., Vergnes, M., Crunelli, V., 1998. J. Physiol. (London) 509 (Pt 2) 449–456. Pinault, D., Smith, Y., Deschenes, M., 1997. Dendrodendritic and axoaxonal synapses in the thalamic reticular nucleus of the adult rat. J. Neurosci. 17, 3215–3233. Resibois, A., Rogers, J.H., 1992. Calretinin in the brain: an immunohistochemical study. Neuroscience 46, 101–134. Scheibel, M.E., Scheibel, A.B., 1972. Specialized organizational patterns within the nucleus reticularis thalami of the cat. Exp. Neurol. 34, 316– 322. Shosaku, A., Kayama, Y., Sumimoto, I., Sugitani, M., Iwama, K., 1989. Prog. Neurobiol. 32, 77–102. Spreafico, R., de Curtis, M., Frassoni, C., Avanzani, G., 1988. Electrophysiological characteristics of morphologically identified reticular thalamic neurons from rat slices. Neuroscience 27, 629–638. Spreafico, R., Battaglia, G., Frassoni, C., 1991. Reticular thalamic nucleus (RTN) of the rat: cytoarchitectural, Golgi, immunocytochemical, and horseradish peroxidase study. J. Comp. Neurol. 304, 478–490. Stehberg, J., Acuna-Goycolea, C., Ceric, F., Torrealba, F., 2001. The visceral sector of the thalamic reticular nucleus in the rat. Neuroscience 106, 745–755. Steriade, M., 2001. The GABAergic reticular nucleus: a preferential target of corticothalamic projections. PNAS 98, 3625–3627. Steriade, M., Deschenes, M., 1984. The thalamus as a neuronal oscillator. Brain Res. 320, 1–63. Steriade, M., Domich, L., Oakson, G., Deschenes, M., 1987. J. Neurophysiol. 57, 260–273. West, M.J., 1999. Stereological methods for estimating the total number of neurons and synapses: issues of precision and bias. TINS 22, 51–61. Williamson, A.M., Ohara, P.T., Ralston, D.D., Milroy, A.M., Ralston, H.R., 1994. Analysis of gamma-aminobutyric acidergic synaptic contacts int he thae´ lamic reticular nucleus of the monkey. J. Comp. Neurol. 349, 182–193. Winsky, L., Montpied, R., Arai, R., Martin, B.M., Jacobowitz, D.M., 1992. Calretinin distribution in the thalamus of the rat: immunohistochemican and in situ hybridization histochemical analyses. Neuroscince 50, 181– 196. Yen, C.T., Conley, M., Hendry, S.H.C., Jones, E.G., 1985. The morphology of physiologically identified GABAergic neurons in the somatic sensory part of the thalamic reticular nucleus in the cat. J. Neurosci. 5, 2254– 2268.