- Email: [email protected]
NPY-, SOM- and VIP-containing interneurons in postnatal development of the rat claustrum
Brain Research Bulletin 76 (2008) 565–571
Contents lists available at ScienceDirect
Brain Research Bulletin journal homepage: www.elsevier.com/locate/brainresbull
Research report
NPY-, SOM- and VIP-containing interneurons in postnatal development of the rat claustrum ∗ ´ ˛ ´ Przemysław Kowianski , Joanna M. Mory´s, Jerzy Dziewiatkowski, Sławomir Wojcik, Justyna Sidor-Kaczmarek, Janusz Mory´s Department of Anatomy and Neurobiology, Medical University of Gda´ nsk, 1 D˛ebinki Street, 80-211 Gda´ nsk, Poland
a r t i c l e
i n f o
Article history: Received 18 September 2007 Received in revised form 3 March 2008 Accepted 16 April 2008 Available online 13 May 2008 Keywords: Amygdaloid body Claustrum Development Interneurons Neuropeptides
a b s t r a c t A growing body of evidence indicates the common origin of the claustrum, endopiriform nucleus, and the basolateral nuclear complex of amygdala from the lateral and ventral parts of the pallium, as the claustroamygdaloid complex. It seems very probable that at least some of the claustral interneurons derive from subcortical sources. The postnatal development of neuropeptide Y-, somatostatin- and vasoactive intestinal polypeptidecontaining interneurons was studied during the 4 postnatal months (P0–P120; P, postnatal day). The study was conducted on 45 Wistar rats of both sexes. Our results indicate that neuropeptide-containing interneurons are not morphologically mature at the moment of birth. The characteristic features of neuronal bodies and the relatively long period of postnatal development may indicate their migration from the subcortical neurogenetic centers. Morphological changes in the neuropil are also reported. Although developmental patterns differ between various neuropeptide-containing neuronal subpopulations, two phases of development can be distinguished in each of them: the early phase (P0–P4) during which undifferentiated neurons and neuropil dominate, and the late phase (P7–P28) during which the characteristic features of an adult-like structure gradually appear. Later these observed developmental changes are terminated. The postnatal development of neuropeptide-containing interneurons is completed after 4 weeks of life. This period, which is important for the structural and functional development of the claustrum, must be taken into account in future studies on this structure. © 2008 Elsevier Inc. All rights reserved.
1. Introduction The claustrum (Cl; dorsal claustrum or claustrum proper) is a telencephalic structure characterized by extensive, reciprocal connections with numerous neocortical areas [14]. A growing body of evidence indicates that it plays a role in processes of cross-modal sensory integration [3,7,12,23], memory [54,55] and epileptogenesis [47]. Because of its numerous cortical connections, the Cl is able to play a particular role enabling the transfer of sensory information of various modalities between cortical areas. The special function of the Cl is also concerned with kindling propagation, as well as with the generalization of seizures [34,35]. The Cl consists of two morphologically distinguished neuronal subpopulations [31]. The prevalent group consists of glutamater-
∗ Corresponding author. Tel.: +48 58 349 1401; fax: +48 58 349 1421. ´ E-mail address: [email protected] (P. Kowianski). 0361-9230/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2008.04.004
gic projecting neurons with spiny dendrites. The second group of GABA-ergic interneurons, represented by variously shaped somata and aspiny dendrites, constitute, in various species, between 7 and 12% of all claustral neurons [5,17,25,50]. Apart from GABA, the occurrence of neuropeptides such as neuropeptide Y (NPY), somatostatin (SOM), vasoactive intestinal polypeptide (VIP) and cholecystokinin (CCK) has been described in the Cl [14,15,27,48,49]. The expression patterns of developmental regulatory genes in mice indicate the origin of the Cl to be from the lateral and ventral pallial divisions [33,43,45]. Neurogenesis of the claustral neurons in the cortical neuroepithelium of the rat was estimated to be between E15 and E16 [4]. However, there is evidence that at least some of the claustral interneurons may originate in the subpallium and migrate tangentially during development [29,43]. According to previously published results, the development of the Cl is not complete at the moment of birth. The increase of its total volume and the loss of nearly 30% of neurons, including apoptotic death, take place after birth [30]. Moreover, characteristic
566
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
morphological changes among interneurons containing calciumbinding proteins (CaBP) and nitric oxide synthase (NOS) have been described [13,22,26]. Neuropeptide-containing interneurons may be very important for the initiation of integrative function and memory processes coordinated by the Cl [12]. The stabilizing influence of peptide neuromodulators on the activity of claustral projecting neurons may be crucial for the development of claustro-cortical connections. Our studies focused on the morphological changes of claustral interneurons, immunoreactive to three neuropeptides (NPY, SOM and VIP), during the first 4 postnatal months.
magnification. The borders of Cl were marked as separate inclusion areas under low magnification (4×). The neuronal profiles were counted and the numerical density (number of cell profiles per mm2 ) was calculated with the aid of a 20× objective lens in systematic random test frames of the selected area using a C.A.S.T. grid system (Computer Assisted Stereological Tool, Olympus, Denmark) and microscope (BX-51, Olympus, Japan). 2.4. Statistics All the calculations were performed using Excel 2000 (Microsoft, USA). Raw data concerning the number of cells per mm2 were collected. Mean values and standard deviations were calculated for each group. The comparisons of neuronal densities between various immunoreactive neuronal populations were performed using a nonparametric multiple comparison test (Steel–Dwas test, Tukey equivalent test).
2. Materials and methods
3. Results Forty-five Wistar rats of both sexes were used in this study. Animal care and treatment guidelines established by the local Ethical Committee, as well as standards defined by the European Communities Council Directive of 24 November 1986 (86/609/EEC), were followed. The animals were divided into nine age groups: P0, P4, P7, P14, P21, P28, P60, P90, and P120, with five animals in each group. The animals were deeply anaesthetized with sodium pentobarbital (thiopental sodium, Biochemie GmBH, Germany; 80 mg/kg of body weight, i.p.) and transcardially perfused with 200 ml of physiological saline (at pH 7.4) followed by 400 ml of 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS, at pH 7.4 and 4 ◦ C). After removal from the skull, the brains were immersed in the same fixative for 120 min and stored in 30% sucrose in 0.1 M PBS (at pH 7.4 and 4 ◦ C) for at least 24 h. Subsequently, the brains were frozen and sectioned coronally into 40-m-thick serial sections on a sliding microtome. 2.1. Immunocytochemical procedure The free-floating sections were processed for fluorescence immunocytochemistry. Three primary antibodies were used in the study: anti-NPY (rabbit; 1:400, Affiniti, UK), anti-SOM (rabbit; 1:300, Euro-Diagnostica AB, Sweden), and anti-VIP (rabbit; 1:100, Cappel ICN Pharmaceuticals Inc., USA). After incubation for 1 h in a blocking solution containing 0.1% bovine serum albumin in PBS and 10% normal goat serum, the sections were incubated overnight in primary antibodies at room temperature. Subsequently, they were rinsed in PBS and incubated in the secondary antibody (goat anti-rabbit; conjugated to indocarbocyanine [Cy-3] used at 1:800, Jackson ImmunoResearch, USA) for 1 h at room temperature. The sections were rinsed in PBS, mounted on slides, dried, and coverslipped with Vectashield (Vector Laboratories Inc., USA). In order to confirm the specificity of staining, the sections were processed by immunocytochemical procedure with the omission of primary or secondary antibody. In all cases the omission tests resulted in a lack of specific labeling. 2.2. Qualitative study In each immunocytochemically distinguished neuronal subpopulation, the shapes of the perikarya and their polarity were studied. Additionally, the morphological features of the neuropil were evaluated. These included the density of neuronal fibers, the presence of immunolabeled ramifications, the presence of immunolabeled points and varicosities, and the intensity of staining. The morphological features of mature neurons were defined based on the increased amount of cytoplasm in relationship to the nucleus and the decreased proportion of unipolar neurons in relationship to all immunolabeled neurons. The morphological features of neuropil maturation were assumed to be: increased neuronal fiber density, the occurrence of immunolabeled ramifications, varicosities and immunoreactive points, and increased intensity of staining. In order to compare the selected parameters of immunoreactive neuronal populations, the following criteria of classification were established. Immunoreactive neurons were classified, according to the shapes of perikarya, as belonging to one of five categories (round, oval, fusiform, triangular, or multiangular). Additionally, three morphological types of neuronal polarity were distinguished (unipolar, bipolar, and multipolar). The sections were examined with a fluorescence microscope (Eclipse E600, Nikon, Japan) equipped with a confocal imaging system (MicroRadiance, Bio-Rad, UK) supplied with an argon laser (excitation 488/514 nm). Image analysis programs (LaserSharp 2000 v. 2.01 and LaserPix v. 4.0, Bio-Rad, UK) were used to prepare the illustrations. 2.3. Quantitative study The quantitative study involved the assessment of the neuronal density in each of the immunoreactive populations. Sections of the anterior (+1.6 mm from bregma), central (−0.26 mm), and posterior (−0.92 mm) parts of the Cl were chosen under 10×
3.1. Neuropeptide-containing neurons in the Cl Neuronal bodies and neuropil containing NPY, SOM, and VIP were present in the Cl during the entire study period. These neurons were randomly distributed in all parts of the structure. Morphological changes of immunoreactive cell bodies and neuropil were observed, which could be divided into two phases of development. In the first phase (P0–P4), immature neurons were observed. These were characterized by small somata, very often of unipolar or bipolar type, with nuclei surrounded by a narrow rim of cytoplasm and short processes. These morphological features may be attributed to migrating neurons. In the second phase (P7–P28), differentiating neurons were observed, with a greater amount of cytoplasm and longer, prevalently ramified dendrites. The intensity of staining increased inhomogenously. The last period of observation (after P28) was characterized by the occurrence of differentiated neurons of various shapes and clearly differentiated dendritic arbors, as well as visible proximal fragments of the axons. 3.2. NPY immunoreactivity in the Cl 3.2.1. Neurons In the first phase (P0–P4), NPY-ir bipolar and unipolar neurons containing a small volume of cytoplasm were present (Fig. 1A). Neuronal somata of oval and fusiform shapes were most numerous. Only short and dilated fragments of one to three neuronal processes were visible. In the second phase (P7–P28), the neuronal types became more differentiated (Fig. 1D). As well as those previously observed, multipolar neurons occurred for the first time. Oval and fusiform shapes of neuronal somata were most numerous in this phase. The amount of cytoplasm increased gradually. Neuronal processes were narrower and longer than before, and their morphological differentiation into dendrites and axons could be traced more easily. After P28, neurons of multipolar and bipolar types and of variously shaped neuronal somata were observed (Fig. 1G). The initial fragments of axon and two to five dendrites originated from neuronal somata, dividing dichotomically into consecutive branches. The numerical density of NPY-ir neurons increased during the first 2 weeks of life (Fig. 2A). 3.2.2. Neuropil In the first phase (P0–P4), NPY-ir neuropil was poorly developed (Fig. 1A). Short fragments of fibers with varicosities and scattered immunoreactive points were present. In the second phase (P7–P28), an increase in the number of immunoreactive elements and the length of immunoreactive fibers was observed (Fig. 1D). After 4 weeks the morphological structure of NPY-ir neuropil reached its mature form (Fig. 1G). The intensity of staining did
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
567
Fig. 1. Confocal laser scanning images of neuropeptide-containing interneurons and neuropil of the Cl at various stages of development (scale bar = 25 m). (A, D and G) NPY-ir neurons: (A) undifferentiated neuron of oval cell body with a dilated fragment of neuronal process (higher magnification; arrow); neuropil consisting of sparsely distributed immunoreactive points and fibers; (D) bipolar and multipolar neurons of various shapes and fragments of immunoreactive fibers with varicosities (higher magnification); (G) bipolar and multipolar (higher magnification) neurons and completely differentiated neuropil. (B, E and H) SOM-ir neurons: (B) undifferentiated, bipolar and unipolar neurons of round and oval cell bodies with dilated fragments of processes (higher magnification; arrow); neuropil consisting of sparsely distributed immunoreactive points and fibers; (E) bipolar and multipolar neurons of oval, fusiform, and triangular (higher magnification) cell bodies and proximal fragments of dendrites; (H) bipolar and multipolar neurons of multiangular (higher magnification), fusiform, and oval shapes of cell bodies, differentiated intensity of staining is characteristic; neuropil represented by sparsely distributed neuronal fibers and immunoreactive points. (C, F and I) VIP-ir neurons: (C) undifferentiated unipolar and bipolar neurons of oval and fusiform cell bodies with proximal dilated fragment of neuronal process (higher magnification; arrow); neuropil consisting of thin and sparsely distributed immunoreactive fibers and points; (F) bipolar and multipolar (higher magnification) neurons of various shapes; neuropil represented by long fragments of immunoreactive fibers and sparsely distributed points; (I) multipolar and bipolar neurons of triangular (higher magnification), oval, and fusiform somata; long fragments of delicate fibers and immunoreactive points.
568
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
The numerical density of SOM-ir neurons was high but did not change significantly during the whole observation period (Fig. 2B). 3.3.2. Neuropil In the first phase of development (P0–P4), SOM-ir neuropil consisted predominantly of sparsely distributed immunoreactive points and fibers (Fig. 1B). Between P7 and P28 (second phase), the neuropil structure only changed slightly (Fig. 1E). In contrast to the other neuropeptide-containing neuropil elements, the SOMir fibers with varicosities were very delicate, short, and scattered. Immunoreactive ramifications were sporadically observed. A network of more numerous fibers was visible at P21. After 4 weeks of life the number and distribution of SOM-ir fibers and immunoreactive points was characteristic for an adult, although it was much more delicate than in the other studied neuropeptide-containing populations (Fig. 1H). 3.4. VIP immunoreactivity in the Cl 3.4.1. Neurons At P0 and P4 (first phase of development) uni- and bipolar neurons with a very narrow rim of cytoplasm were most frequently observed (Fig. 1C). Wide, proximal fragments of processes were present. Later (P7–P28; second phase), neurons with more differentiated morphology were observed. As well as the prevailing bipolar neurons, cells of multi- and unipolar types were also present (Fig. 1F). During this phase, the amount of cytoplasm and intensity of staining increased, whereas proximal fragments of dendrites became more delicate and narrow after P10. After P28 multi- and bipolar types of neurons were most numerous (Fig. 1I). Their somata were prevalently of oval and fusiform shapes. The initial fragments of axons and dendrites were clearly visible. The numerical density of VIP-ir neurons revealed a transient increase, reaching a maximal value at P21 (Fig. 2C).
Fig. 2. Mean values of numerical densities (number of neuronal profiles per mm2 ) of neuropeptide-containing neurons: (A) NPY; (B) SOM; (C) VIP.
not change in older age groups. NPY-ir fibers and axonal terminals on the surface of unlabeled neuronal somata were frequently observed. 3.3. SOM immunoreactivity in the Cl 3.3.1. Neurons During the first phase of development (P0–P4), bipolar and unipolar neurons of oval, round, and fusiform somata were most numerous (Fig. 1B). Small amounts of cytoplasm and short, wide fragments of processes were characteristic for these cells. In older age groups (P7–P28; second phase), SOM-ir neurons were observed with increasing volumes of cytoplasm and various staining intensities (Fig. 1E). Bipolar and multipolar neurons were most numerous. In this phase, all shapes of neuronal somata were represented. Neurons of round somata gradually became the least numerous. Proximal fragments of two to five narrow dendrites were present, as well as fragment of the axon. After P28 bi- and multipolar neurons of fusiform and multiangular somata were the most frequently observed (Fig. 1H). The intensity of neuronal staining was differentiated.
3.4.2. Neuropil The intensity of neuropil staining was weakest in the first phase of development (P0–P4; Fig. 1C). The immunoreactive fragments of fibers were shorter than in the later phases of development. The ramifications became visible from P4 onwards. During the second phase of development (P7–P28), labeled fragments of VIP-ir fibers with varicosities became longer and the number of immunoreactive points increased (Fig. 1F). After P28 neuropil formed a delicate network of fibers and immunoreactive points (Fig. 1I). In groups P60 and P120 immunoreactive fibers and axonal terminals situated on the surface of unlabeled neuronal cell bodies were observed. The intensity of neuropil staining became adult-like after 4 weeks of life. 4. Discussion 4.1. At the moment of birth the morphological development of claustral interneurons is incomplete The most important finding of our studies is the fact that neuropeptide-containing interneurons of the rat Cl are still developing in the earliest postnatal period. Changes in their number and morphology, as well as the morphology of immunoreactive neuropil, can be observed. The sequence of morphological changes allows us to distinguish two phases of development. The first phase of early development, during which undifferentiated neurons are observed, extends from P0 to P4 in all of the studied subpopulations. It might be that some of these neurons represent migrating cells, the neurogenesis of which
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
takes place in subcortical structures, like the ganglionic eminence, during the prenatal period. Their settling down period in the Cl may be extended until P4. In the second phase, some morphological features appear that are characteristic for the adult population, although the frequency and intensity of their occurrence are still different from those observed in the adult structure. This phase extends to P28, during which time neuronal shapes become more differentiated, numerical density of NPY- and VIP-ir neurons increases transiently, and neuropil attains its adult-like structure. These changes correspond to the development and stabilization of claustral connections, processes of synaptogenesis, and synaptic plasticity [22,26,44]. After P28 the morphological changes are complete and the adult-like structure is finally achieved. The dynamics of these changes are characteristic for each immunoreactive neuronal population. 4.2. Postnatal developmental changes of neuropeptide-containing claustral interneurons are observed, but their dynamics are differentiated Studies based on the expression of selected transcription factors, comparative anatomy, and immunocytochemical methods have allowed the introduction of a general concept for the claustroamygdaloid complex, which is made up of the amygdalar basolateral complex (lateral, basolateral, and basomedial amygdalar nuclei) and the claustral complex (dorsal claustrum or claustrum proper and endopiriform nucleus) [28,33,43,52]. The expression of developmental regulatory genes confirms the common origin of the dorsal part of the Cl and the basolateral nucleus of amygdala from the lateral part of the pallium [33,43]. Developmental studies on the basolateral complex of amygdala indicate a character of observed in this structure processes similar to those occurring in the dorsal claustrum [39]. The interneurons of the basolateral complex of amygdala and the Cl are inhomogenous in their immunocytochemical characteristics. Numerous neuropeptides (CCK, NPY, SOM, and VIP), CaBP (such as calbindin D28k, calretinin, parvalbumin), and NOS were detected in the interneurons of the Cl [13,14,15,22,27] and amygdala [24,32]. Both pre- and postnatal development of CaBP- and NOS-ir interneurons of the Cl were studied [13,22,26]. Generally, the changes in neuronal morphology, intensity of staining and number of immunoreactive neurons were revealed in these studies. However, the intensity and dynamics of these processes were not identical in all developing interneurons; they were differentiated and related to the type of immunocytochemical marker. Based on our results, we have distinguished some characteristic differences in the development of neuropeptide-containing interneurons. This is most clearly demonstrated by the changes in numerical density values of NPY- and VIP-ir interneurons, in comparison to the values of SOM-ir neurons. In addition, differences were observed in the most frequently occurring neuronal types (multipolar in NPY- and VIP-ir versus bipolar in SOM-ir neurons). Furthermore, differentiated dynamics of neuropil development were reported, immunoreactive to each neuropeptide. These results may suggest the existence of separate regulatory developmental mechanisms for various interneurons, characterized by neuropeptide content. 4.3. Changes in the number of neuropeptide-containing neurons may indicate the importance of their physiological role during postnatal development The values of numerical density of NPY- and VIP-ir neurons increased transiently in the studied period. In the case of NPY-ir
569
neurons, the maximum values were observed after 2 weeks, and in the case of VIP-ir neurons, 1 week later. The changes in the number of NPY-ir neurons, similar to those observed in the Cl, occured in the occipital region of the rat cerebral cortex [1,9]. The decrease in the number of immunoreactive neurons is predominantly the result of neuronal death and, to a lesser degree, the result of a decrease in neuropeptide expression [1]. The role of NPY during early claustral development may be connected with the regulation of Ca++ ion inflow [2,51] and the regulation of neuronal fiber growth [11]. The influence of NPY on glutamate release may be indicative of a protective action against excitotoxicity. These important functions of NPY might be carried out by means of NPY receptors (type Y1 and Y2), the presence of which was described in the Cl [40]. Transient increases of both VIP mRNA and neuropeptide expression during development were observed in the frontal and parietal regions of the rat cerebral cortex [20]. The decrease in the numerical density of VIP-ir neurons might, according to some authors, be explained primarily by the decrease in neuropeptide content and, to a lesser degree, by neuronal death [10]. The role of VIP in claustral development is concerned with the regulation of neuronal proliferation and differentiation [16,36]. VIP may influence the length of the cell cycle by decreasing the G1 and S phases [21]. The potential function of VIP in relation to the claustral neurons may also be linked with its neuroprotective and neurotrophic function [6,19]. SOM-ir neurons in the Cl, although numerous, do not reveal any significant changes in their numerical density during the postnatal period. The intensity of SOM staining was differentiated throughout our developmental studies. This may indicate specific mechanisms of developmental regulation in SOM-ir neurons. It may also be speculated that differentiated expression of SOM may reveal some kind of relationship with the different place of origin of this subpopulation of neurons. During the early postnatal period, SOM may be connected with the growth control of the neuronal population [37]. It may also influence the process of neuronal migration [46,56]. 4.4. Developmental changes occur in neuropeptide-containing neuropil Maturation of the claustral connections remains under the control of neurotrophic factors produced by claustral neurons and axonal terminals of projecting neurons, originating in various cortical regions and subcortical structures [18,41,53]. Therefore, the length of the period of claustral neuropil development may be related, at least partially, to the development of connected structures, in particular, neocortical regions. According to our observations, neuropil development is completed after 4 weeks of postnatal life. This is concerned with final adjustment of claustral projections originating in various cortical regions and subcortical structures. The principal neurotransmitter of these projections is glutamate [42]. Simultaneously, other types of connections are developed between claustral interneurons, and between interneurons and projecting neurons of the Cl. GABA must be considered as the dominating neurotransmitter of these connections, in colocalization with various kinds of neuromodulators (neuropeptides) [17]. Interestingly, and importantly, the increase of vesicular glutamate transporter (VGLUT2) is thought to be proof of the development of stimulating connections [44] that takes place in the Cl at the moment of birth and early in the postnatal period. At the same time, intensive synaptogenesis is continued in the Cl, which is characterized by the transient increase of SNAP-25 (assessed by immunoblotting; our unpublished data). The maximal
570
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
value of this transient increase occurs at P21. This pattern of SNAP-25 reactivity may reflect the development of intraclaustral connections and, therefore, the number of synapses. Subsequently, SNAP-25 content decreases, probably because of the changing influence of neurotrophic factors and cortical activity. The decrease in synapse formation during the later phase of postnatal life may correspond, to some extent, to the process of neuronal death, partly of apoptotic character, described previously in the rat in the early postnatal period [30]. Finally, the occurrence of NPY- and VIP-ir fibers situated around unlabeled neuronal cell bodies may confirm the inhibitory function of neuropeptide-containing neuropil elements in relation to projecting neurons and interneurons of the Cl. 4.5. The origin and path of migration of claustral neuropeptide-containing interneurons are still unknown Recently published studies point to a subcortical origin, possibly the medial ganglionic eminence (MGE), for a proportion of claustral GABA-ergic interneurons—at least those containing calbindin [29]. MGE and caudal ganglionic eminence (CGE) are the sources of CaBP- and somatostatin-containing neurons migrating to the cortex and numerous structures of the limbic system [8,38]. GABAergic interneurons originating in MGE and CGE were found in the structures of claustroamygdaloid complex. Taking into account the common colocalization of CaBP with neuropeptides in GABA-ergic interneurons of the Cl and amygdaloid body, the dual – cortical and subcortical – origin of the neuropeptide-containing interneurons may not be excluded. Our results indicate that the development of claustral interneurons and neuropil extend into the postnatal period. This process, according to our observations, terminates after the fourth postnatal week. However, we focused only on selected aspects of the complex problem of claustral development, which consist of numerous qualitative and quantitative processes. A specific cutoff point, terminating these processes cannot be established precisely. Despite this, the fundamental role of the early postnatal period for the proper development of the rat Cl must be taken into account in future studies on this structure. Conflict of interest No conflict of interests is claimed by any author of this publication. Acknowledgements This research was supported by funds from the Polish State Committee of Scientific Research grant number: W-705. The technical assistance of Mrs. S. Scislowska and Mr. T. Alexander is greatly appreciated. References [1] J. Antonopoulos, G.C. Papadopoulos, H. Michaloudi, M.E. Cavanagh, J.G. Parnavelas, Postnatal development of neuropeptide Y-containing neurons in the visual cortex of normal- and dark-reared rats, Neurosci. Lett. 145 (1992) 75. [2] A. Bacci, J.R. Huguenard, D.A. Prince, Differential modulation of synaptic transmission by neuropeptide Y in rat neocortical neurons, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 17125. [3] R.B. Banati, G.W. Goerres, C. Tjoa, J.P. Aggleton, P. Grasby, The functional anatomy of visual-tactile integration in man: a study using positron emission tomography, Neuropsychologia 38 (2000) 115–124. [4] S.A. Bayer, J. Altman, Development of the endopiriform nucleus and the claustrum in the rat brain, Neuroscience 45 (1991) 391. [5] H. Braak, E. Braak, Neuronal types in the claustrum of man, Anat. Embryol. 163 (1982) 447.
[6] D.E. Brenneman, G. Glazner, J.M. Hill, J. Hauser, A. Davidson, I. Gozes, VIP neurotrophism in the central nervous system: multiple effectors and identification of a femtomolar-acting neuroprotective peptide, Ann. N.Y. Acad. Sci. 865 (1998) 207. [7] G.A. Calvert, P.C. Hansen, S.D. Iversen, M.J. Brammer, Detection of audio-visual integration sites in humans by application of electrophysiological criteria to the BOLD effect, Neuroimage 14 (2001) 427. [8] R.S. Carney, T.B. Alfonso, D. Cohen, H. Dai, S. Nery, B. Stoica, J. Slotkin, B.S. Bregman, G. Fishell, J.G. Corbin, Cell migration along the lateral cortical stream to the developing basal telencephalic limbic system, J. Neurosci. 26 (2006) 11562–11574. [9] M.E. Cavanagh, J.G. Parnavelas, Development of vasoactive-intestinalpolypeptide-immunoreactive neurons in the rat occipital cortex: a combined immunohistochemical–autoradiographic study, J. Comp. Neurol. 284 (1989) 637. [10] M.E. Cavanagh, J.G. Parnavelas, Development of neuropeptide Y (NPY) immunoreactive neurons in the rat occipital cortex: a combined immunohistochemical–autoradiographic study, J. Comp. Neurol. 297 (1990) 553. [11] G. Chen, A.N. Van Den Pol, Multiple NPY receptors coexist in pre- and postsynaptic sites: inhibition of GABA release in isolated self-innervating SCN neurons, J. Neurosci. 16 (1996) 7711. [12] F.C. Crick, C. Koch, What is the function of the claustrum? Philos. Trans. R. Soc. Lond. B Biol. Sci. 360 (2005) 1271–1279. [13] J.C. Davila, M.A. Real, L. Olmos, I. Legaz, L. Medina, S. Guirado, Embryonic and postnatal development of GABA, calbindin, calretinin, and parvalbumin in the mouse claustral complex, J. Comp. Neurol. 481 (2005) 42. [14] L.R. Edelstein, F.J. Denaro, The claustrum: a historical review of its anatomy, physiology, cytochemistry and functional significance, Cell. Mol. Biol. 50 (2004) 675–702. [15] L.E. Eiden, E. Mezey, R.L. Eskay, M.C. Beinfeld, M. Palkovits, Neuropeptide content and connectivity of the rat claustrum, Brain Res. 523 (1990) 245. [16] G.W. Glazner, P. Gressens, S.J. Lee, G. Gibney, I. Gozes, Y. Gozes, D.E. Brenneman, J.M. Hill, Activity-dependent neurotrophic factor: a potent regulator of embryonic growth and development, Anat. Embryol. 200 (1999) 65. [17] S.M. Gomez-Urquijo, I. Gutierrez-Ibarluzea, J.L. Bueno-Lopez, C. Reblet, Percentage incidence of gamma-aminobutyric acid neurons in the claustrum of the rabbit and comparison with the cortex and putamen, Neurosci. Lett. 282 (2000) 177. [18] C.S. Goodman, Mechanisms and molecules that control growth cone guidance, Annu. Rev. Neurosci. 19 (1996) 341. [19] I. Gozes, I. Divinsky, I. Pilzer, M. Fridkin, D.E. Brenneman, A.D. Spier, From vasoactive intestinal peptide (VIP) through activity-dependent neuroprotective protein (ADNP) to NAP: a view of neuroprotection and cell division, J. Mol. Neurosci. 20 (2003) 315. [20] I. Gozes, Y. Shani, W.H. Rostene, Developmental expression of the VIP-gene in brain and intestine, Brain Res. 388 (1987) 137. [21] P. Gressens, B. Paindaveine, J.M. Hill, P. Evrard, D.E. Brenneman, Vasoactive intestinal peptide shortens both G1 and S phases of neural cell cycle in whole postimplantation cultured mouse embryos, Eur. J. Neurosci. 10 (1998) 1734. [22] S. Guirado, M.A. Real, J.L. Olmos, J.C. Davila, Distinct types of nitric oxideproducing neurons in the developing and adult mouse claustrum, J. Comp. Neurol. 465 (2003) 431–444. [23] N. Hadjikhani, P.E. Roland, Cross-modal transfer of information between the tactile and the visual representations in the human brain: a positron emission tomographic study, J. Neurosci. 18 (1998) 1072–1084. [24] S. Kemppainen, A. Pitkanen, Distribution of parvalbumin, calretinin, and calbindin-D(28k) immunoreactivity in the rat amygdaloid complex and colocalization with gamma-aminobutyric acid, J. Comp. Neurol. 426 (2000) 441–467. [25] P. Kowianski, J. Morys, J. Dziewiatkowski, Z. Karwacki, B. Bobek-Billewicz, O. Narkiewicz, Neurons of the claustrum projecting to the motor and somatosensory cortex in the rabbit—assessment with the method of retrograde transport of fluorescent tracers, Folia Morphol. 55 (1996) 350–352. [26] P. Kowianski, J.M. Morys, S. Wojcik, J. Dziewiatkowski, J. Morys, Postnatal development of NOS-ir neurons in the rat claustrum, Folia Morphol. 62 (2003) 11–17. [27] P. Kowianski, J.P. Timmermans, J. Morys, Differentiation in the immunocytochemical features of intrinsic and cortically projecting neurons in the rat claustrum—combined immunocytochemical and axonal transport study, Brain Res. 905 (2001) 63. [28] I. Legaz, M. Garcia-Lopez, L. Medina, Subpallial origin of part of the calbindinpositive neurons of the claustral complex and piriform cortex, Brain Res. Bull. 66 (2005) 470–474. [29] I. Legaz, G. Gonzalez, F. De Castro, J.L. Rubenstein, L. Puelles, Expression of Dbx1, Neurogenin 2, Semaphorin 5A, Cadherin 8, and Emx1 distinguish ventral and lateral pallial histogenetic divisions in the developing mouse claustroamygdaloid complex, J. Comp. Neurol. 474 (2004) 504–523. ˛ [30] B. Maciejewska, J. Mory´s, B. Berdel, J. Dziewiatkowski, O. Narkiewicz, The development of the rat claustrum—a study using morphometric and in situ DNA end labelling techniques, Eur. J. Anat. 1 (1997) 137. [31] L. Mamos, Morphology of claustral neurons in the rat, Folia Morphol. 43 (1984) 73.
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571 [32] A.J. McDonald, J.C. Pearson, Coexistence of GABA and peptide immunoreactivity in non-pyramidal neurons of the basolateral amygdala, Neurosci. Lett. 100 (1989) 53–58. [33] L. Medina, I. Legaz, G. Gonzalez, F. De Castro, J.L. Rubenstein, L. Puelles, Expression of Dbx1, Neurogenin 2, Semaphorin 5A, Cadherin 8, and Emx1 distinguish ventral and lateral pallial histogenetic divisions in the developing mouse claustroamygdaloid complex, J. Comp. Neurol. 474 (2004) 504–523. [34] P. Mohapel, D.K. Hannesson, L.L. Armitage, G.W. Gillespie, M.E. Corcoran, Claustral lesions delay amygdaloid kindling in the rat, Epilepsia 41 (2000) 1095. [35] P. Mohapel, X. Zhang, G.W. Gillespie, J. Chlan-Fourney, D.K. Hannesson, S.M. Corley, X.M. Li, M.E. Corcoran, Kindling of claustrum and insular cortex: comparison to perirhinal cortex in the rat, Eur. J. Neurosci. 13 (2001) 1501. [36] T.W. Moody, J.M. Hill, R.T. Jensen, VIP as a trophic factor in the CNS and cancer cells, Peptides 24 (2003) 163. [37] J. Nacher, C. Ramirez, A. Molowny, C. Lopez-Garcia, Ontogeny of somatostatin immunoreactive neurons in the medial cerebral cortex and other cortical areas of the lizard Podarcis hispanica, J. Comp. Neurol. 374 (1996) 118. [38] S. Nery, G. Fishell, J.G. Corbin, The caudal ganglionic eminence is a source of distinct cortical and subcortical cell populations, Nat. Neurosci. 5 (2002) 1279–1287. ´ [39] J.L. Olmos, M.A. Real, L. Medina, S. Guirado, J.C. Davila, Distribution of nitric oxide-producing neurons in the developing and adult mouse amygdalar basolateral complex, Brain Res. Bull. 66 (2005) 465–469. [40] R.M.C. Parker, H. Herzog, Regional distribution of Y-receptor subtype mRNAs in rat brain, Eur. J. Neurosci. 11 (1999) 1431. [41] M. Pascual, E. Pozas, M.J. Barallobre, M. Tessier-Lavigne, E. Soriano, Coordinated functions of Netrin-1 and Class 3 secreted Semaphorins in the guidance of reciprocal septohippocampal connections, Mol. Cell. Neurosci. 26 (2004) 24. [42] F. Perez-Cerda, L. Martinez-Millan, C. Matute, Anatomical evidence for glutamate and/or aspartate as neurotransmitters in the geniculo-, claustro-, and cortico-cortical pathways to the cat striate cortex, J. Comp. Neurol. 373 (1996) 422. [43] L. Puelles, E. Kuwana, E. Puelles, A. Bulfone, K. Shimamura, J. Keleher, S. Smiga, J.L. Rubenstein, Pallial and subpallial derivatives in the embryonic chick and mouse telencephalon, traced by the expression of the genes Dlx-2, Emx-1, Nkx2.1, Pax-6, and Tbr-1, J. Comp. Neurol. 424 (2000) 409–438.
571
[44] M.A. Real, J.C. Davila, S. Guirado, Immunohistochemical localization of the vesicular glutamate transporter VGLUT2 in the developing and adult mouse claustrum, J. Chem. Neuroanat. 31 (2006) 169–177. [45] C. Reblet, A. Alejo, R.I. Blanco-Santiago, J. Mendizabal-Zubiaga, M. Fuentes, J.L. Bueno-Lopez, Neuroepithelial origin of the insular and endopiriform parts of the claustrum, Brain Res. Bull. 57 (2002) 495–497. [46] J.P. Schwartz, Z. Ji, J. Epelbaum, Somatostatin as a neurotrophic factor. Which receptor/second messenger transduction system is involved? Perspect. Dev. Neurobiol. 5 (1998) 427. [47] A.H. Sheerin, K. Nylen, X. Zhang, D.M. Saucier, M.E. Corcoran, Further evidence for a role of the anterior claustrum in epileptogenesis, Neuroscience 125 (2004) 57. [48] S. Shiosaka, K. Takatsuki, M. Sakanaka, S. Inagaki, H. Takagi, E. Senba, Y. Kawai, H. Iida, H. Minagawa, Y. Hara, T. Matsuzaki, M. Tohyama, Ontogeny of somatostatin-containing neuron system of the rat: immunohistochemical analysis. II. Forebrain and diencephalons, J. Comp. Neurol. 204 (1982) 211. [49] K.B. Sims, D.L. Hoffman, S.I. Said, E.Z. Zimmerman, Vasoactive intestinal polypeptide (VIP) in mouse and rat brain: an immunocytochemical study, Brain Res. 186 (1980) 165. [50] B. Spahn, H. Braak, Percentage of projection neurons and various types of interneurons in the human claustrum, Acta Anat. 122 (1985) 245. [51] L. Sun, L.H. Philipson, R.J. Miller, Regulation of K+ and Ca++ channels by a family of neuropeptide Y receptors, J. Pharmacol. Exp. Ther. 284 (1998) 625. [52] L.W. Swanson, G.D. Petrovich, What is the amygdala? Trends Neurosci. 21 (1998) 323–331. [53] R. Tuttle, B.L. Schlaggar, J.E. Braisted, D.D. O’Leary, Maturation-dependent upregulation of growth-promoting molecules in developing cortical plate controls thalamic and cortical neurite growth, J. Neurosci. 15 (1995) 3039. [54] A.P. Weible, C. Weiss, J.F. Disterhoft, Connections of the caudal anterior cingulate cortex in rabbit: neural circuitry participating in the acquisition of trace eyeblink conditioning, Neuroscience 145 (2007) 288–302. [55] B.L. Wilhite, T.J. Teyler, C. Hendricks, Functional relation of rodent claustral–entorhinal–hippocampal system, Brain Res. 365 (1986) 54. [56] E. Yacubova, H. Komuro, Stage-specific control of neuronal migration by somatostatin, Nature 415 (2002) 77.
Contents lists available at ScienceDirect
Brain Research Bulletin journal homepage: www.elsevier.com/locate/brainresbull
Research report
NPY-, SOM- and VIP-containing interneurons in postnatal development of the rat claustrum ∗ ´ ˛ ´ Przemysław Kowianski , Joanna M. Mory´s, Jerzy Dziewiatkowski, Sławomir Wojcik, Justyna Sidor-Kaczmarek, Janusz Mory´s Department of Anatomy and Neurobiology, Medical University of Gda´ nsk, 1 D˛ebinki Street, 80-211 Gda´ nsk, Poland
a r t i c l e
i n f o
Article history: Received 18 September 2007 Received in revised form 3 March 2008 Accepted 16 April 2008 Available online 13 May 2008 Keywords: Amygdaloid body Claustrum Development Interneurons Neuropeptides
a b s t r a c t A growing body of evidence indicates the common origin of the claustrum, endopiriform nucleus, and the basolateral nuclear complex of amygdala from the lateral and ventral parts of the pallium, as the claustroamygdaloid complex. It seems very probable that at least some of the claustral interneurons derive from subcortical sources. The postnatal development of neuropeptide Y-, somatostatin- and vasoactive intestinal polypeptidecontaining interneurons was studied during the 4 postnatal months (P0–P120; P, postnatal day). The study was conducted on 45 Wistar rats of both sexes. Our results indicate that neuropeptide-containing interneurons are not morphologically mature at the moment of birth. The characteristic features of neuronal bodies and the relatively long period of postnatal development may indicate their migration from the subcortical neurogenetic centers. Morphological changes in the neuropil are also reported. Although developmental patterns differ between various neuropeptide-containing neuronal subpopulations, two phases of development can be distinguished in each of them: the early phase (P0–P4) during which undifferentiated neurons and neuropil dominate, and the late phase (P7–P28) during which the characteristic features of an adult-like structure gradually appear. Later these observed developmental changes are terminated. The postnatal development of neuropeptide-containing interneurons is completed after 4 weeks of life. This period, which is important for the structural and functional development of the claustrum, must be taken into account in future studies on this structure. © 2008 Elsevier Inc. All rights reserved.
1. Introduction The claustrum (Cl; dorsal claustrum or claustrum proper) is a telencephalic structure characterized by extensive, reciprocal connections with numerous neocortical areas [14]. A growing body of evidence indicates that it plays a role in processes of cross-modal sensory integration [3,7,12,23], memory [54,55] and epileptogenesis [47]. Because of its numerous cortical connections, the Cl is able to play a particular role enabling the transfer of sensory information of various modalities between cortical areas. The special function of the Cl is also concerned with kindling propagation, as well as with the generalization of seizures [34,35]. The Cl consists of two morphologically distinguished neuronal subpopulations [31]. The prevalent group consists of glutamater-
∗ Corresponding author. Tel.: +48 58 349 1401; fax: +48 58 349 1421. ´ E-mail address: [email protected] (P. Kowianski). 0361-9230/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2008.04.004
gic projecting neurons with spiny dendrites. The second group of GABA-ergic interneurons, represented by variously shaped somata and aspiny dendrites, constitute, in various species, between 7 and 12% of all claustral neurons [5,17,25,50]. Apart from GABA, the occurrence of neuropeptides such as neuropeptide Y (NPY), somatostatin (SOM), vasoactive intestinal polypeptide (VIP) and cholecystokinin (CCK) has been described in the Cl [14,15,27,48,49]. The expression patterns of developmental regulatory genes in mice indicate the origin of the Cl to be from the lateral and ventral pallial divisions [33,43,45]. Neurogenesis of the claustral neurons in the cortical neuroepithelium of the rat was estimated to be between E15 and E16 [4]. However, there is evidence that at least some of the claustral interneurons may originate in the subpallium and migrate tangentially during development [29,43]. According to previously published results, the development of the Cl is not complete at the moment of birth. The increase of its total volume and the loss of nearly 30% of neurons, including apoptotic death, take place after birth [30]. Moreover, characteristic
566
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
morphological changes among interneurons containing calciumbinding proteins (CaBP) and nitric oxide synthase (NOS) have been described [13,22,26]. Neuropeptide-containing interneurons may be very important for the initiation of integrative function and memory processes coordinated by the Cl [12]. The stabilizing influence of peptide neuromodulators on the activity of claustral projecting neurons may be crucial for the development of claustro-cortical connections. Our studies focused on the morphological changes of claustral interneurons, immunoreactive to three neuropeptides (NPY, SOM and VIP), during the first 4 postnatal months.
magnification. The borders of Cl were marked as separate inclusion areas under low magnification (4×). The neuronal profiles were counted and the numerical density (number of cell profiles per mm2 ) was calculated with the aid of a 20× objective lens in systematic random test frames of the selected area using a C.A.S.T. grid system (Computer Assisted Stereological Tool, Olympus, Denmark) and microscope (BX-51, Olympus, Japan). 2.4. Statistics All the calculations were performed using Excel 2000 (Microsoft, USA). Raw data concerning the number of cells per mm2 were collected. Mean values and standard deviations were calculated for each group. The comparisons of neuronal densities between various immunoreactive neuronal populations were performed using a nonparametric multiple comparison test (Steel–Dwas test, Tukey equivalent test).
2. Materials and methods
3. Results Forty-five Wistar rats of both sexes were used in this study. Animal care and treatment guidelines established by the local Ethical Committee, as well as standards defined by the European Communities Council Directive of 24 November 1986 (86/609/EEC), were followed. The animals were divided into nine age groups: P0, P4, P7, P14, P21, P28, P60, P90, and P120, with five animals in each group. The animals were deeply anaesthetized with sodium pentobarbital (thiopental sodium, Biochemie GmBH, Germany; 80 mg/kg of body weight, i.p.) and transcardially perfused with 200 ml of physiological saline (at pH 7.4) followed by 400 ml of 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS, at pH 7.4 and 4 ◦ C). After removal from the skull, the brains were immersed in the same fixative for 120 min and stored in 30% sucrose in 0.1 M PBS (at pH 7.4 and 4 ◦ C) for at least 24 h. Subsequently, the brains were frozen and sectioned coronally into 40-m-thick serial sections on a sliding microtome. 2.1. Immunocytochemical procedure The free-floating sections were processed for fluorescence immunocytochemistry. Three primary antibodies were used in the study: anti-NPY (rabbit; 1:400, Affiniti, UK), anti-SOM (rabbit; 1:300, Euro-Diagnostica AB, Sweden), and anti-VIP (rabbit; 1:100, Cappel ICN Pharmaceuticals Inc., USA). After incubation for 1 h in a blocking solution containing 0.1% bovine serum albumin in PBS and 10% normal goat serum, the sections were incubated overnight in primary antibodies at room temperature. Subsequently, they were rinsed in PBS and incubated in the secondary antibody (goat anti-rabbit; conjugated to indocarbocyanine [Cy-3] used at 1:800, Jackson ImmunoResearch, USA) for 1 h at room temperature. The sections were rinsed in PBS, mounted on slides, dried, and coverslipped with Vectashield (Vector Laboratories Inc., USA). In order to confirm the specificity of staining, the sections were processed by immunocytochemical procedure with the omission of primary or secondary antibody. In all cases the omission tests resulted in a lack of specific labeling. 2.2. Qualitative study In each immunocytochemically distinguished neuronal subpopulation, the shapes of the perikarya and their polarity were studied. Additionally, the morphological features of the neuropil were evaluated. These included the density of neuronal fibers, the presence of immunolabeled ramifications, the presence of immunolabeled points and varicosities, and the intensity of staining. The morphological features of mature neurons were defined based on the increased amount of cytoplasm in relationship to the nucleus and the decreased proportion of unipolar neurons in relationship to all immunolabeled neurons. The morphological features of neuropil maturation were assumed to be: increased neuronal fiber density, the occurrence of immunolabeled ramifications, varicosities and immunoreactive points, and increased intensity of staining. In order to compare the selected parameters of immunoreactive neuronal populations, the following criteria of classification were established. Immunoreactive neurons were classified, according to the shapes of perikarya, as belonging to one of five categories (round, oval, fusiform, triangular, or multiangular). Additionally, three morphological types of neuronal polarity were distinguished (unipolar, bipolar, and multipolar). The sections were examined with a fluorescence microscope (Eclipse E600, Nikon, Japan) equipped with a confocal imaging system (MicroRadiance, Bio-Rad, UK) supplied with an argon laser (excitation 488/514 nm). Image analysis programs (LaserSharp 2000 v. 2.01 and LaserPix v. 4.0, Bio-Rad, UK) were used to prepare the illustrations. 2.3. Quantitative study The quantitative study involved the assessment of the neuronal density in each of the immunoreactive populations. Sections of the anterior (+1.6 mm from bregma), central (−0.26 mm), and posterior (−0.92 mm) parts of the Cl were chosen under 10×
3.1. Neuropeptide-containing neurons in the Cl Neuronal bodies and neuropil containing NPY, SOM, and VIP were present in the Cl during the entire study period. These neurons were randomly distributed in all parts of the structure. Morphological changes of immunoreactive cell bodies and neuropil were observed, which could be divided into two phases of development. In the first phase (P0–P4), immature neurons were observed. These were characterized by small somata, very often of unipolar or bipolar type, with nuclei surrounded by a narrow rim of cytoplasm and short processes. These morphological features may be attributed to migrating neurons. In the second phase (P7–P28), differentiating neurons were observed, with a greater amount of cytoplasm and longer, prevalently ramified dendrites. The intensity of staining increased inhomogenously. The last period of observation (after P28) was characterized by the occurrence of differentiated neurons of various shapes and clearly differentiated dendritic arbors, as well as visible proximal fragments of the axons. 3.2. NPY immunoreactivity in the Cl 3.2.1. Neurons In the first phase (P0–P4), NPY-ir bipolar and unipolar neurons containing a small volume of cytoplasm were present (Fig. 1A). Neuronal somata of oval and fusiform shapes were most numerous. Only short and dilated fragments of one to three neuronal processes were visible. In the second phase (P7–P28), the neuronal types became more differentiated (Fig. 1D). As well as those previously observed, multipolar neurons occurred for the first time. Oval and fusiform shapes of neuronal somata were most numerous in this phase. The amount of cytoplasm increased gradually. Neuronal processes were narrower and longer than before, and their morphological differentiation into dendrites and axons could be traced more easily. After P28, neurons of multipolar and bipolar types and of variously shaped neuronal somata were observed (Fig. 1G). The initial fragments of axon and two to five dendrites originated from neuronal somata, dividing dichotomically into consecutive branches. The numerical density of NPY-ir neurons increased during the first 2 weeks of life (Fig. 2A). 3.2.2. Neuropil In the first phase (P0–P4), NPY-ir neuropil was poorly developed (Fig. 1A). Short fragments of fibers with varicosities and scattered immunoreactive points were present. In the second phase (P7–P28), an increase in the number of immunoreactive elements and the length of immunoreactive fibers was observed (Fig. 1D). After 4 weeks the morphological structure of NPY-ir neuropil reached its mature form (Fig. 1G). The intensity of staining did
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
567
Fig. 1. Confocal laser scanning images of neuropeptide-containing interneurons and neuropil of the Cl at various stages of development (scale bar = 25 m). (A, D and G) NPY-ir neurons: (A) undifferentiated neuron of oval cell body with a dilated fragment of neuronal process (higher magnification; arrow); neuropil consisting of sparsely distributed immunoreactive points and fibers; (D) bipolar and multipolar neurons of various shapes and fragments of immunoreactive fibers with varicosities (higher magnification); (G) bipolar and multipolar (higher magnification) neurons and completely differentiated neuropil. (B, E and H) SOM-ir neurons: (B) undifferentiated, bipolar and unipolar neurons of round and oval cell bodies with dilated fragments of processes (higher magnification; arrow); neuropil consisting of sparsely distributed immunoreactive points and fibers; (E) bipolar and multipolar neurons of oval, fusiform, and triangular (higher magnification) cell bodies and proximal fragments of dendrites; (H) bipolar and multipolar neurons of multiangular (higher magnification), fusiform, and oval shapes of cell bodies, differentiated intensity of staining is characteristic; neuropil represented by sparsely distributed neuronal fibers and immunoreactive points. (C, F and I) VIP-ir neurons: (C) undifferentiated unipolar and bipolar neurons of oval and fusiform cell bodies with proximal dilated fragment of neuronal process (higher magnification; arrow); neuropil consisting of thin and sparsely distributed immunoreactive fibers and points; (F) bipolar and multipolar (higher magnification) neurons of various shapes; neuropil represented by long fragments of immunoreactive fibers and sparsely distributed points; (I) multipolar and bipolar neurons of triangular (higher magnification), oval, and fusiform somata; long fragments of delicate fibers and immunoreactive points.
568
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
The numerical density of SOM-ir neurons was high but did not change significantly during the whole observation period (Fig. 2B). 3.3.2. Neuropil In the first phase of development (P0–P4), SOM-ir neuropil consisted predominantly of sparsely distributed immunoreactive points and fibers (Fig. 1B). Between P7 and P28 (second phase), the neuropil structure only changed slightly (Fig. 1E). In contrast to the other neuropeptide-containing neuropil elements, the SOMir fibers with varicosities were very delicate, short, and scattered. Immunoreactive ramifications were sporadically observed. A network of more numerous fibers was visible at P21. After 4 weeks of life the number and distribution of SOM-ir fibers and immunoreactive points was characteristic for an adult, although it was much more delicate than in the other studied neuropeptide-containing populations (Fig. 1H). 3.4. VIP immunoreactivity in the Cl 3.4.1. Neurons At P0 and P4 (first phase of development) uni- and bipolar neurons with a very narrow rim of cytoplasm were most frequently observed (Fig. 1C). Wide, proximal fragments of processes were present. Later (P7–P28; second phase), neurons with more differentiated morphology were observed. As well as the prevailing bipolar neurons, cells of multi- and unipolar types were also present (Fig. 1F). During this phase, the amount of cytoplasm and intensity of staining increased, whereas proximal fragments of dendrites became more delicate and narrow after P10. After P28 multi- and bipolar types of neurons were most numerous (Fig. 1I). Their somata were prevalently of oval and fusiform shapes. The initial fragments of axons and dendrites were clearly visible. The numerical density of VIP-ir neurons revealed a transient increase, reaching a maximal value at P21 (Fig. 2C).
Fig. 2. Mean values of numerical densities (number of neuronal profiles per mm2 ) of neuropeptide-containing neurons: (A) NPY; (B) SOM; (C) VIP.
not change in older age groups. NPY-ir fibers and axonal terminals on the surface of unlabeled neuronal somata were frequently observed. 3.3. SOM immunoreactivity in the Cl 3.3.1. Neurons During the first phase of development (P0–P4), bipolar and unipolar neurons of oval, round, and fusiform somata were most numerous (Fig. 1B). Small amounts of cytoplasm and short, wide fragments of processes were characteristic for these cells. In older age groups (P7–P28; second phase), SOM-ir neurons were observed with increasing volumes of cytoplasm and various staining intensities (Fig. 1E). Bipolar and multipolar neurons were most numerous. In this phase, all shapes of neuronal somata were represented. Neurons of round somata gradually became the least numerous. Proximal fragments of two to five narrow dendrites were present, as well as fragment of the axon. After P28 bi- and multipolar neurons of fusiform and multiangular somata were the most frequently observed (Fig. 1H). The intensity of neuronal staining was differentiated.
3.4.2. Neuropil The intensity of neuropil staining was weakest in the first phase of development (P0–P4; Fig. 1C). The immunoreactive fragments of fibers were shorter than in the later phases of development. The ramifications became visible from P4 onwards. During the second phase of development (P7–P28), labeled fragments of VIP-ir fibers with varicosities became longer and the number of immunoreactive points increased (Fig. 1F). After P28 neuropil formed a delicate network of fibers and immunoreactive points (Fig. 1I). In groups P60 and P120 immunoreactive fibers and axonal terminals situated on the surface of unlabeled neuronal cell bodies were observed. The intensity of neuropil staining became adult-like after 4 weeks of life. 4. Discussion 4.1. At the moment of birth the morphological development of claustral interneurons is incomplete The most important finding of our studies is the fact that neuropeptide-containing interneurons of the rat Cl are still developing in the earliest postnatal period. Changes in their number and morphology, as well as the morphology of immunoreactive neuropil, can be observed. The sequence of morphological changes allows us to distinguish two phases of development. The first phase of early development, during which undifferentiated neurons are observed, extends from P0 to P4 in all of the studied subpopulations. It might be that some of these neurons represent migrating cells, the neurogenesis of which
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
takes place in subcortical structures, like the ganglionic eminence, during the prenatal period. Their settling down period in the Cl may be extended until P4. In the second phase, some morphological features appear that are characteristic for the adult population, although the frequency and intensity of their occurrence are still different from those observed in the adult structure. This phase extends to P28, during which time neuronal shapes become more differentiated, numerical density of NPY- and VIP-ir neurons increases transiently, and neuropil attains its adult-like structure. These changes correspond to the development and stabilization of claustral connections, processes of synaptogenesis, and synaptic plasticity [22,26,44]. After P28 the morphological changes are complete and the adult-like structure is finally achieved. The dynamics of these changes are characteristic for each immunoreactive neuronal population. 4.2. Postnatal developmental changes of neuropeptide-containing claustral interneurons are observed, but their dynamics are differentiated Studies based on the expression of selected transcription factors, comparative anatomy, and immunocytochemical methods have allowed the introduction of a general concept for the claustroamygdaloid complex, which is made up of the amygdalar basolateral complex (lateral, basolateral, and basomedial amygdalar nuclei) and the claustral complex (dorsal claustrum or claustrum proper and endopiriform nucleus) [28,33,43,52]. The expression of developmental regulatory genes confirms the common origin of the dorsal part of the Cl and the basolateral nucleus of amygdala from the lateral part of the pallium [33,43]. Developmental studies on the basolateral complex of amygdala indicate a character of observed in this structure processes similar to those occurring in the dorsal claustrum [39]. The interneurons of the basolateral complex of amygdala and the Cl are inhomogenous in their immunocytochemical characteristics. Numerous neuropeptides (CCK, NPY, SOM, and VIP), CaBP (such as calbindin D28k, calretinin, parvalbumin), and NOS were detected in the interneurons of the Cl [13,14,15,22,27] and amygdala [24,32]. Both pre- and postnatal development of CaBP- and NOS-ir interneurons of the Cl were studied [13,22,26]. Generally, the changes in neuronal morphology, intensity of staining and number of immunoreactive neurons were revealed in these studies. However, the intensity and dynamics of these processes were not identical in all developing interneurons; they were differentiated and related to the type of immunocytochemical marker. Based on our results, we have distinguished some characteristic differences in the development of neuropeptide-containing interneurons. This is most clearly demonstrated by the changes in numerical density values of NPY- and VIP-ir interneurons, in comparison to the values of SOM-ir neurons. In addition, differences were observed in the most frequently occurring neuronal types (multipolar in NPY- and VIP-ir versus bipolar in SOM-ir neurons). Furthermore, differentiated dynamics of neuropil development were reported, immunoreactive to each neuropeptide. These results may suggest the existence of separate regulatory developmental mechanisms for various interneurons, characterized by neuropeptide content. 4.3. Changes in the number of neuropeptide-containing neurons may indicate the importance of their physiological role during postnatal development The values of numerical density of NPY- and VIP-ir neurons increased transiently in the studied period. In the case of NPY-ir
569
neurons, the maximum values were observed after 2 weeks, and in the case of VIP-ir neurons, 1 week later. The changes in the number of NPY-ir neurons, similar to those observed in the Cl, occured in the occipital region of the rat cerebral cortex [1,9]. The decrease in the number of immunoreactive neurons is predominantly the result of neuronal death and, to a lesser degree, the result of a decrease in neuropeptide expression [1]. The role of NPY during early claustral development may be connected with the regulation of Ca++ ion inflow [2,51] and the regulation of neuronal fiber growth [11]. The influence of NPY on glutamate release may be indicative of a protective action against excitotoxicity. These important functions of NPY might be carried out by means of NPY receptors (type Y1 and Y2), the presence of which was described in the Cl [40]. Transient increases of both VIP mRNA and neuropeptide expression during development were observed in the frontal and parietal regions of the rat cerebral cortex [20]. The decrease in the numerical density of VIP-ir neurons might, according to some authors, be explained primarily by the decrease in neuropeptide content and, to a lesser degree, by neuronal death [10]. The role of VIP in claustral development is concerned with the regulation of neuronal proliferation and differentiation [16,36]. VIP may influence the length of the cell cycle by decreasing the G1 and S phases [21]. The potential function of VIP in relation to the claustral neurons may also be linked with its neuroprotective and neurotrophic function [6,19]. SOM-ir neurons in the Cl, although numerous, do not reveal any significant changes in their numerical density during the postnatal period. The intensity of SOM staining was differentiated throughout our developmental studies. This may indicate specific mechanisms of developmental regulation in SOM-ir neurons. It may also be speculated that differentiated expression of SOM may reveal some kind of relationship with the different place of origin of this subpopulation of neurons. During the early postnatal period, SOM may be connected with the growth control of the neuronal population [37]. It may also influence the process of neuronal migration [46,56]. 4.4. Developmental changes occur in neuropeptide-containing neuropil Maturation of the claustral connections remains under the control of neurotrophic factors produced by claustral neurons and axonal terminals of projecting neurons, originating in various cortical regions and subcortical structures [18,41,53]. Therefore, the length of the period of claustral neuropil development may be related, at least partially, to the development of connected structures, in particular, neocortical regions. According to our observations, neuropil development is completed after 4 weeks of postnatal life. This is concerned with final adjustment of claustral projections originating in various cortical regions and subcortical structures. The principal neurotransmitter of these projections is glutamate [42]. Simultaneously, other types of connections are developed between claustral interneurons, and between interneurons and projecting neurons of the Cl. GABA must be considered as the dominating neurotransmitter of these connections, in colocalization with various kinds of neuromodulators (neuropeptides) [17]. Interestingly, and importantly, the increase of vesicular glutamate transporter (VGLUT2) is thought to be proof of the development of stimulating connections [44] that takes place in the Cl at the moment of birth and early in the postnatal period. At the same time, intensive synaptogenesis is continued in the Cl, which is characterized by the transient increase of SNAP-25 (assessed by immunoblotting; our unpublished data). The maximal
570
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571
value of this transient increase occurs at P21. This pattern of SNAP-25 reactivity may reflect the development of intraclaustral connections and, therefore, the number of synapses. Subsequently, SNAP-25 content decreases, probably because of the changing influence of neurotrophic factors and cortical activity. The decrease in synapse formation during the later phase of postnatal life may correspond, to some extent, to the process of neuronal death, partly of apoptotic character, described previously in the rat in the early postnatal period [30]. Finally, the occurrence of NPY- and VIP-ir fibers situated around unlabeled neuronal cell bodies may confirm the inhibitory function of neuropeptide-containing neuropil elements in relation to projecting neurons and interneurons of the Cl. 4.5. The origin and path of migration of claustral neuropeptide-containing interneurons are still unknown Recently published studies point to a subcortical origin, possibly the medial ganglionic eminence (MGE), for a proportion of claustral GABA-ergic interneurons—at least those containing calbindin [29]. MGE and caudal ganglionic eminence (CGE) are the sources of CaBP- and somatostatin-containing neurons migrating to the cortex and numerous structures of the limbic system [8,38]. GABAergic interneurons originating in MGE and CGE were found in the structures of claustroamygdaloid complex. Taking into account the common colocalization of CaBP with neuropeptides in GABA-ergic interneurons of the Cl and amygdaloid body, the dual – cortical and subcortical – origin of the neuropeptide-containing interneurons may not be excluded. Our results indicate that the development of claustral interneurons and neuropil extend into the postnatal period. This process, according to our observations, terminates after the fourth postnatal week. However, we focused only on selected aspects of the complex problem of claustral development, which consist of numerous qualitative and quantitative processes. A specific cutoff point, terminating these processes cannot be established precisely. Despite this, the fundamental role of the early postnatal period for the proper development of the rat Cl must be taken into account in future studies on this structure. Conflict of interest No conflict of interests is claimed by any author of this publication. Acknowledgements This research was supported by funds from the Polish State Committee of Scientific Research grant number: W-705. The technical assistance of Mrs. S. Scislowska and Mr. T. Alexander is greatly appreciated. References [1] J. Antonopoulos, G.C. Papadopoulos, H. Michaloudi, M.E. Cavanagh, J.G. Parnavelas, Postnatal development of neuropeptide Y-containing neurons in the visual cortex of normal- and dark-reared rats, Neurosci. Lett. 145 (1992) 75. [2] A. Bacci, J.R. Huguenard, D.A. Prince, Differential modulation of synaptic transmission by neuropeptide Y in rat neocortical neurons, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 17125. [3] R.B. Banati, G.W. Goerres, C. Tjoa, J.P. Aggleton, P. Grasby, The functional anatomy of visual-tactile integration in man: a study using positron emission tomography, Neuropsychologia 38 (2000) 115–124. [4] S.A. Bayer, J. Altman, Development of the endopiriform nucleus and the claustrum in the rat brain, Neuroscience 45 (1991) 391. [5] H. Braak, E. Braak, Neuronal types in the claustrum of man, Anat. Embryol. 163 (1982) 447.
[6] D.E. Brenneman, G. Glazner, J.M. Hill, J. Hauser, A. Davidson, I. Gozes, VIP neurotrophism in the central nervous system: multiple effectors and identification of a femtomolar-acting neuroprotective peptide, Ann. N.Y. Acad. Sci. 865 (1998) 207. [7] G.A. Calvert, P.C. Hansen, S.D. Iversen, M.J. Brammer, Detection of audio-visual integration sites in humans by application of electrophysiological criteria to the BOLD effect, Neuroimage 14 (2001) 427. [8] R.S. Carney, T.B. Alfonso, D. Cohen, H. Dai, S. Nery, B. Stoica, J. Slotkin, B.S. Bregman, G. Fishell, J.G. Corbin, Cell migration along the lateral cortical stream to the developing basal telencephalic limbic system, J. Neurosci. 26 (2006) 11562–11574. [9] M.E. Cavanagh, J.G. Parnavelas, Development of vasoactive-intestinalpolypeptide-immunoreactive neurons in the rat occipital cortex: a combined immunohistochemical–autoradiographic study, J. Comp. Neurol. 284 (1989) 637. [10] M.E. Cavanagh, J.G. Parnavelas, Development of neuropeptide Y (NPY) immunoreactive neurons in the rat occipital cortex: a combined immunohistochemical–autoradiographic study, J. Comp. Neurol. 297 (1990) 553. [11] G. Chen, A.N. Van Den Pol, Multiple NPY receptors coexist in pre- and postsynaptic sites: inhibition of GABA release in isolated self-innervating SCN neurons, J. Neurosci. 16 (1996) 7711. [12] F.C. Crick, C. Koch, What is the function of the claustrum? Philos. Trans. R. Soc. Lond. B Biol. Sci. 360 (2005) 1271–1279. [13] J.C. Davila, M.A. Real, L. Olmos, I. Legaz, L. Medina, S. Guirado, Embryonic and postnatal development of GABA, calbindin, calretinin, and parvalbumin in the mouse claustral complex, J. Comp. Neurol. 481 (2005) 42. [14] L.R. Edelstein, F.J. Denaro, The claustrum: a historical review of its anatomy, physiology, cytochemistry and functional significance, Cell. Mol. Biol. 50 (2004) 675–702. [15] L.E. Eiden, E. Mezey, R.L. Eskay, M.C. Beinfeld, M. Palkovits, Neuropeptide content and connectivity of the rat claustrum, Brain Res. 523 (1990) 245. [16] G.W. Glazner, P. Gressens, S.J. Lee, G. Gibney, I. Gozes, Y. Gozes, D.E. Brenneman, J.M. Hill, Activity-dependent neurotrophic factor: a potent regulator of embryonic growth and development, Anat. Embryol. 200 (1999) 65. [17] S.M. Gomez-Urquijo, I. Gutierrez-Ibarluzea, J.L. Bueno-Lopez, C. Reblet, Percentage incidence of gamma-aminobutyric acid neurons in the claustrum of the rabbit and comparison with the cortex and putamen, Neurosci. Lett. 282 (2000) 177. [18] C.S. Goodman, Mechanisms and molecules that control growth cone guidance, Annu. Rev. Neurosci. 19 (1996) 341. [19] I. Gozes, I. Divinsky, I. Pilzer, M. Fridkin, D.E. Brenneman, A.D. Spier, From vasoactive intestinal peptide (VIP) through activity-dependent neuroprotective protein (ADNP) to NAP: a view of neuroprotection and cell division, J. Mol. Neurosci. 20 (2003) 315. [20] I. Gozes, Y. Shani, W.H. Rostene, Developmental expression of the VIP-gene in brain and intestine, Brain Res. 388 (1987) 137. [21] P. Gressens, B. Paindaveine, J.M. Hill, P. Evrard, D.E. Brenneman, Vasoactive intestinal peptide shortens both G1 and S phases of neural cell cycle in whole postimplantation cultured mouse embryos, Eur. J. Neurosci. 10 (1998) 1734. [22] S. Guirado, M.A. Real, J.L. Olmos, J.C. Davila, Distinct types of nitric oxideproducing neurons in the developing and adult mouse claustrum, J. Comp. Neurol. 465 (2003) 431–444. [23] N. Hadjikhani, P.E. Roland, Cross-modal transfer of information between the tactile and the visual representations in the human brain: a positron emission tomographic study, J. Neurosci. 18 (1998) 1072–1084. [24] S. Kemppainen, A. Pitkanen, Distribution of parvalbumin, calretinin, and calbindin-D(28k) immunoreactivity in the rat amygdaloid complex and colocalization with gamma-aminobutyric acid, J. Comp. Neurol. 426 (2000) 441–467. [25] P. Kowianski, J. Morys, J. Dziewiatkowski, Z. Karwacki, B. Bobek-Billewicz, O. Narkiewicz, Neurons of the claustrum projecting to the motor and somatosensory cortex in the rabbit—assessment with the method of retrograde transport of fluorescent tracers, Folia Morphol. 55 (1996) 350–352. [26] P. Kowianski, J.M. Morys, S. Wojcik, J. Dziewiatkowski, J. Morys, Postnatal development of NOS-ir neurons in the rat claustrum, Folia Morphol. 62 (2003) 11–17. [27] P. Kowianski, J.P. Timmermans, J. Morys, Differentiation in the immunocytochemical features of intrinsic and cortically projecting neurons in the rat claustrum—combined immunocytochemical and axonal transport study, Brain Res. 905 (2001) 63. [28] I. Legaz, M. Garcia-Lopez, L. Medina, Subpallial origin of part of the calbindinpositive neurons of the claustral complex and piriform cortex, Brain Res. Bull. 66 (2005) 470–474. [29] I. Legaz, G. Gonzalez, F. De Castro, J.L. Rubenstein, L. Puelles, Expression of Dbx1, Neurogenin 2, Semaphorin 5A, Cadherin 8, and Emx1 distinguish ventral and lateral pallial histogenetic divisions in the developing mouse claustroamygdaloid complex, J. Comp. Neurol. 474 (2004) 504–523. ˛ [30] B. Maciejewska, J. Mory´s, B. Berdel, J. Dziewiatkowski, O. Narkiewicz, The development of the rat claustrum—a study using morphometric and in situ DNA end labelling techniques, Eur. J. Anat. 1 (1997) 137. [31] L. Mamos, Morphology of claustral neurons in the rat, Folia Morphol. 43 (1984) 73.
P. Kowia´ nski et al. / Brain Research Bulletin 76 (2008) 565–571 [32] A.J. McDonald, J.C. Pearson, Coexistence of GABA and peptide immunoreactivity in non-pyramidal neurons of the basolateral amygdala, Neurosci. Lett. 100 (1989) 53–58. [33] L. Medina, I. Legaz, G. Gonzalez, F. De Castro, J.L. Rubenstein, L. Puelles, Expression of Dbx1, Neurogenin 2, Semaphorin 5A, Cadherin 8, and Emx1 distinguish ventral and lateral pallial histogenetic divisions in the developing mouse claustroamygdaloid complex, J. Comp. Neurol. 474 (2004) 504–523. [34] P. Mohapel, D.K. Hannesson, L.L. Armitage, G.W. Gillespie, M.E. Corcoran, Claustral lesions delay amygdaloid kindling in the rat, Epilepsia 41 (2000) 1095. [35] P. Mohapel, X. Zhang, G.W. Gillespie, J. Chlan-Fourney, D.K. Hannesson, S.M. Corley, X.M. Li, M.E. Corcoran, Kindling of claustrum and insular cortex: comparison to perirhinal cortex in the rat, Eur. J. Neurosci. 13 (2001) 1501. [36] T.W. Moody, J.M. Hill, R.T. Jensen, VIP as a trophic factor in the CNS and cancer cells, Peptides 24 (2003) 163. [37] J. Nacher, C. Ramirez, A. Molowny, C. Lopez-Garcia, Ontogeny of somatostatin immunoreactive neurons in the medial cerebral cortex and other cortical areas of the lizard Podarcis hispanica, J. Comp. Neurol. 374 (1996) 118. [38] S. Nery, G. Fishell, J.G. Corbin, The caudal ganglionic eminence is a source of distinct cortical and subcortical cell populations, Nat. Neurosci. 5 (2002) 1279–1287. ´ [39] J.L. Olmos, M.A. Real, L. Medina, S. Guirado, J.C. Davila, Distribution of nitric oxide-producing neurons in the developing and adult mouse amygdalar basolateral complex, Brain Res. Bull. 66 (2005) 465–469. [40] R.M.C. Parker, H. Herzog, Regional distribution of Y-receptor subtype mRNAs in rat brain, Eur. J. Neurosci. 11 (1999) 1431. [41] M. Pascual, E. Pozas, M.J. Barallobre, M. Tessier-Lavigne, E. Soriano, Coordinated functions of Netrin-1 and Class 3 secreted Semaphorins in the guidance of reciprocal septohippocampal connections, Mol. Cell. Neurosci. 26 (2004) 24. [42] F. Perez-Cerda, L. Martinez-Millan, C. Matute, Anatomical evidence for glutamate and/or aspartate as neurotransmitters in the geniculo-, claustro-, and cortico-cortical pathways to the cat striate cortex, J. Comp. Neurol. 373 (1996) 422. [43] L. Puelles, E. Kuwana, E. Puelles, A. Bulfone, K. Shimamura, J. Keleher, S. Smiga, J.L. Rubenstein, Pallial and subpallial derivatives in the embryonic chick and mouse telencephalon, traced by the expression of the genes Dlx-2, Emx-1, Nkx2.1, Pax-6, and Tbr-1, J. Comp. Neurol. 424 (2000) 409–438.
571
[44] M.A. Real, J.C. Davila, S. Guirado, Immunohistochemical localization of the vesicular glutamate transporter VGLUT2 in the developing and adult mouse claustrum, J. Chem. Neuroanat. 31 (2006) 169–177. [45] C. Reblet, A. Alejo, R.I. Blanco-Santiago, J. Mendizabal-Zubiaga, M. Fuentes, J.L. Bueno-Lopez, Neuroepithelial origin of the insular and endopiriform parts of the claustrum, Brain Res. Bull. 57 (2002) 495–497. [46] J.P. Schwartz, Z. Ji, J. Epelbaum, Somatostatin as a neurotrophic factor. Which receptor/second messenger transduction system is involved? Perspect. Dev. Neurobiol. 5 (1998) 427. [47] A.H. Sheerin, K. Nylen, X. Zhang, D.M. Saucier, M.E. Corcoran, Further evidence for a role of the anterior claustrum in epileptogenesis, Neuroscience 125 (2004) 57. [48] S. Shiosaka, K. Takatsuki, M. Sakanaka, S. Inagaki, H. Takagi, E. Senba, Y. Kawai, H. Iida, H. Minagawa, Y. Hara, T. Matsuzaki, M. Tohyama, Ontogeny of somatostatin-containing neuron system of the rat: immunohistochemical analysis. II. Forebrain and diencephalons, J. Comp. Neurol. 204 (1982) 211. [49] K.B. Sims, D.L. Hoffman, S.I. Said, E.Z. Zimmerman, Vasoactive intestinal polypeptide (VIP) in mouse and rat brain: an immunocytochemical study, Brain Res. 186 (1980) 165. [50] B. Spahn, H. Braak, Percentage of projection neurons and various types of interneurons in the human claustrum, Acta Anat. 122 (1985) 245. [51] L. Sun, L.H. Philipson, R.J. Miller, Regulation of K+ and Ca++ channels by a family of neuropeptide Y receptors, J. Pharmacol. Exp. Ther. 284 (1998) 625. [52] L.W. Swanson, G.D. Petrovich, What is the amygdala? Trends Neurosci. 21 (1998) 323–331. [53] R. Tuttle, B.L. Schlaggar, J.E. Braisted, D.D. O’Leary, Maturation-dependent upregulation of growth-promoting molecules in developing cortical plate controls thalamic and cortical neurite growth, J. Neurosci. 15 (1995) 3039. [54] A.P. Weible, C. Weiss, J.F. Disterhoft, Connections of the caudal anterior cingulate cortex in rabbit: neural circuitry participating in the acquisition of trace eyeblink conditioning, Neuroscience 145 (2007) 288–302. [55] B.L. Wilhite, T.J. Teyler, C. Hendricks, Functional relation of rodent claustral–entorhinal–hippocampal system, Brain Res. 365 (1986) 54. [56] E. Yacubova, H. Komuro, Stage-specific control of neuronal migration by somatostatin, Nature 415 (2002) 77.