brain research 1627 (2015) 201–215

Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research report

Calretinin and parvalbumin immunoreactive interneurons in the retrosplenial cortex of the rat brain: Qualitative and quantitative analyses Martin Salaja, Rastislav Drugaa,b,c, Jirˇı´ Cermana,d, Hana Kubova´c, Filip Barinkae,n a

Department of Anatomy, Charles University in Prague, 2nd Faculty of Medicine, Prague, Czech Republic Department of Anatomy, Charles University in Prague, 1st Faculty of Medicine, Prague, Czech Republic c Institute of Physiology, Academy of Sciences of the Czech Republic, Prague, Czech Republic d Department of Neurology, Charles University in Prague, 2nd Faculty of Medicine, Prague, Czech Republic e Department of Neurology, University of Regensburg, Bezirksklinikum Regensburg, Regensburg, Germany b

art i cle i nfo

ab st rac t

Article history:

The retrosplenial cortex (RSC) is a mesocortical region broadly involved with memory and

Accepted 26 September 2015

navigation. It shares many characteristics with the perirhinal cortex (PRC), both of which

Available online 9 October 2015

appear to be significantly involved in the spreading of epileptic activity. We hypothesized

Keywords: Retrosplenial cortex Calretinin Parvalbumin Interneurons Calcium-binding proteins Perirhinal cortex

that RSC possesses an interneuronal composition similar to that of PRC. To prove the hypothesis we studied the general pattern of calretinin (CR) and parvalbumin (PV) immunoreactivity in the RSC of the rat brain, its optical density as well as the morphological features and density of CR- and PV-immunoreactive (CRþ and PVþ) interneurons. We also analyzed the overall neuronal density on Nissl-stained sections in RSC. Finally, we compared our results with our earlier analysis of PRC (Barinka et al., 2012). Compared to PRC, RSC was observed to have a higher intensity of PV staining and lower intensity of CR staining of neuropil. Vertically-oriented bipolar neurons were the most common morphological type among CRþ neurons. The staining pattern did not allow for a similarly detailed analysis of somatodendritic morphology of PVþ neurons. RSC possessed lower absolute (i.e., neurons/mm3) and relative (i.e., percentage of the overall neuronal population) densities of CRþ neurons and similar absolute and lower relative densities of PVþ neurons relative to PRC. CR: PV neuronal ratio in RSC (1:2 in area 29 and 1:2.2 in area 30) differed from PRC (1:1.2 in area 35 and 1:1.7 in area 36). In conclusion, RSC, although similar in many aspects to PRC, differs strikingly in the interneuronal composition relative to PRC. & 2015 Elsevier B.V. All rights reserved.

Abbreviations: CaBP,

calcium binding protein(s); CB,

calbindin; CR,

calretinin; cROD,

corrected relative optical density;

PV, parvalbumin; PRC, perirhinal cortex; ROD, relative optical density; RSC, retrosplenial cortex; S.E.M., standard error of mean n Correspondence to: Department of Neurology, University of Regensburg, Bezirksklinikum Regensburg, Universitätstr. 84, 93053 Regensburg, Germany. E-mail address: fi[email protected] (F. Barinka). http://dx.doi.org/10.1016/j.brainres.2015.09.031 0006-8993/& 2015 Elsevier B.V. All rights reserved.

202

1.

brain research 1627 (2015) 201–215

Introduction

In cerebral cortex, complementing the pyramidal cells forming excitatory glutamatergic connections, a large grouping of inhibitory GABAergic interneurons has been described. Various strategies to sort GABAergic interneurons into distinct subgroups have been adopted (for review see Barinka and Druga, 2010; DeFelipe et al. 2013; Markram et al., 2004). Based on the expression of cytosolic calcium-binding proteins (CaBP) calretinin (CR), parvalbumin (PV) and calbindin (CB), it is possible-with some limitations-to divide cortical GABAergic interneurons into three largely non-overlapping populations (Del Rio and DeFelipe, 1996; Gabbott and Bacon, 1996; Kawaguchi and Kubota, 1997; Kubota et al., 1994; Toledo-Rodriguez et al., 2004, 2005; Zaitsev et al., 2005, 2009). In particular, the PV-expressing (PVþ) and CRexpressing (CRþ) interneuronal subgroups exhibit virtually no overlap. CRþ interneurons differ from PVþ interneurons in many aspects (Barinka and Druga, 2010)-with regard to their morphology, their connectivity, their electrophysiological properties and at least in rodents, their site of origin. There are remarkable interareal and interspecies differences in the cellular composition of individual regions which must be taken into account when exploring and discussing the precise function of any cortical area (Barinka et al., 2010; Hof et al., 1993). Recently, we had shown that the transitional perirhinal cortex (PRC) in the rat possesses a specific complement of interneurons with a higher proportion of CRþ interneurons and less extensive dendritic and/or axonal arborization of PVþ interneurons when compared to temporal neocortex (Barinka et al., 2012) and prefrontal cortex (Gabbott et al., 1997). It remains unclear whether this is an evolutionary adaptation for a special function of PRC (a massive inhibitory mechanism with a gating function has been described by Curtis and Pare, 2004) or a general attribute of transitional cortical areas in the rat brain. To further clarify this matter, we herein investigated the interneuronal apparatus of retrosplenial cortex (RSC) in the rat. Retrosplenial cortex shares many structural and functional characteristics with PRC – [1] they both are transitional cortices located between neocortex and archi-/periarchicortex, [2] both are broadly involved in memory functions (Murray and Richmond, 2001; Suzuki, 1996; Vann and Aggleton, 2002; Van Groen and Wyss, 2003; Van Strien et al., 2009; Vann et al., 2009; Vogt et al., 2006;), and [3] both are part of networks involved in the generation and spreading of epileptic activity (Ampuero et al., 2007; Benini et al., 2011; Brevard et al., 2006; Cardoso et al., 2008; De Guzman et al., 2004; Duzel et al., 2006; Englot et al., 2008; Fukumoto et al., 2002; Holmes et al., 1992; Imamura et al., 1998; Kelly and McIntyre, 1996; Raisinghani and Faingold, 2005; Scholl et al., 2013; Sudbury and Avoli, 2007). Thus, we hypothesized that RSC possesses an interneuronal composition similar to that of PRC. In our present work we studied the general pattern of CR and PV immunoreactivity, the optical density of neuropil as well as the morphological features and density of CR- and PV-immunoreactive interneurons in rat RSC. The RSC is located on the medial surface of the hemisphere, while the PRC in the concavity of the rhinal sulcus. Hence, factors like a different packing density of cells could hamper direct comparison of interneuronal density between particular areas. To avoid this problem, in the present study, equivalently to previous study on PRC, the overall

neuronal density was stereologically measured on Nisslstained sections. The respective proportions of CRþ and PVþ interneurons were then expressed as a percentage of the overall (Nissl) neuronal population in cortical layers II–IV and V–VI, and collectively in layers II–VI. In the Discussion, we compare our results with our previous data obtained from PRC as well as with data obtained in various neocortical regions.

2.

Results

2.1. General features of calretinin and parvalbumin immunoreactivity in RSC Determination of the boundaries of cortical areas under study is illustrated in Figs. 1 and 3. General features of CR and PV staining as described below are shown in Figs. 2 and 3.

2.1.1.

Calretinin

Unlike the perirhinal cortex, where the intensity of CR immunopositivity of neuropil was markedly higher than in the neighboring temporal neocortex (Barinka et al., 2012), we noted only a slight difference in the overall staining intensity between RSC and neighboring neocortex (Fig. 2B). However, differences in the pattern of immunoreactivity within individual cortical layers clearly differentiate retrosplenial areas 29 and 30 from the neighboring neocortex (Fig. 3A). In RSC the high layer I staining intensity, typical for all cortical regions, is less apparent than in the neighboring neocortex, and particularly so compared to PRC (Figs. 2B, 3A). Beneath layer I of RSC, a band of low CR immunoreactivity can be seen in layer II and the superficial portion of layer III (Fig. 3A). This band is very sharply delineated both superficially and profoundly in area 29, less sharply so in area 30. This difference is very prominent and enables a reliable distinction between these cortical areas. Further, in area 29, a thin and clearly demarcated band of higher CR immunoreactivity can be seen encompassing the thin layer IV as well as the most superficial portion of layer V (Fig. 3A). Again, the delineation of this highly CR-immunoreactive band is less sharp in area 30, where it also becomes markedly thicker relative to area 29. The deep portion of layer V and all of layer VI demonstrate a lower intensity of immunoreactivity (with the exception of the deepest parts of layer VI in area 29, as described below), which does not significantly change between areas 29 and 30 (Fig. 3A). In the rostral portion of RSC, area 29 ventrally borders the corpus callosum (Fig. 1A–E). The corpus callosum proper is virtually devoid of CR label, while at the transition point between area 29 and corpus callosum a very thin band of higher CRþ neuropil spans the entire thickness of the cortex (Fig. 4A). Beneath this band, a very prominent CRþ field in the most medial portion of the cingulum, continuing in the deep portion of layer VI of area 29 (Fig. 4B) can be seen. It appears to be composed of transversely dissected CRþ fibers running along rostrocaudal axis under the RSC. Ultimately, they disappear in its caudal-most portion. In the caudal portion of RSC, area 29 ventrally borders on dorsal subiculum and, finally, in its most caudal portion, on the postsubiculum (Fig. 1F–I). The border between area 29 and dorsal subiculum is clearly distinguishable: [a] the overall CRþ intensity in area 29 is higher than in subiculum, [b] a band of relatively

brain research 1627 (2015) 201–215

203

Fig. 1 – Schematic drawings of coronal sections at nine rostrocaudal levels with demarcated borders of areas 29 and 30. In C–G, also borders of perirhinal areas 35 and 36 in the extent evaluated in our previous work (Barinka et al., 2012), are demarcated. Approximate anteroposterior (AP) locations relative to bregma according to Paxinos and Watson (2007). M2: secondary motor cortex, MPtA: medial parietal association cortex, V2MM: secondary visual cortex mediomedial, DS: dorsal subiculum, Post: postsubiculum. Scale bar¼2 mm.

low CR immunopositivity in layers II and III of area 29 abruptly transitions to the uniformly lower staining intensity in the subiculum. While the overall intensity of CR staining in the postsubiculum is similar to area 29, the low-intensity CRþ band in layers II and III is not present in the postsubiculum (Fig. 5B). Along its rostrocaudal axis, area 30 borders laterally on secondary motor cortex, medial parietal association cortex and secondary visual cortex (Fig. 1). Although each of these areas differ in their respective CR immunoreactivity patterns, general features which help to define the border between them and RSC area 30 can be found. Firstly, area 30 possesses

a slightly lower CR staining intensity in layer I. As well, the above-described band of higher staining intensity in the superficial portion of layer V in area 30 transitions to a more uniformly higher staining intensity in layers V and VI in the neighboring neocortical areas (Fig. 3A). Rostral to RSC, cingular cortex differs in its CR immunoreactivity due to its far more homogenous labeling across cortical layers without that clearly demarcated band of lower staining intensity in layers II/III seen in the RSC. The overall pattern and intensity of CR staining in RSC does not change significantly along the rostrocaudal axis, but

204

brain research 1627 (2015) 201–215

there is one exception – a slightly higher overall staining intensity in area 29a, which is found only in the caudal sections (Figs. 5B, 6A,C,E). We did not observe any difference in staining patterns between areas 29b and 29c, previously noting that an unequivocal distinction between these two areas was not possible from the CR-immunostained sections.

2.1.2.

Parvalbumin

The PV immunoreactivity pattern in RSC (Figs. 2C, 3B) was not seen to differ from neighboring neocortical areas as prominently as was the case with CRþ material. In area 29, layer I shows low intensity of PV staining. A band of high intensity in layer II sharply demarcated against layer I, and with a gradual decrease of staining intensity in layer III, can be found. Layers III and IV possess low PV immunopositivity, gradually transitioning to a higher intensity in layer V. Finally, there is a lower intensity of staining in layer VI. At the border of area 29 with area 30, an abrupt and prominent increase of the thickness of a band of high staining intensity in layers II and III – along with a slight decrease in the thickness of a high intensity band in layer V – was found (Fig. 3C). Due to this stark change in the immunoreactivity pattern, the border between areas 29 and 30 can be clearly demarcated on PV-stained sections. We were unable to find any difference in the level of PV immunoreactivity between areas 29 b and c. In area 29 a, the individual bands of higher/lower intensity of PV staining partially merge to a more uniform staining pattern (Fig. 5C). Both dorsal subiculum and postsubiculum possess higher overall intensity of staining in all cortical layers when compared to the adjacent area 29 (Fig. 5C). Further, the clearly demarcated bands of higher PV staining intensity as seen in area 29 were no longer visible. The borders between area 30 and secondary motor cortex, medial parietal association cortex and secondary visual cortex were not reliably demonstrable in the PV immunostained material (Fig. 3B). We could not find any significant difference in the overall intensity of labeling in PV-immunostained sections along the rostrocaudal axis (Fig. 6B,D,F).

2.2.

Fig. 2 – Representative low magnification photomicrographs illustrating the pattern of (A) Nissl (cresyl violet), (B) calretinin and (C) parvalbumin staining in cerebral hemisphere of the rat brain. The sections depicted are consecutive, neighboring sections, approximately at 4.6 mm AP from bregma according to Paxinos and Watson (2007). Borders of cortical areas 29, 30 (RSC) and 35, 36 (PRC) are demarcated. Scale bar¼2 mm.

Neuronal morphology

In general, neuronal morphology was more readily distinguishable in CRþ than in PVþ neurons (Fig. 7). Cell bodies, proximal dendrites and distal dendritic segments could be identified in the majority of the CRþ neurons. The dendritic arbor was more intensely stained and longer dendritic segments could be observed in layers V–VI than in layers II–IV. In the majority of PV immunostained neurons, only the cell bodies and occasionally primary dendrites could be identified. Their more distal dendritic segments were not distinguishable due to the interference from highly-immunopositive neuropil. The overall count of immunoreactive neurons in RSC layer I was quite low and therefore did not allow for a systematic morphological analysis. Further, a bias in comparison of neuronal density due to the difference in relative thickness of layer I (in relation to the overall cortical thickness) between RSC and PRC could be introduced by inclusion of layer I in quantitative analysis. Given this, only layers II–VI were examined.

brain research 1627 (2015) 201–215

205

Fig. 3 – Representative photomicrographs illustrating the pattern of (A) calretinin and (B) parvalbumin immunoreactivity and (C) Nissl (cresyl violet) staining in areas 29 and 30 with demarcated areal (arrows) and laminar borders (lines, showing borders between layer I and layer II, and between layers II–IV and layers V–VI). Schematic drawing of areal and laminar boundaries in (D) is based on the section immunostained for PV in (B). Location at approximately  4.05 mm AP from bregma. At this AP level, area 30 borders laterally on the medial parietal association cortex. Scale bar¼ 300 lm.

2.2.1.

Calretinin

In total, 505 CRþ neurons in 16 histological sections were described and divided into three groups according to their morphological type. The relative representation of each type in the different areas/layers can be seen in Fig. 8. 1. Vertically-oriented bipolar or bitufted neurons The most commonly seen shape of CRþ neuron was the bipolar type (Fig. 7A,B). These cells formed the largest group in deep cortical layers (V–VI) in both areas but not in the superficial layers (II–IV) where multipolar type neurons were more common. Their perikarya were usually oval or fusiform,

and two bipolar dendrites were often seen emanating from opposite poles of the cell body. Less commonly, two tufts of dendrites (bitufted) were observed, extending into neighboring cortical layers. 2. Multipolar neurons Multipolar neurons (Fig. 7C) were the second largest group of CRþ neurons in general and the most common type seen in the superficial cortical layers (II–IV) in both areas. Usually three or four dendrites were noted to radially project from a round cell body. Less frequently, multipolar neurons with an oval or inverted pyramidal morphology were found. 3. Horizontally-oriented bipolar or bitufted neurons

206

brain research 1627 (2015) 201–215

Fig. 4 – An overall (A) and detailed (B) view of intensive calretinin immunoreactivity in the medial portion of the cingular bundle, composed in first line of transversely cut fibers. Additionally, many fibers extend to the adjacent area 29; these fibers are cut longitudinally or oblique on coronal sections. Scale bar¼500 lm in (A) and 100 lm in (B).

Horizontally-oriented CRþ neurons (Fig. 7B) numbered in the minority in both areas. When seen, they usually had an oval cell body and either two dendritic tufts or two primary

Fig. 5 – A detailed view of (A) Nissl, (B) calretinin and (C) parvalbumin staining in area 29a and in the neighboring postsubiculum (Post) with depicted border between these two areas. Alv¼ alveus. Short black lines indicate borders between cortical layers I and II–IV, and between layers II–IV and V–VI. Location at approximately  6.3 mm AP from bregma. Scale bar¼300 lm.

dendrites arising from opposite poles.

2.2.2.

Parvalbumin

With respect to PVþ RSC cells, in the majority of observed neurons only their perikarya and short segments of primary dendrites were clearly discernible (Fig. 7D). Cellular processes were well stained, but only rarely were they clearly distinguishable from the highly-PVþ neuropil owing to background interference. Most of these neurons had either a round,

2.3.

Quantitative analysis

2.3.1.

Densities of CRþ and PVþ neurons

Neuronal densities of CRþ and PVþ neurons in areas 29 and 30 were stereologically estimated and expressed as the number of neurons per 1 mm3 of cerebral cortex 7 standard error of mean (S.E.M.), separately and respectively for cortical layers II–IV and V–VI and finally collectively for layers II–VI. The results are summarized in Table 1.

inverted pyramidal or irregularly-shaped body, resembling the morphology of multipolar neurons. Less commonly, neurons with a vertically-oriented oval-shaped body were found, and those in the deep cortical layers.

2.3.1.1. Calretinin. The distribution pattern of CRþ neurons was similar in areas 29 and 30, with high neuronal density in the superficial cortical layers and markedly lower density in

brain research 1627 (2015) 201–215

207

Fig. 6 – Representative photomicrographs illustrating the pattern of calretinin (A, C, E) and parvalbumin (B, D, F) immunoreactivity in the studied areas at different rostrocaudal levels. Location approximately at 2.3 mm (A, B),  4.3 mm (C, D) and 7.05 mm (E, F) AP from bregma according to Paxinos and Watson (2007). Scale bar¼ 500 lm. M2: Secondary motor cortex, V2MM: Secondary visual cortex mediomedial. The differences of immunopositivity for calretinin and for parvalbumin between the rostral and more caudal levels were subtle and did not reach statistical significance in densitometric analysis of neuropil immunoreactivity.

the deep layers. In both superficial and deep layers, higher density of CRþ neurons could be found in area 29 when compared with area 30. The interareal difference was statistically significant.

2.3.1.2. Parvalbumin. The density of PVþ neurons in layers V–VI was significantly higher in area 29 when compared with area 30. The differences in remaining density values (layers II–IV, II–VI) did not reach statistical significance.

208

brain research 1627 (2015) 201–215

Fig. 7 – Examples of the various morphological types of immunoreactive neurons in studied area. (A) Vertically-oriented bipolar CRþ neuron. (B) Vertically- and horizontally-oriented bipolar CRþ neurons. (C) Multipolar CRþ neuron. (D) Example of PVþ neuron in RSC, without clearly discernible somatodendritic morphology. Scale bar¼20 lm.

2.3.2. Overall neuronal densities and percentages that CRþ and PVþ neurons constitute relative to the overall neuronal population The overall neuronal density (estimated on Nissl-stained sections) in superficial layers was significantly higher in area 29 in comparison to area 30. Due to this marked difference in layers II–IV, the overall density in layers II–VI was also

significantly higher in area 29 than in area 30, although very similar neuronal densities in layers V–VI of areas 29 and 30 were found. The trend towards a decrease in CRþ neuronal density along the ventro-dorsal axis of RSC was not retained when the values were expressed as percentages of the total neuronal population. The CRþ interneurons formed 2.2% in area 29

209

brain research 1627 (2015) 201–215

and 2.1% of the overall neuronal population in layers II–VI in area 30 (Table 1). No significant difference in the percentages of the overall neuronal population in areas 29 and 30 was found with respect to PVþ neurons, which formed 4.4% of overall neuronal population in layers II–VI in area 29 and 4.5% in area 30 (Table 1).

2.3.3. Densitometric analysis of calretinin and parvalbumin immunoreactivity A densitometric analysis was performed to quantify the observed differences in the CR and PV immunoreactivity of neuropil in areas 29 and 30 of RSC. The corrected relative optical density (cROD) was measured in both areas (see Experimental procedure, 4.6, for further explanation of terms). The results are summarized in Table 2. On CRþ sections, a cROD of 2.1 was observed in both areas. A markedly higher cROD of 4.5 in area 29 and 4.4 in area 30 was found in PVþ sections (see Fig. 3 for visualization). To examine possible differences in the immunoreactivity for CR and PV along the rostrocaudal axis, we further compared the cROD between five different rostrocaudal levels in both examined areas. No significant differences in the cROD were found in any of the examined immunostained sections (Fig. 6, numerical data not shown).

3.

Fig. 8 – Pie charts showing the proportion of individual morphological types of CRþ interneurons in studied areas and layers.

Discussion

In our earlier work we described a pattern of immunoreactivity for CR and PV, and especially the absolute and relative interneuronal density within the perirhinal cortex of the rat (Barinka et al., 2012). We had shown that the PRC possesses a defining interneuronal signature evidenced by a high density of CRþ neurons not found in the neighboring temporal neocortex, nor in the prefrontal cortex of the rat (Gabbott et al., 1997). In the present work, using the same material and identical methods, we performed an analogous assessment of CR and PV immunoreactivity and interneuronal density in rat RSC. Perirhinal and retrosplenial cortices have much in common. Both are transitional cortical areas bridging neocortex and archicortex. As well, both are in a pivotal location within a complex system of extensive connections between neocortical regions, thalamic nuclei and the hippocampal formation and, as such, are a part of the limbic system (Kealy and Commins, 2011; Suzuki, 1996; Van Groen and Wyss, 2003). Each plays an important-albeit different-role in memory. Finally, both RSC and PRC are involved in epileptogenesis and the spread of seizure activity (see Introduction for citations). The aim of this study was to provide evidence as

Table 1 – Absolute neuronal densities and percentages of CaBP immunoreactive neurons. (A) Absolute neuronal densities per mm3 of retrosplenial cortex 7 S.E.M. Area 29

Nissl CR PV

Area 30

II.–IV.

V.–VI.

II.–VI.

II.–IV.

V.–VI.

II.–VI.

145,96978304X 53637165X 44617223

68,81572160 1174730X 39327310X

93,47773456X 2050754X 40597282

97,48473468 33307221 47157247

67,14672265 953762 30647206

79,90471098 16567106 35787196

(B) Percentages of CaBP-immunoreactive neurons in the overall neuronal population 7 S.E.M. Area 29

CR PV

Area 30

II.–IV.

V.–VI.

II.–VI.

II.–IV.

V.–VI.

II.–VI.

3.7570.29 3.1470.31X

1.6770.07 5.7370.46

2.1870.08 4.470.4

3.4870.24 4.8670.29

1.470.12 4.5870.33

2.0770.13 4.4870.27

210

brain research 1627 (2015) 201–215

to whether RSC (areas 29 and 30), similar in many aspects to the PRC as outlined above, is comprised of a comparable interneuronal cyto-/chemoarchitecture supportive of such functional attributes. While the distribution of CR immunoreactivity in the neuropil of individual cortical layers of RSC was found to be similar to that seen in PRC, the overall intensity of neuropil staining in RSC was similar to the neighboring neocortical areas. This is in stark contrast to PRC, which possesses a significantly higher intensity of CRþ neuropil when compared to neocortex (Figs. 2B, 9A,B). Furthermore, in comparison to PRC with its exceptionally low PVþ intensity in area 35, the intensity of PV staining in RSC was markedly higher and did not remarkably differ from neighboring neocortical fields (Figs. 2C, 9C,D). Similarly, when compared to PRC and temporal neocortex, a high proportion of vertically-oriented bipolar and multipolar CRþ neurons were also found in RSC (Fig. 8). Most of the PVþ neurons possessed round, inverted pyramidal or irregularlyshaped cell bodies resembling multipolar neurons. Unfortunately, a high level of background interference due to the intensity of PV-staining of neuropil did not allow for a Table 2 – Corrected relative optical density 7 S.E.M. of neuropil immunopositive for CR and PV.

CR PV

Area 29

Area 30

2.170.2 4.570.7

2.170.2 4.470.6

representative semi-quantitative morphological analysis of PVþ neurons in the respective areas. Through the use of stereological methods we further evaluated the CRþ and PVþ neuronal densities in areas 29 and 30 and the overall neuronal density of Nissl-stained neurons in these areas. A slightly higher density of CRþ neurons-which was also statistically significant-could be found in layers II–VI of area 29 (2050 neurons/mm3) when compared to area 30 (1650 neurons/mm3). A similar situation (albeit without statistical significance) was found on PVþ material, with 4050 neurons/mm3 in layers II–VI of area 29, and 3600 neurons/mm3 in area 30. The overall neuronal density in layers II–VI was also higher in area 29 (93,500 neurons/mm3) than in the area 30 (79,900 neurons/mm3). When expressed as percentages of the overall neuronal population, CRþ neurons formed 2.2% of layers II–VI of area 29 and 2.1% of the overall neuronal population in area 30. PVþ neurons formed 4.4% in layers II–VI of area 29 and 4.5% of the overall neuronal population in area 30. Thus, the numerical ratio between CRþ and PVþ neuronal populations was calculated to be 1:2 in area 29 and 1:2.2 in area 30. In our earlier work on PRC (Barinka et al., 2012) which utilized the same stereological method and the same tissue sections, we found CRþ neurons to form 6.4% of total neuronal population in area 35 and 5.5% in area 36. The PVþ neurons formed 7.6% in area 35 and 9.4% of the overall neuronal population in area 36 (see Fig. 10 for an RSC-PRC comparison). The numerical ratio between CRþ and PVþ neurons was 1:1.2 in area 35 and 1:1.7 in area 36. Therefore, two mesocortical regions (PRC and RSC) with similar locations, connections and-to some extent even function-differ strikingly in the quality as well as quantity of their interneuronal

Fig. 9 – Comparison of patterns of CR immunopositivity in PRC (A) and RSC (B) as well as PV immunopositivity in PRC (C) and RSC (D). Two consecutive/neighboring coronal sections from one animal were stained for CR and PV respectively and are depicted. Therefore, variations in tissue preparation or staining intensity between PRC and RSC can be excluded. Location approximately at  4.3 mm AP from bregma according to Paxinos and Watson (2007). Scale bar¼500 lm.

brain research 1627 (2015) 201–215

3.1.

Fig. 10 – Relative interneuronal (CRþ and PVþ interneurons) counts expressed as percentages of total neuronal population (as estimated stereologically on Nissl stained material)7S.E.M. in RSC (areas 29 and 30), PRC (areas 35 and 36) as well as in neocortical temporal area Te3V. Table 3 – Relative approximate ratios of CRþ to PVþ neurons in various cortical areas of the adult rat brain as described by various authors. Area

CR: PV

Citation

35 mPFC 36 29 29c 30 Te3V 17 Cg2

1: 1: 1: 1: 1: 1: 1: 1: 1:

Barinka et al. (2012) Gabbott et al. (1997) Barinka et al. (2012) Present work Lema Tome et al. (2007) Present work Barinka et al. (2012) Gonchar and Burkhalter (1997) Lema Tome et al. (2007)

a

1.2 1.4 1.7 2 2a 2.2 2.6 2.9 3a

Only semi-quantitative data available.

composition. Not only is there a higher relative density of both CRþ and PVþ in PRC compared with RSC, there is also a higher CRþ: PVþ neuronal ratio in PRC in comparison with RSC. In Table 3 we present an overview of the relative ratios of CRþ and PVþ neurons in PRC, RSC as well as in other cortical regions of the rat brain previously described. To our knowledge, this is the first description of the stereologically-estimated neuronal and interneuronal density of areas 29 and 30 in the rat. Armando Cardoso with colleagues (Cardoso et al., 2008) stereologically estimated the overall neuronal counts in area 29c in normal and epileptic rats. Most interestingly, they found a prominent decrease in the overall neuronal density in epileptic rats, very similar to that which was previously found in entorhinal cortex. However, differences in methodology, tissue preparation and the extent of examined area did not allow for a direct comparison of their results (in normal rats) with ours. Furthermore, Carla M. Lema Tomé with colleagues (Lema Tomé et al., 2007) described the developmental pattern of PVþ and CRþ neurons in rat RSC. A semi-quantitative analysis without the use of stereology was presented, which again makes a direct comparison with our results problematic. However, the ratio of CRþ to PVþ neuronal counts of approximately 1:2 in adult rats is very similar to our results.

211

Conclusions and functional implications

At present, computational approaches are increasingly being applied to elucidate the mechanisms of function of various regions of the central nervous system (Perin et al., 2013). Having a precise knowledge of the cellular composition of any cerebral area forms one of the basic input parameters and tenets for computational modeling of neuronal networks. In our present study we describe the pattern of CaBP immunoreactivity, the density of CRþ and PVþ neurons as well as the overall neuronal density in rat retrosplenial cortex. Our results, coupled with our previous work and with that of other authors, show that there are significant differences in the interneuronal composition and distribution of heretofore seemingly similar cortical areas. These differences must be taken into account when modeling and evaluating intracortical neuronal networks (Beul and Hilgetag, 2015). Further research is needed for precise evaluation of these differences and of their impact on cortical information processing. Finally, it is hoped that our results may form a basis for further research on RSC and its interneuronal attributes in pathological conditions, especially in the context of epilepsy.

4.

Experimental procedure

4.1.

Animals

The brains of six male, healthy, previously untreated, 3month-old Wistar rats weighing between 350 and 400 g were initially prepared for use in our previous study on interneurons in PRC (Barinka et al., 2012). These brains were also used for the present study. For the qualitative analysis of CaBP immunostaining patterns, material from two archival cases were also consulted (male, 4-month-old Wistar rats). Animals were housed under standard conditions (12 h light/12 h dark cycle, 2271 1C, humidity 50–60%, free access to water and food). Experiments were approved by the Animal Care and Use Committee of the Institute of Physiology of the Academy of Sciences of the Czech Republic. Animal care and experimental procedures were conducted in accordance with the guidelines of the European Community Council directives 86/ 609/EEC and NIH Guidelines (Assurance No. #A5820-01).

4.2.

Animals and tissue preparations

Rats were irreversibly anesthetized and euthanized with urethane (2 g/kg i.p.) and perfused with 0.01 M sodium phosphate buffer (pH 7.4) followed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). The brains were removed, postfixed in buffered 4% paraformaldehyde for 3 h and cryoprotected in a graded series of sucrose solutions in 0.01 M sodium phosphate buffer (pH 7.4) at þ4 1C. The brains were frozen in dry ice and sectioned in the coronal plane (50 mm thick sections, 1-in-5 series) with a Leica CM 1900 cryostat. Sections were stored in a cryoprotective solution (30% ethylene glycol, 25% glycerol in 0.05 M sodium phosphate buffer) at  20 1C until processed. An adjacent series of sections was used for Nissl staining and immunohistochemistry.

212

4.3.

brain research 1627 (2015) 201–215

Immunohistochemical staining

Consecutive serial coronal sections (1-in-5 series) were processed for [1] Nissl (cresyl violet) staining, [2] CR, [3] PV, and [4] CB immunohistochemical staining (n.b., the fifth set of sections was not used in this study). The distance between any two sections stained with the same method was also 250 mm. Due to a very variable and in part very low staining intensity of CB immunoreactive neurons with unsatisfactory discrimination of dendritic tree, CB immunostained sections, unsuitable for quantitative analysis, were not used in the present study. The sections for CR and PV immunohistochemistry were processed as follows: The sections were sequentially incubated in 0.15% hydrogen peroxide in PBS (0.01 M; pH 7.4) for 10 min, rinsed with PBS five times, permeabilized with 0.3% Triton-X100 for 5 min and then incubated with a blocking solution containing 2% normal horse serum (Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature. The sections were then incubated with the primary antibody (mouse anti-PV monoclonal antibody, Cat. No. MAB 1572, PubMed ID 17029253, Millipore; dilution 1:10000; or mouse anti-CR monoclonal antibody, Cat. No. MAB 1568, PubMed ID 18022947, Millipore, dilution 1:8000) in PBS containing 1.5% normal horse serum (S-2000, Vector) and 0.1% Triton-X100 for 48 h at 4 1C and then rinsed five times in PBS. Thereafter, the sections were incubated for 1 h at room temperature with the secondary antibody, biotinylated horse anti-mouse antibody (BA-2001, Vector, dilution 1:50 in PBS) containing 1.5% normal horse serum. After this step, the sections were rinsed with PBS five times and covered with the ABC reagent (Vectastain Kit, Vector) for 1 h at room temperature. After rinsing, sections were incubated for approximately 8 min with a mixture of 0.02% diaminobenzidine and 0.05% hydrogen peroxide in PBS. Lastly, sections were mounted onto gelatin-coated slides, dehydrated and cover-slipped.

4.4.

Areal and laminar boundaries

The retrosplenial cortex nomenclature of Vogt et al. (2004) was used for the present study. It is considered equivalent to Brodmann's nomenclature of human cortical areas (Brodmann, 1909), with granular area 29 and dysgranular area 30 comparatively distinguishable in rat RSC. Most of the cytoarchitectonic landmarks described in that work were reliably identified on Nissl-stained sections. The CR- and PV-labeled sections provided additional landmarks useful for delineating RSC borders, as described in the Results section. Area 29 can further be parcellated into 29a, 29b and 29c. However, the border between areas 29b and 29c could not reliably be distinguished on sections used in the present study. As well, area 29a occupies a very small portion of area 29 and proved not suitable for separate quantitative analysis. Therefore, in the context of quantitative analysis, area 29 was viewed as a singular region without further subdivision. In recent modification of the abovementioned nomenclature of RSC, Vogt and Paxinos (2014), based on subtle differences in cytoarchitecture and connectivity, suggested a further subdivision of areas 29c and 30 into anterior and posterior divisions. We could not consistently distinguish the cytoarchitectonic attributes necessary for such subparcellation in our Nissl-stained sections.

In the rostrocaudal axis, our quantitative analysis of RSC was restricted to the coronal planes from approximately  1.8 mm to approximately  6.55 mm relative to bregma (Paxinos and Watson, 2007), nearly spanning its entire rostrocaudal extent (Fig. 1). In the remaining small caudal portion of RSC, the plane of section was not perpendicular to the surface of cerebral cortex. That lack of symmetry could have biased the estimation of neuronal density within individual layers and was a reason why we did not include it in our quantitative analysis.

4.5.

Qualitative analysis

The qualitative analysis was performed on immunohistochemically- and Nissl-stained sections using image analyzing software (Cell*F, Olympus) and a digital camera (Olympus DP 72) attached to the microscope (Olympus BX 51). Photos and linear drawings for this publication were prepared using Corel Draw X3 software. Qualitative analyses were performed on both cerebral hemispheres. In the qualitative description of immunohistochemical staining for CR and PV, the term “intensity” of neuropil staining is used throughout the manuscript. The formulation “optical density” of neuropil is reserved for quantitative measurement of neuropil staining as described below.

4.6.

Quantitative analysis

All methods of quantitative analysis used in the present study were identical with those applied in our study of PRC interneurons (Barinka et al., 2012). In conjunction with our use of the same material, this facilitated a direct comparison of results obtained for the PRC and RSC.

4.6.1.

Densitometric analysis

Densitometric analysis of calretinin and parvalbumin immunopositivity in cortical areas 29 and 30 was performed at five coronal levels (2.05 mm,  3.05 mm,  4.05 mm,  5.05 and 6.05 mm AP relative to Bregma; Paxinos and Watson, 2007) evenly-spaced along the rostrocaudal axis. The measurements were performed on six animals. Therefore, 30 sections in all were analyzed (5 sections from each of the six animals). Images were captured using image analyzing software (QuickPHOTO MICRO 2.3, Promicra) and a digital camera (Olympus DP 72) attached to the microscope (Olympus BX 51; 10x objective). To avoid differences in the light intensity of the captured images, all images were captured at the same light intensity setting in the microscope and then converted to grayscale. Densitometric analysis was performed using Densita software (version 3.42; MicroBrightField, Inc.), as described elsewhere (Barinka et al., 2012). Briefly, the relative optical density (ROD) was calculated for each studied area and section. A lower ROD value means that more light is coming through the tissue, and a higher value means more light is blocked in the tissue and therefore not detected. We calculated ROD for each studied area and section. Cortical layer I was not included in the study. To eliminate the possible biasing influence of slightly varying intensities of immunopositivity between different sections and animals, the ROD was corrected for by dividing the values measured in individual areas under study by the values acquired from the corpus callosum. The corpus callosum was selected as the reference structure because it is only weakly immunolabeled

brain research 1627 (2015) 201–215

by antibodies to CR and PV and has well-defined borders. On each section studied, the central part of the corpus callosum (i.e., that between both hemispheres)-with the omission of the ventral and dorsal surfaces-was used as a reference. The optical density value acquired here was then used for the correction of values acquired in RSC on the same section. This corrected staining index (corrected relative optical density, cROD) allowed for the comparison between different sections and subjects.

4.6.2.

Stereological analysis

Principles and methods of unbiased stereology as previously described (West et al., 1991) were applied to achieve unbiased estimates of total neuronal density and of interneuronal densities in the studied areas. An Olympus BX 51 microscope equipped with a motorized stage and a MicroBrightField (MBF Bioscience) Stereo Investigator stereology system (version 11.01.2) were used. Neuronal numbers were measured separately in layers II–IV and V–VI, and then collectively as a sum in layers II–VI. Cortical layer I was not included in our analysis due to the very low counts of CaBP-labeled interneurons. It is of particular note that PVþ cells were virtually absent in layer I. Overall neuronal density was measured in the Nissl-stained sections. In all, 20–22 coronal sections spaced 250 mm apart were contained in the above-described rostrocaudal extent of analyzed tissue in each of the six animals included in the quantitative analysis. Cell counting was performed on every other section of these 20–22 consecutive Nissl-stained sections, with the starting-point being randomly set as either section 1 or 2. Therefore, 10–11 sections were analyzed for each animal. A sampling grid and optical dissector were adjusted to obtain a relatively constant number of cells sampled and a coefficient of error – CE of o0.07 (Gundersen and Jensen, 1987). For areas 29 and 30 we used a sampling grid of 100  100 mm2 for layers II–IV and of 150  150 mm2 for layers V–VI. For each x–y step, cell counts were derived from a known fraction of 25  25 mm2 in layers II–IV and 30  30 mm2 in layers V–VI (an “optical dissector”). Neuronal nuclei in thick sections were counted using a 100x oil immersion objective (NA 1.4), except for those nuclei that were already in focus at the upper surface of the section. Glial and endothelial cells were excluded from the counts. Neurons were distinguished from glial and endothelial cells based on their typical cytomorphological features, as described in (Barger et al., 2012; Morgan et al., 2014). The total number of neurons in each studied area/layer was calculated using the formula: N ¼ ΣQ  1=ssf  1=asf  1=tsf; where section sampling fraction (ssf) was 1/10, area sampling fraction (asf – area of counting frame divided by area of sampling grid) was 0.0625 in layers II–IV and 0.04 in layers V–VI of both areas 29 and 30 and finally, tissue sampling fraction (tsf – the height of the mounted section thickness divided by the dissector height) was 1. The total neuronal density (i.e., the density of all neurons of all morphological types) was calculated by dividing the total number of neurons by the total volume sampled. The total volume sampled was calculated by adding the crosssectional surface of the area of interest in each analyzed section and multiplying this number by the section interval and by the section thickness.

213

A slightly modified strategy was used for counting CRþ and PVþ interneurons. The same number of sections as in the case of overall neuronal density was analyzed-ssf was 1/ 10. The immunopositivity of cytoplasm in labeled cells was high and occasionally higher than that of the nucleus. Therefore, whole cell bodies rather than simply neuronal nuclei were counted in the overall mounted section height (tsf¼1), again omitting the neurons that were already in focus on the surface. Furthermore, due to the relatively low number of immunoreactive interneurons in the studied areas/layers, all such neurons in all areas of interest in each of the analyzed sections were counted using a 40x immersion oil objective (NA 1.0). Therefore, the asf was 1. The interneuronal density was calculated in the same manner as the overall neuronal density. To exclude the possibility of bias in the interareal comparison of interneuronal density which could be introduced by differences in neuronal packing density, we further expressed the relative interneuronal counts in individual areas/layers as percentages of the overall neuronal population. Quantitative analysis (i.e., densitometric and stereological measurements) was performed on the right cerebral hemisphere.

4.6.3.

Statistical analysis

Analysis of variance (ANOVA) followed by Tukey–Kramer multiple comparison test was used for the comparison of corrected relative optical densities, neuronal densities and interneuronal percentages. Probability (p) values o0.05 were considered as being statistically significant.

Acknowledgment We would like to thank to Mrs. B. Čejková for excellent technical assistance. We further wish to kindly thank to Dr. L. Edelstein for reading the manuscript and for English corrections. This work was supported by Grant Agency of Charles University, Grant no. 35407.

r e f e r e nc e s

Ampuero, E., Dagnino-Subiabre, A., Sandoval, R., Zepeda-Carreno, R., Sandoval, S., Viedma, A., Aboitiz, F., Orrego, F., Wyneken, U., 2007. Status epilepticus induces region-specific changes in dendritic spines, dendritic length and TrkB protein content of rat brain cortex. Brain Res. 1150, 225–238. Barger, N., Stefanacci, L., Schumann, C.M., Sherwood, C.C., Annese, J., Allman, J.M., Buckwalter, J.A., Hof, P.R., Semendeferi, K., 2012. Neuronal populations in the basolateral nuclei of the amygdala are differentially increased in humans compared with apes: a stereological study. J. Comp. Neurol. 520, 3035–3054, http://dx.doi.org/10.1002/cne.23118. Barinka, F., Druga, R., 2010. Calretinin expression in the mammalian neocortex: a review. Physiol. Res. 59, 665–677. Barinka, F., Druga, R., Marusic, P., Krsek, P., Zamecnik, J., 2010. Calretinin immunoreactivity in focal cortical dysplasias and in non-malformed epileptic cortex. Epilepsy Res. 88, 76–86, http: //dx.doi.org/10.1016/j.eplepsyres.2009.09.021. Barinka, F., Salaj, M., Rybar, J., Krajcovicova, E., Kubova, H., Druga, R., 2012. Calretinin, parvalbumin and calbindin

214

brain research 1627 (2015) 201–215

immunoreactive interneurons in perirhinal cortex and temporal area Te3V of the rat brain: qualitative and quantitative analyses. Brain Res. 1436, 68–80, http://dx.doi.org/10.1016/j. brainres.2011.12.014. Benini, R., Longo, D., Biagini, G., Avoli, M., 2011. Perirhinal cortex hyperexcitability in pilocarpine-treated epileptic rats. Hippocampus 21, 702–713, http://dx.doi.org/10.1002/hipo.20785. Beul, S.F., Hilgetag, C.C., 2015. Towards a “canonical” agranular cortical microcircuit. Front. Neuroanat. 8, 165, http://dx.doi. org/10.3389/fnana.2014.00165. Brevard, M.E., Kulkarni, P., King, J.A., Ferris, C.F., 2006. Imaging the neural substrates involved in the genesis of pentylenetetrazol-induced seizures. Epilepsia 47, 745–754. Brodmann K., 1909. Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues Barth, Leipzig Cardoso, A., Madeira, M.D., Paula-Barbosa, M.M., Lukoyanov, N.V., 2008. Retrosplenial granular b cortex in normal and epileptic rats: a stereological study. Brain Res. 1218, 206–214, http://dx. doi.org/10.1016/j.brainres.2008.04.074. De Curtis, M., Pare, D., 2004. The rhinal cortices: a wall of inhibition between the neocortex and the hippocampus. Prog. Neurobiol. 74, 101–110 doi: S0301-0082(04)00145-5 [pii] 10.1016/ j.pneurobio.2004.08.005. De Guzman, P., D’antuono, M., Avoli, M., 2004. Initiation of electrographic seizures by neuronal networks in entorhinal and perirhinal cortices in vitro. Neuroscience 123, 875–886 doi: S0306452203008728 [pii]. Defelipe, J., Lopez-Cruz, P.L., Benavides-Piccione, R., Bielza, C., Larranaga, P., Anderson, S., Burkhalter, A., Cauli, B., Fairen, A., Feldmeyer, D., Fishell, G., Fitzpatrick, D., Freund, T.F., Gonzalez-Burgos, G., Hestrin, S., Hill, S., Hof, P.R., Huang, J., Jones, E.G., Kawaguchi, Y., Kisvarday, Z., Kubota, Y., Lewis, D. A., Marin, O., Markram, H., Mcbain, C.J., Meyer, H.S., Monyer, H., Nelson, S.B., Rockland, K., Rossier, J., Rubenstein, J.L., Rudy, B., Scanziani, M., Shepherd, G.M., Sherwood, C.C., Staiger, J.F., Tamas, G., Thomson, A., Wang, Y., Yuste, R., Ascoli, G.A., 2013. New insights into the classification and nomenclature of cortical GABAergic interneurons. Nat. Rev. Neurosci. 14, 202–216, http://dx.doi.org/10.1038/nrn3444. Del Rio, M.R., Defelipe, J., 1996. Colocalization of calbindin D-28k, calretinin, and GABA immunoreactivities in neurons of the human temporal cortex. J. Comp. Neurol. 369, 472–482. Duzel, E., Schiltz, K., Solbach, T., Peschel, T., Baldeweg, T., Kaufmann, J., Szentkuti, A., Heinze, H.J., 2006. Hippocampal atrophy in temporal lobe epilepsy is correlated with limbic systems atrophy. J. Neurol. 253, 294–300. Englot, D.J., Mishra, A.M., Mansuripur, P.K., Herman, P., Hyder, F., Blumenfeld, H., 2008. Remote effects of focal hippocampal seizures on the rat neocortex. J. Neurosci. 28, 9066–9081, http: //dx.doi.org/10.1523/JNEUROSCI.2014-08.2008. Fukumoto, S., Tanaka, S., Tojo, H., Akaike, K., Takigawa, M., 2002. Perirhinal cortical lesion suppresses the secondary generalization in kainic acid-induced limbic seizure. Psychiatry Clin. Neurosci. 56, 561–567 doi: 1055 [pii]. Gabbott, P.L., Bacon, S.J., 1996. Local circuit neurons in the medial prefrontal cortex (areas 24a,b,c, 25 and 32) in the monkey: I. Cell morphology and morphometrics. J. Comp. Neurol. 364, 567–608. Gabbott, P.L., Dickie, B.G., Vaid, R.R., Headlam, A.J., Bacon, S.J., 1997. Local-circuit neurones in the medial prefrontal cortex (areas 25, 32 and 24b) in the rat: morphology and quantitative distribution. J. Comp. Neurol. 377, 465–499. Gonchar, Y., Burkhalter, A., 1997. Three distinct families of GABAergic neurons in rat visual cortex. Cereb. Cortex 7, 347–358.

Gundersen, H.J., Jensen, E.B., 1987. The efficiency of systematic sampling in stereology and its prediction. J. Microsc. 147, 229–263. Hof, P.R., Nimchinsky, E.A., Celio, M.R., Bouras, C., Morrison, J.H., 1993. Calretinin-immunoreactive neocortical interneurons are unaffected in Alzheimer’s disease. Neurosci. Lett. 152, 145–148. Holmes, K.H., Bilkey, D.K., Laverty, R., 1992. The infusion of an NMDA antagonist into perirhinal cortex suppresses amygdalakindled seizures. Brain Res. 587, 285–290 doi: 0006-8993(92) 91009-4 [pii]. Imamura, S., Tanaka, S., Tojo, H., Fukumoto, S., Uetsuhara, K., Kuratsu, J., Takigawa, M., 1998. Kainic acid-induced perirhinal cortical seizures in rats. Brain Res. 800, 323–327 S0006899398005137 [pii]. Kawaguchi, Y., Kubota, Y., 1997. GABAergic cell subtypes and their synaptic connections in rat frontal cortex. Cereb. Cortex 7, 476–486. Kealy, J., Commins, S., 2011. The rat perirhinal cortex: a review of anatomy, physiology, plasticity, and function. Prog. Neurobiol. 93, 522–548, http://dx.doi.org/10.1016/j.pneurobio.2011.03.002. Kelly, M.E., Mcintyre, D.C., 1996. Perirhinal cortex involvement in limbic kindled seizures. Epilepsy Res. 26, 233–243 doi: S09201211(96)00056-3 [pii]. Kubota, Y., Hattori, R., Yui, Y., 1994. Three distinct subpopulations of GABAergic neurons in rat frontal agranular cortex. Brain Res. 649, 159–173. Lema Tome, C.M., Miller, R., Bauer, C., Nottingham, C., Smith, C., Blackstone, K., Brown, L., Bryan, R., Leigh, A., Brady, M., Busch, J., Turner, C.P., 2007. Decline in age-dependent, MK801induced injury coincides with developmental switch in parvalbumin expression: cingulate and retrosplenial cortex. Dev. Psychobiol. 49, 606–618. Markram, H., Toledo-Rodriguez, M., Wang, Y., Gupta, A., Silberberg, G., Wu, C., 2004. Interneurons of the neocortical inhibitory system. Nat. Rev. Neurosci. 5, 793–807. Morgan, J.T., Barger, N., Amaral, D.G., Schumann, C.M., 2014. Stereological study of amygdala glial populations in adolescents and adults with autism spectrum disorder. PLoS One 9, e110356, http://dx.doi.org/10.1371/journal.pone.0110356. Murray, E.A., Richmond, B.J., 2001. Role of perirhinal cortex in object perception, memory, and associations. Curr. Opin. Neurobiol. 11, 188–193 doi: S0959-4388(00)00195-1 [pii]. Paxinos, G., Watson, C., 2007. The Rat Brain in Stereotaxic Coordinates, 6th ed. Oxford: Elsevier Academic Press, Amsterdam. Perin, R., Telefont, M., Markram, H., 2013. Computing the size and number of neuronal clusters in local circuits. Front. Neuroanat. 7, 1, http://dx.doi.org/10.3389/fnana.2013.00001. Raisinghani, M., Faingold, C.L., 2005. Evidence for the perirhinal cortex as a requisite component in the seizure network following seizure repetition in an inherited form of generalized clonic seizures. Brain Res. 1048, 193–201 S0006-8993(05) 00676-1 [pii] 10.1016/j.brainres.2005.04.070. Scholl, E.A., Dudek, F.E., Ekstrand, J.J., 2013. Neuronal degeneration is observed in multiple regions outside the hippocampus after lithium pilocarpine-induced status epilepticus in the immature rat. Neuroscience 252, 45–59, http://dx.doi.org/ 10.1016/j.neuroscience.2013.07.045. Sudbury, J.R., Avoli, M., 2007. Epileptiform synchronization in the rat insular and perirhinal cortices in vitro. Eur. J. Neurosci. 26, 3571–3582 doi: EJN5962 [pii] 10.1111/j.1460-9568.2007.05962.x. Suzuki, W.A., 1996. The anatomy, physiology and functions of the perirhinal cortex. Curr. Opin. Neurobiol. 6, 179–186 doi: S09594388(96)80071-7 [pii]. Toledo-Rodriguez, M., Blumenfeld, B., Wu, C., Luo, J., Attali, B., Goodman, P., Markram, H., 2004. Correlation maps allow neuronal electrical properties to be predicted from single-cell

brain research 1627 (2015) 201–215

gene expression profiles in rat neocortex. Cereb. Cortex 14, 1310–1327. Toledo-Rodriguez, M., Goodman, P., Illic, M., Wu, C., Markram, H., 2005. Neuropeptide and calcium-binding protein gene expression profiles predict neuronal anatomical type in the juvenile rat. J. Physiol. 567, 401–413. Van Groen, T., Wyss, J.M., 2003. Connections of the retrosplenial granular b cortex in the rat. J. Comp. Neurol. 463, 249–263. Van Strien, N.M., Cappaert, N.L., Witter, M.P., 2009. The anatomy of memory: an interactive overview of the parahippocampalhippocampal network. Nat. Rev. Neurosci. 10, 272–282 nrn2614 [pii] 10.1038/nrn2614. Vann, S.D., Aggleton, J.P., 2002. Extensive cytotoxic lesions of the rat retrosplenial cortex reveal consistent deficits on tasks that tax allocentric spatial memory. Behav. Neurosci. 116, 85–94. Vann, S.D., Aggleton, J.P., Maguire, E.A., 2009. What does the retrosplenial cortex do?. Nat. Rev. Neurosci. 10, 792–802, http: //dx.doi.org/10.1016/j.nicl.2014.08.021. Vogt, B.A., Paxinos, G., 2014. Cytoarchitecture of mouse and rat cingulate cortex with human homologies. Brain Struct. Funct. 219, 185–192, http://dx.doi.org/10.1007/s00429-012-0493-3.

215

Vogt, B.A., Vogt, L., Farber, N.B., 2004. Cigulate cortex and disease models. In: Paxinos, G. (Ed.), The Rat Nervous System 3th Edition Elsevier Academic Press, San Diego, pp. 705–727 London. Vogt, B.A., Vogt, L., Laureys, S., 2006. Cytology and functionally correlated circuits of human posterior cingulate areas. Neuroimage 29, 452–466. West, M.J., Slomianka, L., Gundersen, H.J., 1991. Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator. Anat. Rec. 231, 482–497. Zaitsev, A.V., Gonzalez-Burgos, G., Povysheva, N.V., Kroner, S., Lewis, D.A., Krimer, L.S., 2005. Localization of calcium-binding proteins in physiologically and morphologically characterized interneurons of monkey dorsolateral prefrontal cortex. Cereb. Cortex 15, 1178–1186. Zaitsev, A.V., Povysheva, N.V., Gonzalez-Burgos, G., Rotaru, D., Fish, K.N., Krimer, L.S., Lewis, D.A., 2009. Interneuron diversity in layers 2–3 of monkey prefrontal cortex. Cereb. Cortex 19, 1597–1615, http://dx.doi.org/10.1093/cercor/bhn198.