Modeling trophoblast differentiation using equine chorionic girdle vesicles.

The chorionic girdle of the equine conceptus is comprised of specialized trophoblast cells which, at day 36-38 of equine pregnancy, gain an invasive phenotype and invade the endometrium to form endometrial cups. Studies of equine endometrial cups remain difficult to perform because of the invasive techniques required to obtain cup tissue and because sampling requires termination of the pregnancy. In this study we developed a system to model trophoblast differentiation and trophoblast-immune interactions in vitro and in vivo. We utilized a method of culturing chorionic girdle pieces in serum-free medium to promote spontaneous formation of vesicle structures enriched for terminally differentiated binucleate cells that secreted equine chorionic gonadotrophin (eCG). Immunohistochemical staining and scanning electron microscopy showed that the cells of the vesicles closely resembled the outer layers of chorionic girdle immediately prior to invasion. Chorionic girdle vesicles were harvested after 72h in culture and ectopically transplanted via injection into the vulvar mucosa of recipient mares. At 7, 14, 21 and 28days after transplantation, biopsies of the injection sites were obtained. Immunohistochemical labeling of cryostat sections of the biopsies with a panel of monoclonal antibodies to horse trophoblast molecules demonstrated survival, differentiation, and presence of trophoblast cells for at least 21days. Serial sections of the biopsies labeled with antibodies to the equine lymphocyte surface markers CD4 and CD8, together with lymphocyte microcytotoxicity assays, revealed that the recipients mounted both cellular and humoral antibody immune responses to the transplanted trophoblast cells. This new method for culturing equine chorionic girdle trophoblast cells, and for transplanting trophoblast vesicles to ectopic sites, should allow identification of key aspects of trophoblast differentiation and the interactions that occur between invasive trophoblast and the maternal immune system.

Control of expression of major histocompatibility complex genes in horse trophoblast.

In most mammals, the fetus limits its presentation of paternal antigens to the mother by suppressing the cell-surface expression of proteins of the major histocompatibility complex (MHC) on trophoblast. In the horse, however, functional, polymorphic MHC class I antigens are expressed at high levels on the invasive trophoblast cells of the chorionic girdle between Days 32 and 36 of pregnancy, although not on the adjacent noninvasive trophoblast of the chorion and allantochorion membranes. In this study, the control of MHC class I gene expression was investigated in invasive and noninvasive horse trophoblast, and the MHC class I loci expressed by invasive trophoblast were identified. Northern blot hybridization of Day 33-34 conceptus tissue revealed both transcriptional and posttranscriptional regulation of cell-surface MHC class I expression in horse trophoblast. The invasive MHC class I-positive trophoblast showed levels of steady-state mRNA nearly as high as those in lymphoid tissues from adult horses, whereas noninvasive MHC class I-negative trophoblast also contained transcripts for MHC class I, but at lower levels similar to those present in adult horse nonlymphoid tissue. We also cloned and sequenced polymerase chain reaction products from the transmembrane and cytoplasmic regions of MHC class I transcripts in chorionic girdle and lymphocytes, and determined that horse invasive trophoblast appears to transcribe the same MHC class I loci transcribed in lymphocytes, including both polymorphic and nonpolymorphic loci. These data from the horse demonstrate that functional alloantigen presentation by trophoblast can be a normal part of early pregnancy.

Molecular and functional characterization of genes encoding horse MHC class I antigens.

Sequence and functional analyses were undertaken on two cDNAs and a genomic clone encoding horse major histocompatibility complex (MHC) class I molecules. All of the clones were isolated from a single horse that is homozygous for all known horse MHC class I and class II antigens. The two cDNAs (clones 8-9 and 1-29) were isolated from a lymphocyte library and encode polymorphic MHC antigens from two loci. The genomic cosmid clone, isolated from a sperm library, contains the 8-9 gene. All three genes were expressed in mouse L-cells and were recognized by alloantisera and, for the cDNAs, by alloreactive cytotoxic T lymphocytes. A total of 3815 bp of the genomic clone were sequenced, extending from 429 bp upstream (5') of the leader peptide through the 3' untranslated region. Promoter region motifs and an intron-exon structure characteristic of MHC class I genes of other species were found. A subclone containing 407 bp of the promoter region was inserted into a chloramphenicol acetyl transferase reporter plasmid, tested in transient transfection assays, and found to have promoter activity in heterologous cells. This genomic clone will enable detailed studies of MHC class I gene regulation in horse trophoblasts, and in horse retroviral infections.

Sex ratio bias in postpartum-conceived Norway rat litters is produced by embryonic loss in midpregnancy.

In rats, dams that conceive in their postpartum oestrus and then lose their firstborn litter bias the sex ratio of the litter toward females in utero. The present study identifies the source of litter sex ratio bias in these postpartum pregnant non-lactating dams. The female bias arises first through the postconception loss of embryos, and second, the loss occurs in midpregnancy between the attachment of the blastocyst to the uterine wall on day 5 and full metrial gland development on day 14. Some pregnancies were restricted to one uterine horn to see if this loss (and thus the opportunity for litter sex ratio biasing) was influenced by local factors operating within the uterine horn. Embryonic loss was more closely associated with the number of embryos implanting in a single horn than with the number implanting in the litter, demonstrating that local crowding within a horn is sufficient for the preferential loss of male embryos. This loss did not cause an obvious decrease in the size of the live-born litter because only those horns with a surfeit of embryos lost them. This process was the same in the right and left horns; both carried and lost the same numbers of embryos. A dam that conceives in her postpartum oestrus and then loses her suckling litter forgoes the implantation delay and uterine healing caused by lactation. Male embryos are less successful at implanting in a uterus only recently vacated by a previous litter.

Calretinin neurons in human medial prefrontal cortex (areas 24a,b,c, 32', and 25).

The calcium-binding protein calretinin (CR) is present in a subpopulation of local-circuit neurons in the mammalian cerebral cortex containing gamma-aminobutyric acid. This light microscopic investigation provides a detailed qualitative and quantitative morphological analysis of CR-immunoreactive (CR+) neurons in the medial prefrontal cortex (mPFC; areas 24a,b,c, 32', and 25) of the normal adult human. The morphology of CR+ neurons and their areal and laminar distributions were consistent across human mPFC. The principal organisational features of CR+ labelling were the marked laminar distribution of immunoreactive somata and the predominantly vertical orientation of labelled axon-like and dendritic processes. Several types of CR- neurons were present in layer 1, including horizontally aligned Cajal-Retzius cells. In layers 2-6, CR+ neurons displayed a variety of morphologies: bipolar cells (49% of CR+ population), vertically bitufted cells (35%), and horizontally bitufted cells (3.5%). These neuron types were mainly located in layer 2/upper layer 3, and their dendritic processes were commonly aspiny and sometimes highly beaded. Aspiny (8%) and sparsely spiny multipolar (5%) CR+ neurons were also found. The mean somatic profile diameter of CR+ cells was 11.6 +/- 0.3 microm (mean +/- S.D). CA+ puncta formed pericellular baskets around unlabelled circular somatic profiles in layers 2/3 and around unlabelled pyramidal-shaped somata in layers 5/6. The somatic sizes of these unlabelled cell populations were significantly different. Immunolabelled puncta were also found in close contact with CR+ somata. Cortical depth distribution histograms and laminar thickness measurements defined the proportions of the overall CR- cell population in each layer: 7% in layer 1, 78% in layers 2/3, 14% in layers 5/6, and 1% in the white matter. In the tangential plane, CR+ neurons were distributed uniformly at all levels of the cortex. By using stereological counting procedures on immunoreacted Nissl-stained sections, CR+ neurons were estimated to constitute a mean 8.0% (7.2-8.7%) of the total neuron population in each cortical area. These data are compared with similar information obtained for the mPFC in monkey and rat (Gabbott and Bacon [1996b] J. Comp. Neurol. 364:657-608; Gabbott et al., [1997] J. Comp. Neurol. 377:465-499). This study provides important morphological insights into a neurochemically distinct subclass of local-circuit inhibitory neurons in the human mPFC.

Calcineurin immunoreactivity in prelimbic cortex (area 32) of the rat.

Calcineurin (CN) is a Ca2+/calmodulin-dependent phosphatase present in brain tissue. In this immunocytochemical study of rat prelimbic cortex (area 32), a gradient of CN immunolabelling was found in the somata and processes of pyramidal cells and in interneurones. Some apical dendritic spines were weakly immunoreactive. The results suggest that CN is differentially present in prelimbic cortical neurones--this may reflect possible differences in intracellular Ca2+ signalling mechanisms.

Vasoactive intestinal polypeptide containing neurones in monkey medial prefrontal cortex (mPFC): colocalisation with calretinin.

Neurones immunoreactive for vasoactive intestinal polypeptide (VIP) were studied in monkey medial prefrontal cortex. The majority (78.0%) of VIP+ neurones were bipolar cells located mainly in layers 2/3. Calretinin (CR) immunoreactivity was colocalised in 80.5% of VIP+ neurones. Furthermore, VIP+ puncta formed pericellular baskets around GABA immunonegative somata in layers 2/3. These results indicate that VIP/CR are colocalised in some bipolar cells and superficial pyramidal somata are likely targets of VIP+ neurones.

Local-circuit neurones in the medial prefrontal cortex (areas 25, 32 and 24b) in the rat: morphology and quantitative distribution.

This paper is a light microscopical study describing the detailed morphology and quantitative distribution of local circuit neurones in areas 25, 32, and 24b of the medial prefrontal cortex (mPFC) in the rat. Cortical interneurones were identified immunocytochemically by their expression of calretinin (CR), parvalbumin (PV), and calbindin D-28k (CB) immunoreactivity. Neurones immunoreactive for gamma-aminobutyric acid (GABA) were also investigated, as were interneurones containing reduced nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase activity. Several distinct classes of CR+, PV+, and CB+ neurones were identified; the most frequent were: bipolar/bitufted CR+ cells in upper layer 3; multipolar PV+ neurones in layers 3 and 5; and bitufted/multipolar CB+ neurones in lower layer 3. CB+ neurones resembling Martinotti and neurogliaform cells were also present in layers 5/6. The morphologies and depth distributions of each cell type were consistent across the three areas of mPFC studied. Seven classes of diaphorase-reactive mPFC neurone are described; these cells were composed about 0.8% of the total neurone population and had a peak distribution located in mid- to lower layer 5 in each area. In areas 32 and 25, three defined bands of diffuse NADPH diaphorase staining were located in layer 2 and in upper and deep layer 5. Diaphorase reactivity was very infrequently colocalised with either CR, PV, or CB immunoreactivities. The numerical densities of neurones (N(V), number of cells per mm3) in each layer were calculated stereologically. The mean total neuronal N(V) estimate for areas 25, 32, and 24b was 51,603 +/- 3,324 (mean +/- S.D.; n = 8). Significant interareal differences were detected. From cortical thickness data and neuronal N(V) estimates, the absolute number of neurones under 1 mm2 of cortical surface (N(C)) have been derived. The mean N(C) value for areas 25, 32, and 24b was 57,328 +/- 7,505 neurones. In immunolabelled Nissl-stained sections, CR+ neurones constituted an overall 4.0%, PV+ cells 5.6%, and CB+ 3.4% of the total neurone populations in mPFC. GABA+ cells represented a mean of 16.2% (14.8-17.2%) of neurones in areas 25, 32 and 24b. The absolute numbers of CR+, PV+, CB+, and GABA+ neurones within individual layers in a column of cortex under 1 mm2 of cortical surface (N(L)) have also been derived, with significant interareal differences in N(L) values being detected. The data provide the structural basis for a qualitative and quantitative definition of local cortical circuits in the rat mPFC.

The organisation of dendritic bundles in the prelimbic cortex (area 32) of the rat.

Using an antibody against microtubule associated protein 2 (MAP-2; a specific marker for neuronal dendrites), this paper reports the structural organisation of pyramidal cell apical dendrites in the rat prelimbic (PL) cortex. In the coronal plane, MAP-2-immunoreactive apical dendrites of pyramidal neurons in layers 5, 3 and 2 were found bundled together as they ascended radially through the cortex. These bundles of dendrites dispersed in upper layer 2 to form apical dendritic tufts in layer 1. In tangential cross-section, the immunolabelled bundles were organised into a latticework of discrete clusters of differentially sized profiles. At the boundary between layers 3 and 5, clusters were composed of 26 +/- 8 dendritic profiles (group mean value +/- S.D., five animals), whereas clusters in layer 2 contained 55 +/- 15 profiles. The number of clusters per unit surface area was not significantly different throughout layers 5, 3 and 2 (760 +/- 75 per mm2) with the average centre-to-centre intercluster distance in these layers being 44.2 +/- 4.9 microns. The data indicate that apical dendritic bundles are a feature of the radial organisation of PL cortex. These structural subunits may subserve specific integrative functions in the PL area of the rat medial prefrontal cortex.

Amygdala input to medial prefrontal cortex (mPFC) in the rat: a light and electron microscope study.

This paper describes the termination pattern and synaptic connectivity of the pathway from the basolateral nucleus of the amygdala (BLA) to the medial prefrontal cortex (mPFC; areas 25, 32, and 24b) of the rat. Discrete injections of the anterograde tracer Phaseolus vulgaris-leucoagglutinin (PHA-L) were made in the BLA and detailed light microscopical observations made of the distribution of PHA-L labelled fibres and boutons in the mPFC. Labelled fibres were distributed in two tiers: predominantly within deep layer 1/layer 2 and also in layers 5/6. Fibre plexi in layers 2 and 5 were highly varicose. Electron microscopical examination of 120 labelled boutons in area 32 (60 in layer 2 and 60 in layer 5) indicated that 116 (97%) established asymmetrical synaptic contacts with dendritic spines and 4 (3%) were in synaptic contact with small dendritic shafts. No significant differences in target structures were found between layers 2 and 5. The results indicate that BLA input to mPFC in the rat predominantly innervates spine bearing dendrites in layers 2 and 5. This suggests that the neuronal operations of these processes are influenced by direct feedforward excitation from the BLA.

Localisation of NADPH diaphorase activity and NOS immunoreactivity in astroglia in normal adult rat brain.

This study demonstrates the co-localisation of NADPH diaphorase activity and GFAP immunoreactivity in non-neuronal cells in weakly fixed brain sections from normal adult rats. The presence of GFAP immunoreactivity in these cells indicates that they are astroglia. In addition, cells possessing the morphological characteristics of astroglia were weakly immunoreactive for the endothelial isoform of nitric oxide synthase (eNOS)--these cells also co-localised NADPH diaphorase activity. Furthermore, cells immunoreactive for eNOS displayed GFAP immunoreactive processes. This cytochemical evidence strongly suggests that resting astroglia are potential sources of nitric oxide--a powerful modulator of cell activity.

Local circuit neurons in the medial prefrontal cortex (areas 24a,b,c, 25 and 32) in the monkey: II. Quantitative areal and laminar distributions.

The companion paper (Gabbott and Bacon [1996] J. Comp. Neurol.) describes the morphology of calretinin (CR)-, parvalbumin (PV)-, calbindin (CB)-, and GABA-immunoreactive neurons, and NADPH diaphorase-reactive cells, in the medial prefrontal cortex (mPFC; areas 24a, 24b, 24c, 25 and 32) of the adult monkey. Since these local circuit neurons play crucial functional roles, the aim of this study was to provide supportive quantitative data defining their areal and laminar distribution in mPFC. The numerical densities of neurons (Nv, number of cells per mm3) in each area and layer were calculated stereologically. The mean total neuronal NV estimates across mPFC was 55,727 +/- 3,319 per mm3 (mean +/- S.D.; n = 3); values ranged from 50,489 +/- 8,374 per mm3 (area 24a) to 59,938 +/- 7,214 per mm3 (area 24c). Interareal differences were not significant. Cortical depth measurements and neuronal NV estimates for each area allowed the absolute number of neurons in a column of cortex under 1 mm2 of surface to be calculated; values varied between 86,457 +/- 15,063 (area 24a) and 128,464 +/- 24,050 (area 24c). Using immunolabelled Nissl-stained sections of mPFC, CR+ neurons constituted 11.2%, PV+ neurons 5.9%, and CB+ neurons 5.0% of the total neuron population. GABA+ neurons represented an overall 24.9% (23.5-27.3%) of neurons in the mPFC. Differences between areas were not significant. The cortical depth distribution histograms of CR+, PV+, CB+, and GABA+ cell populations in each area were derived and the percentage of a given cell population in each layer subsequently calculated. Peaks in the cortical depth distributions of CR+ and CB+ neurons occurred in layer 2 and upper layer 3, respectively; the peak distribution of PV+ neurons occurred between lower layer 3 and upper layer 5. The depth distribution of GABA+ cells reflected the combined distributions of CR+, PV+ and CR+ neurons. In all areas, the majority (74.4-84.0%) of the GABA cell population was located in layers 2/3. The depth distributions for each cell type were similar between areas. Diaphorase-reactive neurons accounted for 0.25% (0.2-0.32%) of all cortical neurons in mPFC and were distributed in two horizontal strata, in midlayer 3 and in mid/upper layer 6. A large population of diaphorase-reactive cells was present in the white matter. The absolute numbers of CR+, PV+, CB+ and GABA+ neurons within individual layers in a column of cortex under 1 mm2 and 50 x 50 microns of cortical surface have been derived. The data presented provide the basis for a quantitative definition of cortical circuits in monkey mPFC.

Local circuit neurons in the medial prefrontal cortex (areas 24a,b,c, 25 and 32) in the monkey: I. Cell morphology and morphometrics.

This paper provides a comprehensive morphological description of local circuit neurons in the medial prefrontal cortex (mPFC: areas 24a, 24b, 24c, 25 and 32) of the monkey. Cortical interneurons were identified immunocytochemically by the expression of the calcium binding proteins calretinin (CR), parvalbumin (PV) and calbindin D-28k (CB). Interneurons were also identified using GABA immunocytochemistry. The areal and laminar distributions of CR, PV, and CB cells were consistent across mPFC; their morphological characteristics identified them as local circuit neurons. Throughout layers 2-6: CR immunoreactivity labelled double bouquet and bipolar neurons, PV was localised in large and small basket neurons and in chandelier (axoaxonic) cells, while CB immunoreactivity was present in double bouquet, Martinotti, and neurogliaform neurons. In addition, some cells in layer 1 (including Cajal-Retzius neurons) were CR immunoreactive. Calbindin immunoreactivity also labelled a population of large nonpyramidal neurons deep in the cortex. Other types of CR, PV and CB cells were also immunolabelled. A small population of layer 3 pyramidal cells was weakly CB immunoreactive. Peak cell densities occurred in layer 2/upper layer 3 for CR+ neurons and in upper to midlayer 3 for CB+ cells. PV+ neuron density peaked in midcortex. These observations support and extend a similar study of monkey prefrontal cortex (Condé et al. [1994] J. Comp. Neurol. 341:95-116). The morphologies and combined cortical depth distributions of CR+, PV+, and CB+ neurons were similar to GABA-immunolabelled cells. Local circuit neurons in mPFC displaying NADPH diaphorase activity composed less than 0.25% of the total neuron population, and were distributed in two horizontal strata, in mid- to lower layer 3 and in lower layer 5/upper layer 6. CR, PV and CB immunoreactivity was colocalised in NADPH diaphorase-reactive neurons. The interrelationships between CR+, PV+ and CB+ neurons were investigated using dual immunocytochemistry. CR+ puncta were found to be closely associated with the cell bodies and proximal processes of PV+ neurons, whereas CR+ puncta were located more distally over processes from CB+ cells. Additionally, PV+ puncta were found closely apposed to PV+ somata and processes and CR+ puncta abutted against CR+ cell bodies. The companion paper (Gabbott and Bacon [1996] J. Comp. Neurol.) presents quantitative data regarding the areal and laminar distributions of the identified cell classes in mPFC. Such data provide a realistic structural framework with which to investigate neuronal operations in monkey mPFC.

Co-localisation of NADPH diaphorase activity and GABA immunoreactivity in local circuit neurones in the medial prefrontal cortex (mPFC) of the rat.

This study provides evidence that neurones in the medial prefrontal cortex of the rat (mPFC areas 24b, 25, and 32) containing strong NADPH diaphorase reactivity also contain GABA immunoreactivity. Also demonstrated is the co-localisation of NADPH diaphorase activity with immunoreactivity for the neuronal isoform of nitric oxide synthase (nNOS) in mPFC neurones. Qualitative and quantitative analyses in the light and electron microscopes indicate that strongly NADPH diaphorase reactive cells are a subpopulation of GABAergic local circuit neurones and constitute a very small proportion (0.6-1.1%) of neurones in rat mPFC. These results suggest that NADPH diaphorase reactive cells in rat mPFC can influence neural activity via GABA-mediated and NO-mediated mechanisms.

Dendritic spine density of NADPH diaphorase reactive neurons in the medial prefrontal cortex (mPFC) of the rat.

The density of spines has been calculated over the processes of NADPH diaphorase reactive neurons in the medial prefrontal cortex (mPFC) of the rat. Quantitative data indicate that diaphorase reactive dendrites ranged from being virtually aspiny to possessing moderate numbers of spines (0.66 +/- 0.23 spines/microns; mean +/- S.D.). The size and shape of dendritic spines varied from long thin 'drum sticks' (the most frequent type) to short stubby protrusions.

Two types of interneuron in the dorsal lateral geniculate nucleus of the rat: a combined NADPH diaphorase histochemical and GABA immunocytochemical study.

The rationale for this study was to provide a comprehensive light microscopical description of the morphology of diaphorase-reactive neurons and neuropil elements in the dorsal lateral geniculate nucleus (dLGN) of the rat. An additional objective was to quantitatively assess whether a subpopulation of the diaphorase-reactive neurons, previously shown to be GABA-immunoreactive, constitute a distinct type of local-circuit neuron in the rat dLGN. Diaphorase activity was localised in a population of predominantly bipolar fusiform neurons. These cells were weak to moderately stained and possessed the morphological features of intrinsic inhibitory neurons, previously called class B neurons in the rat dLGN. Quantitative estimates indicated that the diaphorase-reactive neurons constituted approximately 10% of the total neuron composition of the dLGN. The majority (about 83%) of the diaphorase-reactive cells were located in the lateral half of the nucleus. In addition, a dense plexus of diaphorase-reactive varicose fibres was found throughout the dLGN lying between the oriented fibre bundles coursing dorsoventrally through the LGN. Diaphorase-reactive punctae were found to be closely associated with the somata and proximal dendritic segments of nonreactive neurons and also with the stained proximal dendritic segments of diaphorase-reactive dLGN neurons. The source of the diaphorase-reactive fibres in the dLGN was unknown. Evidence suggests, however, that they are of extrinsic origin. The GABA-immunoreactive nature of the diaphorase neurons in the dLGN was demonstrated by colocalising GABA immunoreactivity within the somata of diaphorase-reactive cells. The majority (> 90%) of diaphorase-reactive dLGN neurons were GABA-immunopositive. Also present was a distinct population of GABA-immunopositive neurons that were not diaphorase-reactive. In this study, cells that were solely GABA-immunopositive have been called class B1 neurons, while cells that were both diaphorase-reactive and GABA-immunoreactive have been called class B2 neurons. Size-frequency distributions of somatic profile areas established that the two populations of GABA-immunoreactive neuron were significantly different. Class B1 neurons constituted 57%, with class B2 cells representing 43% of all GABA-immunostained neurons in the rat dLGN. The characteristic morphological features, neurochemical identity and frequency of the diaphorase-reactive neurons in the rat dLGN indicate that they represent a subpopulation of inhibitory interneurons with the ability to affect intrinsic dLGN operations and thalamocortical interactions using the neuromodulator nitric oxide.

Multiple factors determine the sex ratio of postpartum-conceived Norway rat litters.

When Norway rat litters are nursed communally, postpartum-conceived litters born asynchronously are female biased at birth. To determine if one or more mechanisms produced this sex ratio bias, we studied litters conceived at a postpartum estrus, and systematically varied the presence or absence of a litter suckling during gestation. Four different factors biased the sex ratio of postpartum-conceived litters. Postpartum condition, implantation in a single uterine horn, and very large litter size favored the birth of female pups. The presence of suckling pups during gestation, on the other hand, brought the sex ratio up to parity. Each factor operated at or before implantation, and apparently did so asymmetrically in the two horns of the uterus. Thus, a litter's sex ratio at birth, even a sex ratio of parity, can be the result of multiple mechanisms of bias, some working in opposition to each other. Theories of sex ratio biasing must incorporate multiple mechanisms rather than focusing on single mechanisms within a given species.

An oriented framework of neuronal processes in the ventral lateral geniculate nucleus of the rat demonstrated by NADPH diaphorase histochemistry and GABA immunocytochemistry.

This study investigated the morphology and quantitative distribution of neurons containing NADPH diaphorase activity in the ventral lateral geniculate nucleus of the rat. The pattern of diaphorase staining revealed a strongly reactive lateral subdivision and a weakly staining medial subdivision. A characteristic feature of the diaphorase staining in the lateral part was its "stripe-like" appearance. These "diaphorase stripes" resulted from regions of strong somatic and neuropil diaphorase activity lying between unstained fibre bundles coursing dorsoventrally through the nucleus. Two distinct populations of diaphorase reactive cell types were present--class A and class B neurons. The ratio of class A to class B diaphorase neurons was approximately 14:1 (A:B). Diaphorase reactive neurons made up 73% of the total neuron population in the lateral subdivision, and 31% in the medial subdivision. A third population of cells was found exclusively in the optic tract--class C neurons. Quantitative analyses in the coronal and sagittal planes indicated that the principal processes of both class A and class B neurons were oriented preferentially--either parallel with, or perpendicular to the outlying optic tract. Diaphorase enzyme histochemistry in combination with GABA immunocytochemistry demonstrated the co-localization of GABA immunoreactivity in the majority of class B neurons, whereas class A and class C neurons were GABA immunonegative. Furthermore a large population of GABA-immunoreactive neurons was present that were not stained for diaphorase activity. From this and previous studies, it can be concluded that a high proportion of the diaphorase reaction class A neurons are geniculotectal projection cells, while diaphorase reaction class B neurons represent a numerically small subpopulation of "local-circuit" inhibitory neurons. Since diaphorase activity co-localizes with nitric oxide synthase, the results indicate the likely involvement of nitric oxide in the neuronal operations of both subpopulations of geniculotectal projection neurons and "local-circuit" GABAergic neurons in the rat's ventral lateral geniculate nucleus.

Histochemical localization of NADPH-dependent diaphorase (nitric oxide synthase) activity in vascular endothelial cells in the rat brain.

This study investigated the localization of NADPH-dependent diaphorase activity within vascular endothelial cells in the rat brain. Light microscope observations showed that in addition to neurons and neuronal processes stained histochemically for NADPH-dependent diaphorase activity, endothelial cells in many medium to large diameter (20-100 microns) blood vessels were also stained. These vessels were either attached to the pial surface or contained within the substance of the tissue. In vascular endothelia, the formazan end-product of the diaphorase reaction was deposited as discrete clusters of darkly stained punctae that were located around the nucleus of these cells. Correlated light- and electron-microscopical examination revealed that the sites of formazan deposition occurred in regions of endothelial cytoplasm devoid of smooth and rough endoplasmic reticulum and of mitochondria. Since endothelial NADPH dependent diaphorase activity co-localizes with the activity of nitric oxide synthase (the synthetic enzyme for nitric oxide) these observations suggest that in vascular endothelial cells nitric oxide synthase may be a highly localized soluble cytosolic enzyme not structurally associated with any subcellular organelle. In addition, specific regions of the smooth muscle cells encircling the larger diameter blood vessels clearly demonstrated NADPH dependent diaphorase activity. Unmyelinated fibres and fibre-plexi surrounding blood vessels on the pial surface were also stained. The results of this study show specific NADPH dependent diaphorase activity in vascular endothelial cells in the rat brain. Therefore, together with neurons, endothelial cells may control nitric oxide-dependent vasodilation thereby regulating local blood flow in the brain.

A monosynaptic pathway from an identified vasomotor centre in the medial prefrontal cortex to an autonomic area in the thoracic spinal cord.

Chemical microstimulation (1 mM L-glutamate or 25 mM KCl) of the medial prefrontal cortex of anaesthetized rats produced falls in systolic and diastolic blood pressure of similar magnitude, without a change in heart rate. Application of the lectin Phaseolus vulgaris leucoagglutinin by iontophoresis from an adjacent barrel of the same micropipette revealed a direct projection to the central autonomic area of the thoracic spinal cord from this vasomotor area, which is equivalent to the region called prelimbic cortex by Krettek and Price [J. comp. Neurol. (1977) 171, 157-192] or Cg3 by Paxinos and Watson [The Rat Brain in Stereotaxic Coordinates (1986)]. Labelled axons descended in the dorsal corticospinal tract in the cervical spinal cord, where they displayed a few varicosities. In the thoracic spinal cord, labelled fibres occurred bilaterally in the gray matter, predominantly in the central autonomic area, where they displayed many varicosities. Electron microscope studies revealed that the anterogradely labelled varicosities in the central autonomic area were vesicle-filled boutons that formed asymmetric synaptic contacts. The synaptic targets were small dendrites or dendritic protrusions that were characterized by a high incidence of multivesicular bodies and coated vesicles. We conclude that a monosynaptic pathway that originates from a physiologically-defined vasomotor area in the medial prefrontal cortex terminates on a characteristic type of neuron in the central autonomic area of the thoracic spinal cord.