Sonic hedgehog control of size and shape in midbrain pattern formation.

Little is known about how patterns of cell types are organized to form brain structures of appropriate size and shape. To study this process, we employed in vivo electroporation during midbrain development to create ectopic sources of Sonic Hedgehog, a signaling molecule previously shown to specify different neuronal cell types in a concentration-dependent manner in vitro. We provide direct evidence that a Sonic Hedgehog source can control pattern at a distance in brain development and demonstrate that the size, shape, and orientation of the cell populations produced depend on the geometry of the morphogen source. Thus, a single regulatory molecule can coordinate tissue size and shape with cell-type identity in brain development.

Patterning the mammalian cerebral cortex.

When and how is the area map of the cerebral cortex set up during development? Recent studies indicate that regional pattern emerges early in cortical neurogenesis, and that this pattern does not require cues from extrinsic innervation. Studies of mutant mice indicate a role for embryonic signaling centers and for specific transcription factors in regionalizing the cortex. Thus, it is increasingly probable that the cortex is partitioned using the same types of mechanisms--and in some cases, the same gene families--that are used in patterning other parts of the embryo. This emerging model is likely to be the basis for many future studies. However, new evidence also confirms the special nature of the cerebral cortex, in that cues from developing connections appear to modify and refine the final area map.

Dorsoventral patterning of the telencephalon is disrupted in the mouse mutant extra-toes(J).

Little is known about the mechanisms that control the development of regional identity in the mammalian telencephalon. The Gli family of transcription factor genes is involved in the regulation of pattern at many sites in the embryo and is expressed in the embryonic mouse telencephalon. We have analyzed telencephalic patterning in the extra-toes (J) (Xt(J)) mouse mutant, which carries a deletion in the Gli family member Gli3. We report that dorsoventral patterning of the telencephalon is dramatically disrupted in the Xt(J) mutant. Specific dorsal telencephalic cell types and gene expression patterns are lost in homozygous Xt(J) mutants, and features of ventral telencephalic identity develop ectopically in the dorsal telencephalon. This partial ventralization of the dorsal telencephalon does not appear to be induced by an expansion of Sonic hedgehog expression in the telencephalon, but may be due to a loss of Bmp and Wnt gene expression in a putative dorsal telencephalic signaling center, the cortical hem. Our findings suggest that in dorsal telencephalon Gli3 is needed to repress ventral telencephalic identity.

The hem of the embryonic cerebral cortex is defined by the expression of multiple Wnt genes and is compromised in Gli3-deficient mice.

In the developing vertebrate CNS, members of the Wnt gene family are characteristically expressed at signaling centers that pattern adjacent parts of the neural tube. To identify candidate signaling centers in the telencephalon, we isolated Wnt gene fragments from cDNA derived from embryonic mouse telencephalon. In situ hybridization experiments demonstrate that one of the isolated Wnt genes, Wnt7a, is broadly expressed in the embryonic telencephalon. By contrast, three others, Wnt3a, 5a and a novel mouse Wnt gene, Wnt2b, are expressed only at the medial edge of the telencephalon, defining the hem of the cerebral cortex. The Wnt-rich cortical hem is a transient, neuron-containing, neuroepithelial structure that forms a boundary between the hippocampus and the telencephalic choroid plexus epithelium (CPe) throughout their embryonic development. Indicating a close developmental relationship between the cortical hem and the CPe, Wnt gene expression is upregulated in the cortical hem both before and just as the CPe begins to form, and persists until birth. In addition, although the cortical hem does not show features of differentiated CPe, such as expression of transthyretin mRNA, the CPe and cortical hem are linked by shared expression of members of the Bmp and Msx gene families. In the extra-toesJ (XtJ) mouse mutant, telencephalic CPe fails to develop. We show that Wnt gene expression is deficient at the cortical hem in XtJ/XtJ mice, but that the expression of other telencephalic developmental control genes, including Wnt7a, is maintained. The XtJ mutant carries a deletion in Gli3, a vertebrate homolog of the Drosophila gene cubitus interruptus (ci), which encodes a transcriptional regulator of the Drosophila Wnt gene, wingless. Our observations indicate that Gli3 participates in Wnt gene regulation in the vertebrate telencephalon, and suggest that the loss of telencephalic choroid plexus in XtJ mice is due to defects in the cortical hem that include Wnt gene misregulation.

Isoform-specific immunological detection of newt retinoic acid receptor delta 1 in normal and regenerating limbs.

Retinoic acid (RA) exerts a variety of effects on the regenerating urodele limb including positional respecification of the blastema. The major RA receptor expressed in the newt limb and blastema is the delta 1 isoform and, in order to detect delta 1 in this context, we have made five affinity-purified antibodies against fusion proteins and peptides from non-overlapping regions of the molecule. These antibodies have been evaluated by reaction with transfected COS-7 cells, newt limb cells in culture and newt limb tissue sections. The most informative antibodies were RP6, directed against N-terminal region A sequence, and RP8, directed against C-terminal sequence. In western blots of blastemal extracts, delta 1 protein was detected as two major bands of immunoreactivity at positions consistent with the employment of two candidate methionine initiators identified by cDNA sequencing. Staining of adult limb sections with RP6 and RP8 showed reactivity in half of the nuclei in epidermal and mesenchymal tissues, a heterogeneity that was observed with adjacent nuclei in muscle fibres. In the regenerating limb, nuclei in the blastemal mesenchyme and wound epidermis were strongly reactive, although no axial variation in expression was detected.

Delta retinoic acid receptor isoform delta 1 is distinguished by its exceptional N-terminal sequence and abundance in the limb regeneration blastema.

In amphibian limb regeneration memory for position in the proximal-distal axis can be respecified by retinoic acid. The favoured candidates to mediate this effect are the retinoic acid receptors (RARs) and of the RARs identified in the regeneration blastema, the delta receptor is the most abundant. The presence in blastemal mesenchyme of at least two delta receptor isoforms, delta 1 and delta 2, alternatively spliced at the A-B junction, was demonstrated in expression studies and by PCR cloning. The delta 1 receptor is abundant in regenerative structures such as the limb and tail, whereas the delta 2 and alpha receptors show a more uniform pattern of expression across adult newt tissues. Full-length cloning of the delta 1 receptor established the presence of an unusually long open reading frame and N-terminal sequence that appears unique among vertebrate retinoic acid receptors. Transient transfection of expression constructs into COS cells followed by Western blotting confirmed the existence of at least three potential initiation sites for delta 1 translation. The possibility that delta 1 RAR expression may specify positional memory directly was tested in RNase protection experiments. delta 1 receptor message is increased on amputation, but does not exhibit a pronounced differential distribution along the proximal-distal axis in normal and regenerating limbs, nor does it show a persistent alteration in expression levels following a dose of retinoic acid sufficient to respecify position. The possibility that the morphogenetic effects of RA may be mediated through receptor interactions is raised by the finding that single mesenchymal blastemal cells in culture can express multiple RAR subtypes (delta 1 and alpha) and isoforms (delta 1 and delta 2).

Retinoids and their targets in vertebrate development.

The elaboration of the effect of retinoic acid on limb morphogenesis has prompted renewed investigation into the teratology of retinoic acid treatment, with the hope that such analysis might give insight into mechanisms of vertebrate patterning. Retinoids, their nuclear receptors and their cytoplasmic binding proteins are now known to be deployed throughout development, but the extent to which they are natural agents of morphogenesis remains obscure. The study of retinoic acid receptors may offer molecular insight into gene regulation underlying vertebrate pattern formation.

Compartmental organization of the thalamostriatal connection in the cat.

The compartmental organization of the thalamostriatal connection in the cat was studied by labelling thalamic fibers in anterograde axonal transport experiments and comparing their striatal distributions with the arrangement of striosomes and matrix tissue identified by histochemical staining methods. When analyzed according to their principal compartmental targets in dorsal striatum, the thalamic deposits indicated the existence of medial and lateral divisions within the thalamostriatal projection. Nuclei of the medial division, which includes parts of the thalamic midline, projected primarily to striosomes. The lateral division, which embraces the anterior and posterior intralaminar groups, the rostral ventral tier nuclei, and parts of the posterior lateral nuclear complex, predominantly innervated matrix tissue. In the dorsal division of the nucleus accumbens, the medial system preferentially terminated in zones that stain heavily in butyrylcholinesterase and substance P preparations, but fibers from both the medial and the lateral systems largely avoided the histochemically marked compartments such as the border islands of the nucleus accumbens that are seen elsewhere in the ventral striatum. Medial division: Thalamic deposits involving the paraventricular and rhomboid nuclei of the thalamic midline elicited labelling of striosomes and, invariably, ventral extrastriosomal matrix, the nucleus accumbens, and the amygdala. This projection was topographically organized: rostral thalamic deposits elicited labelling in the medial caudate nucleus and the medial nucleus accumbens. More caudal injections produced more lateral labelling. Lateral division: The lateral division is composed of at least three projection systems distinguished by their patterns of matrix innervation. Deposits involving the anterior intralaminar nuclei and the striatally projecting cells located lateral to the stria medullaris (anterior intralaminar complex) produced an even, diffuse labelling of the matrix tissue and weak labelling of the striosomes. Injections placed in the ventroanterior, ventrolateral, and ventromedial nuclei (rostral ventral complex) elicited fibrous labelling of matrix tissue that often showed nonstriosomal inhomogeneities. Deposits involving the centromedian and parafascicular nuclei (posterior intralaminar complex) produced a highly variable pattern of matrix labelling that included both homogeneous and decidedly patchy innervations of the extrastriosomal matrix. Each of these lateral thalamostriatal systems showed a similar spatial organization, whereby dorsoventral and mediolateral thalamic axes were roughly preserved in the projection to striatum.

A simple ordering of neocortical areas established by the compartmental organization of their striatal projections.

The compartmental organization of corticostriatal projections from the fronto-orbito-insular cortex was studied in the cat. Cortical areas in this field were found to have a highly organized projection to the striatum, selectively innervating striosomes dorsally and predominantly avoiding them ventrally within their striatal fields of termination. These observations have two important implications for striatal processing. First, some cortical areas preferentially terminate in different compartments in different parts of the striatum. Therefore, the sources of input to striosomes and matrix are not categorical but switch according to the striatal region considered. Second, three properties of the bicompartmental termination pattern--one-dimensionality, common polarization, and multiple positions at which the pattern switched from "fills" to "avoids"--allowed us to order the corticostriatal projections with respect to one another. This ordering of the striatal projections of cortical areas implies an ordering of the cortical areas themselves, one that is independent of transcortical connections. For the corticostriatal projections described in this report, the ordering is [parietal, dorsomedial prefrontal, ventrolateral prefrontal, insular, rostral temporal] cortex. Our analysis suggests that a major function of striatal compartmentalization is to segregate and then bring together inputs from cortical areas at different positions in this ordering. The ordering may also serve as a simple format for specifying corticostriatal connections in development.

Identification of a novel retinoic acid receptor in regenerative tissues of the newt.

In urodele amphibians, the progenitor cells that regenerate amputated limbs (known as the blastema) normally replace only the missing structures. After systemic delivery of retinoic acid (RA), more proximal structures are also formed, indicating that RA can control position specification in the proximal-distal axis of the regenerating limb. According to dose and experimental context, retinoids can also re-specify the anteroposterior axis of the limb, induce deletions of skeletal elements, or block re-growth completely. To study the molecular basis of these morphogenetic effects, we screened complementary DNA libraries of newt regenerative tissues (limbs and tails) for hormone nuclear receptors activated by RA. Two functional retinoic acid receptors (RARs) were identified, one of which is the newt homologue of the human alpha-receptor (RAR alpha). The second receptor, called RAR delta, is novel. Sequence analysis suggests that the composite newt RAR previously reported is chimaeric, consisting of 5'RAR-beta-like and 3' RAR delta clones. We conclude that multiple RARs are expressed during limb regeneration in amphibians and suggest that receptor heterogeneity may underlie the different effects of retinoids on limb morphogenesis.

High efficiency transformation of E. coli by high voltage electroporation.

E. coli can be transformed to extremely high efficiencies by subjecting a mixture of cells and DNA to brief but intense electrical fields of exponential decay waveform (electroporation). We have obtained 10(9) to 10(10) transformants/micrograms with strains LE392 and DH5 alpha, and plasmids pUC18 and pBR329. The process is highly dependent on two characteristics of the electrical pulse: the electric field strength and the pulse length (RC time constant). The frequency of transformation is a linear function of the DNA concentration over at least six orders of magnitude; and the efficiency of transformation is a function of the cell concentration. Most of the surviving cells are competent with up to 80% transformed at high DNA concentration. The mechanism does not appear to include binding of the DNA to the cells prior to entry. Possible mechanisms are discussed and a simple procedure for the practical use of this technique is presented.

Fibers from the basolateral nucleus of the amygdala selectively innervate striosomes in the caudate nucleus of the cat.

The compartmental organization of the amygdalostriatal projection was studied in the cat by comparing staining patterns seen by cholinesterase enzyme histochemistry with the distribution of fibers labelled with a horseradish peroxidase-wheat germ agglutinin conjugate or by incorporation of 35S-methionine or 3H-leucine. Fibers from the basolateral nucleus of the amygdala were found to innervate selectively acetylcholinesterase-poor striosomes demonstrated in the caudate nucleus and butyrylcholinesterase-rich zones observed in the anterodorsal nucleus accumbens. In no case were the amygdalar fibers fully restricted to striosomes, but the nature and degree of labelling of the striatal matrix, as well as the range of the labelled fibers in dorsal striatum, varied with the positions of the injection sites.

Direct demonstration of a correspondence between the dopamine islands and acetylcholinesterase patches in the developing striatum.

The distribution of dopamine-containing processes in the striatum of fetal and neonatal cats was studied by immunohistochemical and glyoxylic acid histofluorescence methods and compared to the distribution of acetylcholinesterase (acetylcholine acetylhydrolase, EC 3.1.1.7) observed by thiocholine histochemistry in the same or serially adjoining sections. Both methods for demonstrating the dopamine innervation revealed the characteristic patchwork of dopamine "islands" in the caudoputamen, in which catecholamine histofluorescence or tyrosine hydroxylase [tyrosine 3-monooxygenase; L-tyrosine, tetrahydropteridine:oxygen oxidoreductase (3-hydroxylating), EC 1.14.16.2]-like immunoreactivity was concentrated into 0.2- to 0.6-mm-wide patches. Both methods also demonstrated a high degree of patterning of the dopamine innervation in the ventral striatum, including the nucleus accumbens septi. A detailed and striking match was found between these configurations and the compartmental distribution of acetylcholinesterase observed in the caudoputamen and ventral striatum of the same brains. The correspondence between the dopamine and acetylcholinesterase figures was most obvious in the fetal brains, in which the background acetylcholinesterase staining was lightest, but matches between the dopamine islands and acetylcholinesterase patches could still be seen in the kittens. There was no clear alignment of striatal cell bodies stained for acetylcholinesterase with either the dopamine or the acetylcholinesterase-positive patches. Nor was there an obvious correspondence between dopamine and acetylcholinesterase in the striatal background matrix. We conclude that, at least during ontogenesis, it is the clustered arrangements of dopamine and acetylcholinesterase that are, in particular, tightly linked, and we suggest that information about the maturation of these clusters may be crucial in assessing the functions of striatal dopamine and acetylcholinesterase in the adult.

The fronto-striatal projection in the cat and monkey and its relationship to inhomogeneities established by acetylcholinesterase histochemistry.

The distribution of cortico-striatal fibers from the frontal cortex was studied by autoradiography in cat and monkey and compared with the patterns of striatal acetylcholinesterase activity observed in the same brains. The clustering patterns of labeled fronto-caudate fibers were found to correspond to a large extent to the inhomogeneities in enzyme activity identified by the acetylthiocholinesterase staining method.

Clumping of acetylcholinesterase activity in the developing striatum of the human fetus and young infant.

The distribution of acetylcholinesterase activity (acetylcholine acetylhydrolase, EC 3.1.1.7) in the developing human striatum has been studied histochemically in autopsy material from fetal brains of estimated gestational ages 16-29 weeks (180-1000 g) and from the brains of infants 2 days to 4 months old. The findings provide evidence that striatal acetylcholinesterase activity in the human fetus and neonate is concentrated in a network of densely stained zones that appear in cross section as variably shaped 0.5- to 1.0-mm-wide dark patches distributed in a lighter background matrix. An orderly arrangement of the patches seemed well established in the putamen by the 16th-18th week of gestation (crown-rump length 14-15 cm) but in the caudate nucleus the pattern was still not fully elaborated at these ages. The lateroventral part of the caput was mainly dark and its rostromedial margin, though rich early on in pseudocholinesterase activity, was still without strong acetylcholinesterase activity as late as 20-21 weeks (crown-rump length 16-20 cm). The ganglionic eminence at these ages was sharply divided into a dorsal part with little cholinesterase activity and a ventral part with a high content of pseudocholinesterase. Little information was gained about striatal development during late stages of gestation, but in three 5- to 7-month fetal specimens not only dark patches but also patches with dark perimeters and pale centers were present. Clumping of cholinesterase activity appeared at birth and up to the third month of postnatal life. The patches in both caudate nucleus and putamen were dark and of fairly uniform tint in the striatum of the young infant and the matrix staining was darker than in the fetuses. Around the fourth postnatal month hints of the mature pattern were present, with zones of low cholinesterase activity appearing against a dark background in the caudate nucleus and (in one case) a nearly homogeneous staining pattern appearing in the putamen.