Thalamocortical Axons Are Influenced by Chemorepellent and Chemoattractant Activities Localized to Decision Points along Their Path

Thalamocortical Axons Are Influenced by Chemorepellent and Chemoattractant Activities Localized to Decision Points along Their Path

Developmental Biology 208, 430 – 440 (1999) Article ID dbio.1999.9216, available online at http://www.idealibrary.com on Thalamocortical Axons Are In...

512KB Sizes 0 Downloads 35 Views

Developmental Biology 208, 430 – 440 (1999) Article ID dbio.1999.9216, available online at http://www.idealibrary.com on

Thalamocortical Axons Are Influenced by Chemorepellent and Chemoattractant Activities Localized to Decision Points along Their Path Janet E. Braisted, Rebecca Tuttle, and Dennis D. M. O’Leary Molecular Neurobiology Laboratory, The Salk Institute, 10010 North Torrey Pines Road, La Jolla, California 92037

Thalamocortical axons (TCAs), which originate in dorsal thalamus, project ventrally in diencephalon and then dorsolaterally in ventral telencephalon to their target, the neocortex. To elucidate potentially key decision points in TCA pathfinding and hence the possible localization of guidance cues, we used DiI-tracing to describe the initial trajectory of TCAs in mice. DiI-labeled TCAs extend ventrally on the lateral surface of ventral thalamus. Rather than continuing this trajectory onto the lateral surface of the hypothalamus, TCAs make a sharp lateral turn into ventral telencephalon. This behavior suggests that the hypothalamus is repulsive and the ventral telencephalon attractive for TCAs. In support of this hypothesis, we find that axon outgrowth from explants of dorsal thalamus is biased away from hypothalamus and toward ventral telencephalon when cocultured at a distance in collagen gels. The in vivo DiI analysis also reveals a broad cluster of retrogradely labeled neurons in the medial part of ventral telencephalon positioned within or adjacent to the thalamocortical pathway prior to or at the time TCAs are extending through it. The axons of these neurons extend into or through dorsal thalamus and appear to be coincident with the oppositely extending TCAs. These findings suggest that multiple cues guide TCAs along their pathway from dorsal thalamus to neocortex: TCAs may fasciculate on the axons of ventral telencephalic neurons as they extend through ventral thalamus and the medial part of ventral telencephalon, and chemorepellent and chemoattractant activities expressed by hypothalamus and ventral telencephalon, respectively, may cooperate to promote the turning of TCAs away from hypothalamus and into ventral telencephalon. © 1999 Academic Press Key Words: axon guidance; axon scaffold; dorsal thalamus; globus pallidus; hypothalamus; internal capsule; perireticular cells; ventral telencephalon; ventral thalamus.

INTRODUCTION Growth cones use multiple guidance cues that act at a distance or locally to navigate to their targets within the developing brain (Goodman, 1996). Long-range cues consist of soluble molecules secreted by the intermediate or final targets of advancing growth cones and are thought to act at a distance by diffusing and establishing a concentration gradient that growth cones advance either up or down. In contrast, local cues consist of molecules that are localized to the extracellular matrix surrounding the cell that secretes them or are anchored to the cell membrane and guide growth cones via a contactmediated mechanism. Diffusible and contact-mediated guidance cues steer growth cones in one of two ways: by repelling advancing growth cones and steering them away from inappropriate regions along their pathway or

430

by attracting growth cones and directing them toward their intermediate or final targets. The major axon tract in the mammalian forebrain, the internal capsule, traverses the ventral telencephalon and interconnects neocortex with subcortical structures. The internal capsule is composed mainly of thalamocortical and corticothalamic axons. Thalamocortical axons originate in dorsal thalamus, extend ventrally through ventral thalamus, and turn dorsolaterally to extend through ventral telencephalon to their target, the neocortex. Corticothalamic axons originating in the neocortex take a similar, but opposite, path to their target, the dorsal thalamus. During development, these two projections first converge in the ventral telencephalon. Little is known about the cues that control the pathfinding of thalamocortical axons, although several studies suggest that axonal scaffolds may be involved. For example, it has 0012-1606/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

431

Thalamocortical Axon Pathfinding

been proposed that axons originating from cells within the mantle zone derived from the ganglionic eminences in the ventral telencephalon that project into dorsal thalamus serve as a scaffold to guide thalamocortical axons into the ventral telencephalon (Metin and Godement, 1996; Molnar et al., 1998). In addition, the “handshake hypothesis” proposes that thalamocortical axons use cortical efferent axons as a scaffold to navigate from the ventral telencephalon to their target, the neocortex (Molnar and Blakemore, 1995; Molnar et al., 1998). Diffusible attractant activities have been suggested to play a role in the guidance of cortical efferent axons to their intermediate target, the ventral telencephalon. Embryonic cortical axons orient both in vivo (Richards et al., 1997) and in vitro (Metin and Godement, 1996; Richards et al., 1997) toward the developing ventral telencephalon. In vitro assays demonstrate that the chemoattractant netrin-1, which is expressed in the ventral telencephalon at the time cortical axons are extending through it (Serafini et al., 1996; Metin and Godement, 1996), mimics the ventral telencephalon-derived, chemoattractant activity (Metin et al., 1997; Richards et al., 1997). These data suggest that netrin-1 may contribute to the guidance of cortical axons into and through the developing ventral telencephalon. Chemorepellents also have been implicated in the guidance of forebrain axons. For example, in vitro studies suggest that a midline-derived, diffusible repellent activity deflects axons of the olfactory tract laterally (Pini, 1993). Comparable studies examining the potential roles of chemoattractants and chemorepellents on thalamocortical axon pathfinding have not been reported. The principal goal of this study was to investigate the potential contribution of guidance cues, both attractant and repellent, to thalamocortical axon pathfinding. As a first step toward this goal, we describe the early trajectory of thalamocortical axons in mice using DiI tracing. We find that thalamocortical axons do not penetrate the hypothalamus at any time during development. Instead, they make a sharp lateral turn near the ventral thalamic– hypothalamic border and enter the ventral telencephalon. These in vivo data suggest that the hypothalamus is repulsive and the ventral telencephalon is attractive for thalamocortical axons. Supporting these hypotheses, we find that dorsal thalamic axons are repelled by hypothalamus and attracted by ventral telencephalon in vitro. In addition, our DiI-labeling results confirm and extend upon previous reports (Metin and Godement, 1996; Molnar et al., 1998) of a scaffold of ventral telencephalic axons present at the correct time and place to guide thalamocortical axons through ventral thalamus and the ventral telencephalon via a contact-mediated mechanism. Together these data suggest that thalamocortical axons use multiple guidance cues, both diffusible and contact-mediated, to navigate from the dorsal thalamus to the neocortex.

MATERIALS AND METHODS Animals Embryonic day (E) 12.5 to E15 mice were obtained from timedpregnant ICR females (Harlan Sprague–Dawley). Embryos were staged according to Butler and Juurlink (1987). The day of insemination was designated E0.

DiI Labeling Embryo heads were immersion fixed in 4% paraformaldehyde overnight at 4°C. Brains were bisected along the midline, and tissues caudal to the diencephalon were removed to expose the caudal surface of dorsal thalamus. For E12.5 and E13.5 brains, a solution of the axonal tracer, DiI (1,19-dioctadecyl-3,3,39,39tetramethylindocarbocyanine perchlorate; Molecular Probes) (Honig and Hume, 1989) in dimethylformamide was pressureinjected into the dorsal thalamus through a glass micropipet. Multiple injections were made from the caudal and medial aspects of the dorsal thalamus with the intent of maximizing the number of thalamocortical axons labeled. For E15 brains, several crystals of DiI were placed into the dorsal thalamus. Embryos were left in 4% paraformaldehyde at 30°C for 2 or more weeks to allow the DiI to transport the full length of the axons. Brains were then embedded in 3% low-melting-point agar (Seaplaque; FMC Bioproducts), and 100-mm coronal sections were cut using a Vibratome. Sections were counterstained with a 0.02% solution of the nuclear stain bisbenzimide (Sigma), and labeled cells and processes were photographed under rhodamine (DiI) and UV (bisbenzimide) illumination.

Explant Preparation Timed-pregnant mice were anesthetized with sodium pentobarbital and placed on a thermoregulated heating pad. Abdomens were sterilized and cesarean-sectioned to expose the uterus. Embryos (E13.5–E14) were decapitated and the heads transferred to cold L15 medium supplemented with 0.6% glucose (L15– glucose). Brains were removed from the skulls, embedded in 3% low-melting-point agar in L15– glucose, and sectioned coronally at 200 mm using a Vibratome. Embryos were processed two to three at a time. Sections in which the internal capsule bridges the diencephalon and ventral telencephalon (e.g., Figs. 3B and 3B9) were collected, the neocortex was removed, and a cut was made between the diencephalon and the ventral telencephalon (Fig. 4A). The resulting two pieces served as hypothalamic and ventral telencephalic explants. Dorsal thalamus was isolated from more caudal sections by making two cuts— one through the external medullary lamina/medial lemniscus to remove the ventral thalamus and hypothalamus and a second to remove the epithalamus. The resulting piece of dorsal thalamic tissue was bisected to yield two explants— one lateral and one medial.

Collagen Gel Assay Collagen was extracted from 8- to 10-cm-long, juvenile rat tails. Collagen solution was prepared by mixing 900 ml of collagen extract with 100 ml of 103 MEM and 11–18 ml of a 7.5% solution of sodium bicarbonate. For cocultures, 25 ml of collagen solution was spread over the center of each well of a four-well dish (Nunc) and allowed to gel. Explants were placed onto this base, excess

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

432

Braisted, Tuttle, and O’Leary

L15– glucose was removed, 75 ml of collagen solution was added on top, and then the explants were positioned 150 –300 mm apart. The orientation of dorsal thalamic explants relative to hypothalamus or ventral telencephalon was random. Dorsal thalamus was cultured alone as a control to determine its normal growth in the absence of any other tissue. We did not coculture dorsal thalamus with an “irrelevant” test tissue since we found that two distinct tissues, both of which lie along the in vivo pathway of thalamocortical axons, elicit opposing effects on extending thalamocortical axons in vitro (see Results). The collagen was allowed to gel before addition of 500 m l of DMEM/F12 containing penicillin– streptomycin, 0.6% D-glucose, 2 mM glutamine, and 5% rat serum.

Analysis of Axon Growth in Vitro After 1.5 days in a humidified, 37°C, CO 2 incubator, cocultures were photographed using phase-contrast optics. Dorsal thalamic axon outgrowth was assigned to one of three categories: explants that had more axon outgrowth either “toward” or “away from” the test explant or explants that had similar amounts of axon outgrowth on both of these sides and were therefore scored as “symmetric.” In cases in which dorsal thalamic explants were cultured alone the left side of the dish was designated away and the right side toward. Since the hypothalamic and ventral telencephalic explants are morphologically distinguishable, scoring of cocultures could not be done blind. Therefore, two or three individuals scored the cultures independently. Data were analyzed statistically using the x 2 test.

RESULTS

FIG. 1. Trajectory of thalamocortical axons (TCAs) at E15. Coronal section through E15 mouse forebrain showing DiI-labeled neurons and processes (B and high magnification inset of boxed area) or bisbenzimide nuclear counterstain (A). Multiple DiI crystals (asterisks) were implanted in the dorsal thalamus (dTh). Anterogradely labeled TCAs project ventrolaterally in ventral thalamus (vTh), make a sharp lateral turn (arrowhead) near the dorsal border of hypothalamus (Hy), and then extend dorsolaterally through ventral telencephalon (vTel) toward their target, the neocortex (Ctx). Retrogradely labeled neurons are apparent in vTh (large arrow in B) and vTel (a few indicated with small arrows in B inset). In Figs. 1–3, only the right half of section is shown; dorsal is up, midline is to the left. ic, internal capsule. Scale bars: B, 500 mm; B inset, 50 mm.

Characterization of the Initial Trajectory of Thalamocortical Axons Dorsal thalamic neurons are generated between E10 and E13 in mice (Angevine, 1970). By E14 the first thalamocortical axons have reached the neocortex (Bicknese et al., 1994; Me´ tin and Godement, 1996). To define the thalamocortical axon pathway, these axons were anterogradely labeled by placing DiI into the dorsal thalamus of E12.5–E15 mice. In this paper, the term “ventral telencephalon” refers to a region extending from the ventral surface of the telencephalon to the lateral ventricle; this region includes medial, lateral, and caudal ganglionic eminences, as well as structures ventral to them, such as the caudate putamen, globus pallidus, and amygdala.

At E15, the thalamocortical axon projection is similar to that of the adult (Fig. 1). DiI-labeled axons are predominantly thalamocortical, as only a few retrogradely labeled cells are found in cortex (not shown). Thalamocortical axons extend ventrolaterally through the ventral thalamus and, as they approach the hypothalamus, make a sharp lateral turn and extend dorsolaterally through ventral telencephalon to neocortex (Fig. 1). Many retrogradely labeled neurons are present in the lateral part of the ventral thalamus, and a few are apparent in the medial part of the ventral telencephalon (Fig. 1B, inset). At this age, the pathway of thalamocortical axons through the ventral tel-

FIG. 2. Thalamocortical axon pathway at E12.5. A rostral-to-caudal (A-to-C) series of coronal sections from a single E12.5 brain injected with DiI in dorsal thalamus (dTh) (asterisks) showing DiI-labeled neurons and processes (A9–C9), bisbenzimide counterstain (A–C), and higher magnification images of boxed areas in A and A9–C9 (A0–C0). At rostral levels, a broad cluster of retrogradely labeled neurons is present in the medial part of ventral telencephalon (vTel) and appears to extend to the diencephalic–telencephalic border (A0 and boxed area in A and A9). Arrows in A0 indicate four retrogradely labeled neurons in the plane of focus; other labeled cell bodies are out of the plane of focus. At intermediate and caudal levels, DiI-labeled neurons are present in the lateral part of the ventral thalamus (vTh; arrows in B0 and C0). Also, at intermediate levels, retrogradely labeled vTel and vTh axons and anterogradely labeled TCAs extend along the lateral surface of the vTh (large arrowhead in B9); DiI-labeled vTel axons and TCAs cross the diencephalic–telencephalic border at the ventralmost aspect of the vTel (small arrowhead in B9). At caudal levels, DiI-labeled axons originating mainly from neurons in the dorsal part of the vTh extend dorsally into dTh (large arrowhead in C0). Also at caudal levels, labeled axons form a fascicle at the lateral surface of vTh (small arrowheads in C0). Labeled axons and neurons are also apparent in lateral habenula (open arrowhead in C9). Scale bars: A9, 400 mm; A0, 50 mm; B0, 115 mm; C0, 160 mm.

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

Thalamocortical Axon Pathfinding

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

433

434

Braisted, Tuttle, and O’Leary

encephalon is positioned beneath the proliferative zone of the ganglion eminence and above a thick “mantle zone” composed of postmitotic cells. Labeling patterns were examined at two earlier ages, E12.5 and E13.5, to determine the extent and disposition of anterogradely labeled thalamocortical axons and retrogradely labeled neurons along their pathway. In E12.5 brains, few thalamocortical axons have entered the ventral telencephalon, although a projection from ventral telencephalon to (and/or through) dorsal thalamus is evident. A broad cluster of retrogradely labeled neurons is present in the medial part of the ventral telencephalon (Figs. 2A–2A0). These neurons are likely to be globus pallidus neurons for several reasons: they are located in a region that will develop into the globus pallidus (Paxinos, 1995), the axons of globus pallidus neurons have been shown to project to dorsal thalamus and lateral habenula at later ages (Young et al., 1984; Groenewegen, 1988), and these cells appear to colocalize with a cell population that expresses Nkx2.1 (Tuttle et al., 1999), a marker for the globus pallidus (Shimamura et al., 1995; Olsson et al., 1997; Kohtz et al., 1998). At E12.5, no labeled axons are apparent lateral to these retrogradely labeled neurons. However, medial to the neurons a discrete bundle of DiI-labeled axons extends from the ventralmost aspect of the ventral telencephalon into the diencephalon near the ventral thalamic– hypothalamic border (Figs. 2B–2B0). These axons originate predominantly from ventral telencephalic neurons. In the ventral part of ventral thalamus, a discrete fascicle of labeled axons is present on the lateral surface, lateral to retrogradely labeled ventral thalamic neurons (Figs. 2C9 and 2C0). In the dorsal part of ventral thalamus, labeled axons are less fasciculated and project through retrogradely labeled neurons (Figs. 2C9 and 2C0). Thus, the DiI-labeled axons in ventral thalamus consist of retrogradely labeled axons of ventral thalamic and ventral telencephalic neurons. It is also likely that a proportion of the DiI-labeled axons in ventral thalamus are anterogradely labeled thalamocortical axons, as dorsal thalamic neurons in mice are generated between E10 and E13 (Angevine, 1970), and are therefore likely to have initiated axons by E12.5. The DiI-labeled neurons are likely to be reticular thalamic and zona incerta neurons, as both of these nuclei have been shown to project to dorsal thalamus. Neurons of the ventral lateral geniculate nucleus, which project to pretectum and superior colliculus, have axons that pass through dorsal thalamus (Ricardo, 1981; Jones, 1985) and therefore might also be retrogradely labeled by the DiI injections. However, at these early ages, identification of ventral thalamic nuclei by histological criteria is difficult. In summary, at E12.5, few thalamocortical axons project into ventral telencephalon, although the axons of ventral telencephalic and ventral thalamic neurons extend along the lateral aspect of ventral thalamus and enter dorsal thalamus. In E13.5 brains, many thalamocortical axons project into ventral telencephalon. A tight fascicle of thalamocortical axons is present on the lateral surface of the ventral

thalamus (Fig. 3C9). The axon bundle extends lateral to as well as through retrogradely labeled neurons in the dorsal part of ventral thalamus (Fig. 3C9), but extends lateral to retrogradely labeled neurons in the ventral part of ventral thalamus (Fig. 3C9). As the axon bundle approaches the ventral thalamic– hypothalamic border, it turns laterally and extends into ventral telencephalon (Fig. 3B9), but does not reach neocortex (Fig. 3A9). Retrogradely labeled ventral telencephalic neurons are apparent at the distal tips of the labeled axons (Fig. 3A9, inset), but appear to be less numerous than at E12.5 (Fig. 2A0). Thus, the tightly fasciculated bundle of DiI-labeled axons in the ventral telencephalon and ventral thalamus must consist predominantly of anterogradely labeled thalamocortical axons, in addition to a few retrogradely labeled axons of ventral telencephalic, and possibly ventral thalamic, neurons. At E13.5, cells of the ventral telencephalic mantle zone lay both dorsal and ventral to the path of thalamocortical axons forming the internal capsule (Fig. 3B). These findings indicate that thalamocortical axons do not extend into the hypothalamus, but instead make a sharp lateral turn at or near its dorsal border to enter the ventral telencephalon. The thalamocortical axon pathway from dorsal thalamus through the medial aspect of the ventral telencephalon is coincident with the axons of ventral telencephalic neurons that project to dorsal thalamus, and at the early time points examined (E12.5 and E13.5), ventral telencephalic neurons labeled from dorsal thalamus are found in front of the advancing thalamocortical axons. In contrast, it appears that most of the axons and cell bodies of ventral thalamic neurons that project into dorsal thalamus lie medial to the thalamocortical axon path. Our data indicate that the first thalamocortical axons reach ventral telencephalon around E12.5 and that a substantial number of axons arrive over the next day. The first thalamocortical axons appear to turn and extend into the ventral telencephalon at its ventral surface. However, by E13.5, a large population of cells is interposed between the thalamocortical axon pathway and the ventral surface of the ventral telencephalon, presumably as the result of cell migration that occurs subsequent to the extension of the early thalamocortical axons.

Chemoattractant and Chemorepellent Activities Are Expressed by Tissues along the Thalamocortical Axon Pathway Our in vivo data demonstrate that thalamocortical axons extending along the lateral surface of the ventral thalamus fail to continue along the lateral surface of the hypothalamus and instead make a sharp lateral turn into the ventral telencephalon. We therefore hypothesized that the lateral hypothalamus and the medial part of the ventral telencephalon express repellent and attractant activities, respectively, that promote the turning of thalamocortical axons away from the hypothalamus and into the ventral telencephalon. To test this hypothesis, we carried out collagen

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

435

Thalamocortical Axon Pathfinding

gel assays using explants from E13.5–E14 mice, ages at which thalamocortical axons are making this pathfinding decision in vivo. Explants of dorsal thalamus were cocultured adjacent to the lateral surface of the hypothalamus or the medial surface of the ventral telencephalon (Fig. 4A). Axon outgrowth was analyzed as described in Fig. 4B (also see Materials and Methods). In dorsal thalamic– hypothalamic cocultures, dorsal thalamic axon outgrowth is biased away from the hypothalamus (Figs. 4D and 4F): 55% of the dorsal thalamic explants have more axon outgrowth on the side facing away from the hypothalamus, and only 12% have more axon outgrowth on the side facing toward the hypothalamus. In contrast, in dorsal thalamic–ventral telencephalic cocultures, dorsal thalamic axon outgrowth is biased toward the ventral telencephalon (Figs. 4E and 4F): 55% of the dorsal thalamic explants have more axon outgrowth on the side facing toward the ventral telencephalon, and only 14% have more axon outgrowth on the side facing away from the ventral telencephalon. When dorsal thalamic explants are cultured alone as a control, axon outgrowth is not biased: 52% of the explants have symmetric axon outgrowth (Figs. 4C and 4F); the remainder are divided evenly between away (left side of dish) and toward (right side of dish) categories. The biases in axon growth observed in the cocultures are statistically significant (see legend to Fig. 4). Thus the hypothalamus releases a diffusible activity that repels dorsal thalamic axons, and the ventral telencephalon releases a diffusible activity that attracts them.

DISCUSSION The data presented here suggest that a variety of guidance cues, contact-mediated and diffusible, attractant and repellent activities, may act to control the pathfinding of thalamocortical axons. The localization of these potential guidance cues relative to the path of thalamocortical axons within the developing forebrain is summarized in Fig. 5.

FIG. 3. Trajectory of thalamocortical axons at E13.5. A rostral-tocaudal (A-to-C) series of coronal sections from a single E13.5 brain injected with DiI in dorsal thalamus (dTh; asterisks) showing DiI-labeled cells and processes (A9–C9), or bisbenzimide counterstain (A–C). (A9 inset) Higher magnification of boxed area in A9. DiI-labeled TCAs can be traced from their origin in the dTh to an area approximately midway through the ventral telencephalon

(vTel; arrowhead in A9). At rostral levels, retrogradely labeled vTel neurons are apparent just lateral to the TCAs (boxed area in A and A9, arrows in A9 inset); their axon pathway is coincident with that of TCAs. At intermediate levels, DiI-labeled TCAs and vTel axons form a bundle at the ventral telencephalic– diencephalic junction, near the ventral thalamic– hypothalamic border (arrowhead B9). In the lateral part of ventral thalamus (vTh), DiI-labeled axons and retrogradely labeled neurons are apparent (representative neurons indicated with arrows A9 to C9). In the dorsal part of vTh, the TCAs project through retrogradely labeled neurons (black arrowhead in C9). TCAs form a tight fascicle on the surface of the ventral part of the vTh (white arrowhead in C9), lateral to retrogradely labeled neurons. Scale bars: A9 inset, 25 mm; C9, 500 mm. ic, internal capsule.

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

436

Braisted, Tuttle, and O’Leary

FIG. 4. Dorsal thalamic axon growth is influenced in vitro by chemoattractant and chemorepellent activities released by tissues along the thalamocortical axon pathway. (A) Explants of dorsal thalamus (dTh) were cultured for 1.5 days in collagen gels either alone or at a distance from the lateral surface of hypothalamus (dTh–Hy) or the medial surface of ventral telencephalon (dTh–vTel). Hy and vTel were dissected from living sections of E13.5–E14 diencephalon; dTh explants were dissected from sections caudal to the level illustrated. Darkly shaded area indicates thalamocortical axon projection. (B) For analysis of axon outgrowth, the area surrounding the dTh explant was divided into four quadrants relative to the test explant (dashed lines). Dorsal thalamic axon outgrowth from the quadrants facing toward (T) and away from (A) the test tissue were compared (for details, see Materials and Methods). (C) dTh cultured alone exhibited unbiased axon outgrowth (left side of the dish was designated away, and the right side toward). However, axon outgrowth from dTh was directed away from Hy (D) and toward vTel (E). (F) Quantification of dTh axon responses. DTh axon outgrowth was significantly different when cultured alone compared to either dTh–Hy or dTh–vTel cocultures (P , 0.02 and 0.04, respectively). DTh axon outgrowth in dTh–Hy cocultures was also significantly different from that in dTh–vTel cocultures (P , 0.0001). N values are in parentheses. (Ctx, neocortex; vTh, ventral thalamus.) Scale bar, 200 mm.

Potential Role for a Ventral Telencephalic Axonal Scaffold in Thalamocortical Axon Pathfinding We show that ventral telencephalic neurons, probably belonging to the globus pallidus, project axons into the dorsal thalamus before, or near, the time at which the first thalamocortical axons reach the ventral telencephalon. We also show that this ventral telencephalic axonal projection appears to coincide spatially with the oppositely extending thalamocortical axons. A previous study by Metin and Godement (1996) similarly showed that neurons in the ventral telencephalon of E11.5–12.5 hamster are retrogradely labeled from dorsal thalamus. They suggested that these labeled neurons may be perireticular cells, similar to those previously reported to be retrogradely labeled from dorsal thalamus in postnatal, but not embryonic, rats (Mitrofanis and Baker, 1993) and ferrets (Mitrofanis, 1994a,b). However, the recent study of Molnar et al. (1998) describes ventral telencephalic cells positioned within the nascent internal capsule that can be retrogradely labeled from dorsal thalamus in E14 rats, which is developmentally

equivalent to E12.5 mouse, the earliest age that we have examined. There are a number of differences between our findings in mouse and those of Me´tin and Godement (1996) in hamster and Molnar et al. (1998) in rat. We find a substantially greater number of cells in the ventral telencephalon retrogradely labeled from dorsal thalamus, and these cells have a broader mediolateral distribution extending to the diencephalic border. Further, these neurons are likely to be part of the globus pallidus (see Tuttle et al., 1999). Nonetheless, the spatial and temporal features of the axonal projection of the ventral telencephalic neurons described by all three studies (present study, mouse; Me´tin and Godement (1996), hamster; Molnar et al. (1998), rat), are consistent with the idea that this axonal projection, and possibly in addition the cell bodies themselves, may act as a scaffold to guide thalamocortical axons through the developing ventral thalamus and the medial part of the ventral telencephalon. Consistent with this suggestion, in mice deficient for Mash-1, this population of ventral telencephalic cells ap-

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

437

Thalamocortical Axon Pathfinding

FIG. 5. Schematic of a coronal section showing the early pathway of thalamocortical axons from dorsal thalamus (dTh) into ventral telencephalon (vTel), relative to the localization of potential axon guidance cues identified in this study. Axons (TCAs; red lines) of dTh neurons (red dots) extend ventrally and laterally through ventral thalamus (vTh). Near the border between vTh and hypothalamus (Hy), TCAs make a sharp, lateral turn and extend into the vTel. Axons (blue lines) from neurons (blue dots) in vTel project into dTh; their trajectory appears to coincide with the TCA trajectory. The axons of the vTel neurons may function as a scaffold to direct TCAs through vTh and medial vTel. In vitro studies indicate that the Hy releases a chemorepellent (purple 2) and the vTel releases a chemoattractant (green 1) for TCAs. We propose that these guidance activities may promote the turning of TCAs away from the Hy and into the vTel.

pears to be missing and thalamocortical axons fail to extend into the ventral telencephalon (Tuttle et al., 1999).

Ventral Telencephalic Chemoattractant Activity Our in vitro data suggest that at least a part of the medial aspect of the ventral telencephalon releases a diffusible attractant activity for thalamocortical axons. We propose that in vivo this ventral telencephalic chemoattractant, together with the hypothalamic chemorepellent, promotes the turning of thalamocortical axons away from hypothalamus and into the ventral telencephalon. The present study provides the first evidence that the ventral telencephalon produces a chemoattractant for dorsal thalamic axons. Previous studies have reported that the neocortex releases a soluble trophic activity that stimulates the outgrowth of dorsal thalamic axons in vitro (Blakemore and Molnar, 1990; Rennie et al., 1994; Lotto and Price, 1995). In these studies, axon outgrowth was enhanced from all sides of the

thalamic explant, but was often densest from the side facing the cortical explant, presumably because the concentration of the activity was greater on that side even though twodimensional substrates were used rather than collagen gels. Molnar and Blakemore (1995) have suggested that this cortical-derived trophic activity might diffuse long distances not only to promote the initiation of dorsal thalamic axon extension, but also to provide a directional cue for the growth of these axons toward the nascent internal capsule (i.e., ventral telencephalon). Netrin-1, a well-characterized chemoattractant (Serafini et al., 1994; Kennedy et al., 1994), is a potential candidate for the ventral telencephalic chemoattractant activity. Netrin-1 is expressed in the ventral telencephalon at the time thalamocortical axons are navigating through it (Serafini et al., 1996; Me´tin et al., 1997) and has been shown in vitro to act as a chemoattractant for axonal populations that project toward the ventral midline (Kennedy et al., 1994; Shirasaki et al., 1995, 1996), as well as for retinal (Deiner et al., 1997) and cortical axons (Me´tin et al., 1997; Richards et al., 1997). In collagen gels, embryonic rodent cortical explants show biased axon outgrowth toward tissues expressing netrin-1, including ventral telencephalon and floor plate, as well as netrin-1-transfected cells (Me´tin et al., 1997; Richards et al., 1997). Analyses of netrin-1-deficient mice (Serafini et al., 1996; Deiner et al., 1997), and mice lacking DCC (Fazeli et al., 1997; Deiner et al., 1997), a netrin-1 receptor that mediates its attractant effects (KeinoMasu et al., 1996), indicate that netrin-1 is required for the proper pathfinding of a number of axonal populations in vivo. However, DCC is expressed at only low levels in a restricted part of dorsal thalamus (Gad et al., 1997), and preliminary analyses of netrin-1- and DCC-deficient mice indicate that thalamocortical axons turn properly as they approach the hypothalamus and extend into the ventral telencephalon (Braisted, Catalano, Stimac, TessierLavigne, Shatz, and O’Leary, unpublished observations). Defects are apparent in these mice, but they occur at points more distal in the thalamocortical axon pathway than the pathfinding decision point addressed in the current study. For example, the path of thalamocortical axons through the ventral telencephalon is narrower than normal. Thus, if netrin-1 can act as a ventral telencephalic chemoattractant for thalamocortical axons, it must be only one of multiple guidance cues for thalamocortical axons and is not required for their proper turning away from the hypothalamus and into the ventral telencephalon.

Hypothalamic Chemorepellent Activity Our in vitro data suggest that the hypothalamus releases a diffusible activity that can repel thalamocortical axons at a distance. We suggest that this repellent activity promotes the sharp turning of thalamocortical axons as they approach the hypothalamus and, by acting in concert with other cues, promotes their extension laterally into ventral telencephalon. If the hypothalamic activity is diffusible, why do

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

438

Braisted, Tuttle, and O’Leary

thalamocortical axons initially extend toward the source of the repellent activity in vivo, growing up the putative repellent gradient emanating from the hypothalamus, before turning to grow into the ventral telencephalon? One possibility is that the repellent gradient is effective only very near the hypothalamus, because it either can diffuse only a short distance in vivo (or possibly not at all) or is present at an effective concentration only near its source. Alternatively, thalamocortical axons may approach the hypothalamus before an effective concentration gradient is established, or the response of thalamocortical axons to the repellent activity may have a late onset or require prolonged exposure. Another possibility is that the repellent activity by itself is insufficient to cause the turning of thalamocortical axons in vivo, and additional cues, such as the attractant activity or axonal scaffold originating in ventral telencephalon, are required to complement it. A previous study has demonstrated that the rat and mouse hypothalami are repulsive for retinal axons (Tuttle et al., 1998). In that study, we suggested that the repellent activity prevents retinal axons from invading the hypothalamus and, in conjunction with other cues, causes them to course over its surface in a tight fascicle. In the present study, we suggest that the repellent activity helps direct thalamocortical axon growth away from the hypothalamus. This difference in axon behavior is likely due to the presence of other cues that in sum lead to different behaviors exhibited by retinal and thalamic axons, but could also in part be due to different sensitivities of these axons to the repellent activity, or, alternatively, that the repellent activities defined for retinal and thalamic axons are due to two distinct molecules with different distributions. Semaphorin III/D (semaIII/D) and netrin-1 have been characterized as axonal chemorepellents (Tessier-Lavigne and Goodman, 1996; Mark et al., 1997). The ephrins have also been shown to be axon repellents (O’Leary and Wilkinson, 1999), but they require cell attachment for function and cannot act in a soluble form unless artificially clustered (Davis et al., 1994)—thus they cannot account for the soluble hypothalamic repellent activity that we have identified. Netrin-1, a bifunctional guidance molecule (Colamarino and Tessier-Lavigne, 1995), is an unlikely candidate for the hypothalamic repellent acting on thalamocortical axons on the basis of its expression pattern (Serafini et al., 1996). On the other hand, expression data and functional studies suggest that semaIII/D is a potential candidate for the hypothalamic chemorepellent. Although expression data are limited, semaIII/D is reported to be expressed in the developing hypothalamus (Zhou et al., 1997), and preliminary findings indicate that semaIII/D can collapse growth cones of dorsal thalamic axons (Bagnard et al., 1996). However, neuropilin-1 and -2, receptors for semaIII/D (He and Tessier-Lavigne, 1997; Kolodkin et al., 1997), do not appear to be present in embryonic dorsal thalamus (Kawakami et al., 1996; Chen et al., 1997), although it is possible that other semaIII/D receptors have yet to be identified (Feiner et al., 1997; Koppel et al., 1997). If

semaIII/D is a hypothalamic repellent, it must be only a nonessential contributor to the turning of thalamocortical axons as they approach the hypothalamus since analyses of mice deficient for semaIII/D (Behar et al., 1996) indicate that thalamocortical axon pathfinding is normal in its absence (Catalano et al., 1998).

Conclusion In conclusion, we have identified a chemorepellent activity released by the hypothalamus and a chemoattractant activity released by the ventral telencephalon that influence thalamocortical axon growth in vitro. These activities are localized to a critical juncture in the pathway of thalamocortical axons, where they make an abrupt turn away from the hypothalamus to extend laterally into the ventral telencephalon. It will be of importance to undertake future studies to identify the molecules responsible for these activities, which likely cooperate to turn thalamocortical axons. In addition, our findings suggest that ventral telencephalic neurons, and their axons that extend into the dorsal thalamus, may form a scaffold that helps to guide thalamocortical axons over this portion of their pathway through a contact-mediated mechanism. A definitive test of the role of this potential scaffold in thalamocortical axon guidance may pose a significant challenge.

ACKNOWLEDGMENTS This work was supported by NIH Grant RO1 NS31558 (D.D.M.O’L.) and NIH Fellowship F32 EY06769 (J.E.B.). We thank John Burrill for comments on the manuscript.

REFERENCES Angevine, J. B., Jr. (1970). Time of neuron origin in the diencephalon of the mouse. An autoradiographic study. J. Comp. Neurol. 139, 129 –187. Bagnard, D., Betz, H., Gotz, M. A., Puschel, W., and Bolz, J. (1996). Do semaphorins act as guidance molecules for afferent and efferent projections in the developing cerebral cortex? Soc. Neurosc. Abst. 22, 1473. Behar, O., Golden, J. A., Mashimo, H., Schoen, F. J., and Fishman, M. C. (1996). Semaphorin III is needed for normal patterning and growth of nerves, bones and heart. Nature 383, 525–528. Bicknese, A. R., Sheppard, A. M., O’Leary, D. D. M., and Pearlman, A. L. (1994). Thalamocortical axons extend along a chondroitin sulfate proteoglycan-enriched pathway coincident with the neocortical subplate and distinct from the efferent path. J. Neurosci. 14, 3500 –3510. Blakemore, C., and Molnar, Z. (1990). Factors involved in the establishment of specific interconnections between thalamus and cerebral cortex. Cold Spring Harbor Symp. Quant. Biol. 55, 491–504. Butler, H., and Juurlink, B. H. J. (1987). ”An Atlas for Staging Mammalian and Chick Embryos.“ CRC Press, Boca Raton, FL. Catalano, S. M., Messersmith, E. K., Goodman, C. S., Shatz, C. J., and Chedotal, A. (1998). Many major CNS axon projections

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

439

Thalamocortical Axon Pathfinding

develop normally in the absence of semaphorin III. Mol. Cell. Neurosci. 11, 173–182. Chen, H., Chedotal, A., He, Z., Goodman, C. S., and TessierLavigne, M. (1997). Neuropilin-2, a novel member of the neuropilin family, is a high affinity receptor for the semaphorins Sema E and Sema IV but not Sema III. Neuron 19, 547–559. Colamarino, S. A., and Tessier-Lavigne, M. (1995). The axonal chemoattractant netrin-1 is also a chemorepellent for trochlear motor axons. Cell 81, 621– 629. Davis, S., Gale, N. W., Aldrich, T. H., Maisonpierre, P. C., Lhotak, V., Pawson, T. Goldfarb, M., and Yancopoulos, G. D. (1994). Ligands for EPH-related receptor tyrosine kinases that require membrane attachment or clustering for activity. Science 266, 816 – 819. Deiner, M. S., Kennedy, T. E., Fazeli, A., Serafini, T., TessierLavigne, M., and Sretavan, D. W. (1997). Netrin-1 and DCC mediate axon guidance locally at the optic disc: Loss of function leads to optic nerve hypoplasia. Neuron 19, 575–589. Fazeli, A., Dickinson, S. L., Hermiston, M. L., Tighe, R. V., Steen, R. G., Small, C. G., Stoeckli, E. T., Keino-Masu, K., Masu, M., Rayburn, H., Simons, J., Bronson, R. T., Gordon, J. I., TessierLavigne, M., and Weinberg, R. A. (1997). Phenotype of mice lacking functional Deleted in colorectal cancer (DCC) gene. Nature 386, 796 – 810. Feiner, L., Koppel, A. M., Kobayashi, H., and Raper, J. A. (1997). Secreted chick semaphorins bind recombinant neuropilin with similar affinities but bind different subsets of neurons in situ. Neuron 19, 539 –545. Gad, J. M., Keeling, S. L., Wilks, A. F., Tan, S.-S., and Cooper, H. M. (1997). The expression patterns of guidance receptors, DCC and Neogenin, are spatially and temporally distinct throughout mouse embryogenesis. Dev. Biol. 192, 258 –273. Goodman, C. S. (1996). Mechanisms and molecules that control growth cone guidance. Annu. Rev. Neurosci. 19, 341–377. Groenewegen, H. J. (1988). Organization of the afferent connections of the mediodorsal thalamic nucleus in the rat, related to the mediodorsal–prefrontal topography. Neuroscience 24, 379 – 431. He, Z., and Tessier-Lavigne, M. (1997). Neuropilin is a receptor for the axonal chemorepellent Semaphorin III. Cell 90, 739 –751. Honig, M. G., and Hume, R. I. (1989). Dil and DiO: Versatile fluorescent dyes for neuronal labelling and pathway tracing [see comments]. Trends Neurosci. 12, 333–335, 340 –341. Jones, E. G. (1985). ”The Thalamus.“ Plenum, New York. Kawakami, A., Kitsukawa, T., Takagi, S., and Fujisawa, H. (1996). Developmentally regulated expression of a cell surface protein, neuropilin, in the mouse nervous system. J. Neurobiol. 29, 1–17. Keinu-Masu, K., Masu, M., Hinck, L., Leonardo, E. D., Chan, S. S.-Y., Culotti, J. G., and Tessier-Lavigne, M. (1996). Deleted in Colorectal Cancer (DCC) encodes a netrin receptor. Cell 87, 175–185. Kennedy, T. E., Serafini, T., de la Torre, J. R., and Tessier-Lavigne, M. (1994). Netrins are diffusible chemotrophic factors for commissural axons in the embryonic spinal cord. Cell 78, 425– 435. Kohtz, J. D., Baker, D. P., Corte, G., and Fishell, G. (1998). Regionalization within the mammalian telencephalon is mediated by changes in responsiveness to Sonic Hedgehog. Development 125, 5079 –5089. Kolodkin, A. L., Levengood, D. V., Rowe, E. G., Tai, Y. T., Giger, R. J., and Ginty, D. D. (1997). Neuropilin is a semaphorin III receptor. Cell 90, 753–762. Koppel, A. M., Feiner, L., Kobayashi, H., and Raper, J. A. (1997). A 70 amino acid region within the semaphorin domain activates

specific cellular response of semaphorin family members. Neuron 19, 531–537. Lotto, R. B., and Price, D. J. (1995). The stimulation of thalamic neurite outgrowth by cortex-derived growth factors in vitro: The influence of cortical age and activity. Eur. J. Neurosci. 7, 318 – 328. Mark, M. D., Lohrum, M., and Puschel, A. W. (1997). Patterning neuronal connections by chemorepulsion: The semaphorins. Cell Tissue Res. 290, 299 –306. Me´tin, C., Dele´glise, D., Serafini, T., Kennedy, T. E., and TessierLavigne, M. (1997). A role for netrin-1 in the guidance of cortical efferents. Development 124, 5063–5074. Me´tin, C., and Godement, P. (1996). The ganglionic eminence may be an intermediate target for corticofugal and thalamocortical axons. J. Neurosci. 16, 3219 –3235. Mitrofanis, J. (1994). Development of the pathway from the reticular and perireticular nuclei to the thalamus in ferrets: A DiI study. Eur. J. Neurosci. 6, 1864 –1882. Mitrofanis, J., and Baker, G. E. (1993). Development of the thalamic reticular and perireticular nuclei in rats and their relationship to the course of growing corticofugal and corticopetal axons. J. Comp. Neurol. 338, 575–587. Molnar, Z., and Blakemore, C. (1995). How do thalamic axons find their way to the cortex? Trends Neurosci. 18, 389 –397. Molnar, Z., Adams, R., and Blakemore, C. (1998). Mechanisms underlying the early establishment of thalamocortical connections in the rat. J. Neurosci. 18, 5723–5745. O’Leary, D. D. M., and Wilkinson, D. G. (1999). Eph receptors and ephrins in neural development. Curr. Opin. Neurobiol. 9, 65–73. Olsson, M., Campbell, K., and Turnbull, D. H. (1997). Specification of mouse telencephalic and mid-hindbrain progenitors following heterotopic ultrasound-guided embryonic transplantation. Neuron 19, 761–772. Paxinos, G. (1995). ”The Rat Nervous System.“ Academic Press, San Diego. Pini, A. (1993). Chemorepulsion of axons in the developing mammalian central nervous system. Science 261, 95–98. Puschel, A. W., Adams, R. H., and Betz, H. (1996). The sensory innervation of the mouse spinal cord may be patterned by differential expression of and differential responsiveness to semaphorins. Mol. Cell. Neurosci. 7, 419 – 431. Rennie, S., Lotto, R. B., and Price, D. J. (1994). Growth-promoting interactions between the murine neocortex and thalamus in organotypic co-cultures. Neuroscience 61, 547–564. Ricardo, J. A. (1981). Efferent connections of the subthalamic region in the rat. II. The zona incerta. Brain Res. 214, 43– 60. Richards, L. J., Koester, S. E., Tuttle, R., and O’Leary, D. D. M. (1997). Directed growth of early cortical axons is influenced by a chemoattractant released from an intermediate target. J. Neurosci. 17, 2445–2458. Serafini, T., Kennedy, T. E., Galko, M. J., Mirzayan, C., Jessell, T. M., and Tessier-Lavigne, T. M. (1994). The netrins define a family of axon outgrowth-promoting proteins homologous to C. elegans UNC-6. Cell 78, 409 – 424. Serafini, T., Colamarino, S. A., Leonardo, E. D., Wang, H., Beddington, R., Skarnes, W. C., and Tessier-Lavigne, M. (1996). Netrin-1 is required for commissural axon guidance in the developing vertebrate nervous system. Cell 87, 1001–1014. Shimamura, K., Hartigan, D. J., Martinez, S., Puelles, L., and Rubenstein, J. L. (1995). Longitudinal organization of the anterior neural plate and neural tube. Development 121, 3923–3933.

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.

440

Braisted, Tuttle, and O’Leary

Shirasaki, R., Mirzayan, C., Tessier-Lavigne, M., and Murakami, F. (1996). Guidance of circumferentially growing axons by netrindependent and -independent floor plate chemotropism in the vertebrate brain. Neuron 17, 1079 –1088. Shirasaki, R., Tamada, A., Katsumata, R., and Murakami, F. (1995). Guidance of cerebellofugal axons in the rat embryo: Directed growth toward the floor plate and subsequent elongation along the longitudinal axis. Neuron 14, 961–972. Taniguchi, M., Yuasa, S., Fujisawa, H., Naruse, I., Saga, S., Mishina, M., and Yagi, T. (1997). Disruption of semaphorin III/D gene causes severe abnormality in peripheral nerve projection. Neuron 19, 519 –530. Tessier-Lavigne, M., and Goodman, C. S. (1996). The molecular biology of axon guidance. Science 274, 1123–1133. Tuttle, R., Braisted, J. E., Richards, L. J., and O’Leary, D. D. M. (1998). Retinal axon guidance by region-specific cues in diencephalon. Development 125, 791– 801.

Tuttle, R., Nakagawa, Y., and O’Leary, D. D. M. (1999). Defects in thalamocortical axon pathfinding correlate with altered cell domains in embryonic Mash-1 deficient mice. Development, in press. Young, W. S. d., Alheid, G. F., and Heimer, L. (1984). The ventral pallidal projection to the mediodorsal thalamus: A study with fluorescent retrograde tracers and immunohistofluorescence. J. Neurosci. 4, 1626 –1638. Zhou, L., White, F. A., Lentz, S. I., Wright, D. E., Fisher, D. A., and Snider, W. D. (1997). Cloning and expression of a novel murine semaphorin with structural similarity to insect semaphorin I. Mol. Cell. Neurosci. 9, 26 – 41. Received for publication November 17, 1998 Revised January 14, 1999 Accepted January 20, 1999

Copyright © 1999 by Academic Press. All rights of reproduction in any form reserved.