Netrin signal is produced in leech embryos by segmentally iterated sets of central neurons and longitudinal muscle cells.

Abstract. Netrins are secreted molecules capable of attracting or repelling growing axons. They and their receptors, along with other netrin-interacting proteins, are widely conserved among animals from a broad range of phyla. We have raised and purified an antibody against a recently cloned leech netrin, which has allowed us to characterize embryonic netrin expression by cells in peripheral tissues and in the central nervous system. During early gangliogenesis, netrin expression was detected at particularly high levels in five bilateral pairs of central neurons. Towards the end of the period of axonal outgrowth, netrin expression was observed to be restricted to only six central neurons, comprising two bilateral pairs and two unpaired cells. A pair of netrin-producing central neurons, the bipolar cells, was identified by their expression of the antigen recognized by the monoclonal antibody Laz1-1. Double staining of sensory afferents from segmental sensilla with the monoclonal antibody Lan3-2 and the bipolar cells with the netrin antibody revealed that the terminals of these afferents grow up to the bipolar cells and turn anteriorly or posteriorly, without extending any further medially. Peripheral netrin expression was found to be restricted to longitudinal muscle cells in the ventral half of the body wall. Extracellular, secreted netrin was detected in a broad longitudinal stripe located symmetrically with respect to the ventral midline. The pattern of expression of netrin in leech embryos is consistent with observed expression patterns in other animals, suggesting that developmental netrin functions are conserved among all bilateral animals.

Possible role of the receptor protein tyrosine phosphatase HmLAR2 in interbranch repulsion in a leech embryonic cell.

Accumulating evidence indicates that receptor protein tyrosine phosphatases (rPTPs) play major roles in growth cone migration. We have previously shown that the growth cones of the multiple parallel processes of an identified leech embryonic cell, the Comb cell (CC), express high levels of a leukocyte antigen-related (LAR)-like rPTP, HmLAR2. Embryonic injection of a polyclonal antibody to the receptor's ectodomain resulted in reduced process outgrowth and in processes crossing over each other, a behavior that is seldom observed in normal or control animals. Here we present results of injecting a soluble Fc-HmLAR2 ectodomain fusion protein into embryos in order to bind the endogenous ligands of HmLAR2. Single injections of the Fc-chimeric protein into the developing embryo resulted, 12 to 24 h postinjection, in clear morphological abnormalities, ranging from abnormally directed CC processes and crossovers to apparent growth cone collapse. At later times, 2 to 5 days post injection, growth cones appeared to have recovered and processes had continued to extend, but effects of the earlier guidance errors remained, with the CCs displaying a relatively high incidence of proximal guidance errors. When injected into the germinal plate of developing embryos, the fusion protein was found to bind selectively to the processes of the CCs themselves, in contrast to control injections of Fc alone or closely related Fc-tagged proteins, which did not decorate the CCs. Double-labeling experiments revealed an early phase of Fc-HmLAR2 labeling (within 20 min after application), during which the growth cones and filopodia of the CC showed significant binding of the receptor ectodomain, and a later phase (1-2 h after injection), when most of the label was redistributed away from the growth cones and into the proximal processes of the CC. In culture, HmLAR2-transfected COS cells were found to selectively bind the Fc-recombinant protein, but not Fc-tagged proteins bearing other closely related receptor ectodomains, demonstrating that the HmLAR2 ectodomain is capable of interacting homophilically. Together, our observations demonstrate that the rPTP HmLAR2 is critically involved in CC process extension through its participation in the regulation of growth cone structure, migration, and navigation. Moreover, since our experiments also indicate that HmLAR2 can bind to itself, we hypothesize that HmLAR2 has a key role in the mechanism of mutual repulsion that maintains the parallel growth of adjacent CC projections.

RNAi of the receptor tyrosine phosphatase HmLAR2 in a single cell of an intact leech embryo leads to growth-cone collapse.

Receptor protein tyrosine phosphatases (RPTPs) are important for growth-cone migration [1-5], but their specific roles have yet to be defined. Previously, we showed that the growth cones of the Comb cell, an embryonic cell in the leech, express high levels of an RPTP called HmLAR2 [6,7]. Here, we report the use of RNA interference (RNAi) to block expression of HmLAR2 in individual Comb cells in the developing embryo. HmLAR2 mRNA levels were reduced in the soma, processes and growth cones of Comb cells injected with double-stranded RNA (dsRNA) for HmLAR2, but no decrease was detected when control dsRNAs were injected. Consistent with this observation, the level of phosphotyrosine increased significantly in the growth cones of Comb cells injected with HmLAR2 dsRNA. Within 24 hours, the growth cones of treated cells showed a distinct collapsed phenotype, with sharp reductions in lamellipodial surface area and in numbers of filopodia. These experiments indicate a key role for LAR-like RPTPs in maintaining the integrity of the growth cone.

The role of a LAR-like receptor tyrosine phosphatase in growth cone collapse and mutual-avoidance by sibling processes.

Among the many cells or parts of cells that a growth cone may encounter during its embryonic migrations are other processes or parts of its parent cell. Such an event can be expected to be relatively frequent in the genesis of neuronal arbors, for instance, where the density of innervation of a target region can be quite high. Few experimental studies have addressed the very interesting question of whether a process "recognizes" siblings in some unique way, in a manner that can be distinguished from, say, how it interacts with unrelated cells. One example can be found in the leech, where sibling branches in the terminal fields of identified mechanosensory cells avoid each other strictly while permitting some significant continuing contact and overlap with homologues, a phenomenon that has been dubbed "self-avoidance." Another example has been reported in cultured Helisoma neurons, where severing a branch of a neuron allows sibling neurites to form electrical junctions with it, although normally sibling neurites do not do so. In both of these instances, coincidental activity was proposed as one means to achieve recognition of self and as possibly leading to the blocking of a continuing interaction among the parts, although alternative explanations were indeed considered possible.

Cellular expression of a leech netrin suggests roles in the formation of longitudinal nerve tracts and in regional innervation of peripheral targets.

Netrins are secreted, diffusible proteins that direct axonal growth. To study the functions of netrins in the relatively simple and easily accessible nervous system of the leech Hirudo medicinalis, we have cloned a leech netrin and have characterized its expression during embryogenesis. By probing a leech cDNA library at low stringency with chick netrin probes, we have identified a complete cDNA clone that bears significant sequence similarity to netrins of other species. In situ hybridization and dye filling of individual neurons show that this leech netrin is expressed by several identifiable central neurons in every segmental ganglionic primordium during early stages of embryogenesis. Some of these neurons, including the bipolar cells which are thought to be involved in setting up longitudinal tracts, express this gene only transiently during embryogenesis, while others continue to express it in the adult. In addition, leech netrin is expressed by ventral but not dorsal longitudinal muscle cells in each segment before central neurons project their axons to the periphery. These highly specific expression patterns are consistent with the hypothesis that leech netrin plays a role in forming the major interganglionic neuronal tracts and in defining ventral versus dorsal domains of peripheral innervation.

Lox6, a leech Dfd ortholog, is expressed in the central nervous system and in peripheral sensory structures.

Sequence analysis of a newly isolated Hirudo medicinalis cDNA containing an Antennapedia (Antp)-class homeobox suggests that the corresponding gene, Lox6, is an ortholog of the Drosophila Deformed (Dfd) gene. In situ hybridization of whole-mounted preparations shows that the major sites of Lox6 expression during embryogenesis are the central nervous system (CNS) and the peripheral sensory system. Lox6 mRNA can be detected in a subset of neurons in each ganglion from the subesophageal ganglion (RG2) to the most posterior ganglion, with the highest level of expression seen in RG3. Peripherally, Lox6 is expressed principally in the primordia of the sensillae and in the eyes. This pattern of expression of Lox6 suggests that one of its functions may be to contribute to the diversification of neuronal phenotypes.

The leech receptor protein tyrosine phosphatase HmLAR2 is concentrated in growth cones and is involved in process outgrowth.

Developing neurons extend long processes to specific distal targets using extracellular molecules as guidance cues to navigate through the embryo. Growth cones, specialized structures at the tip of the extending processes, are thought to accomplish this navigation through receptors that recognize guidance cues and modulate growth accordingly. In Drosophila, several receptor tyrosine phosphatases (rPTPs), including DLAR, have been shown to participate in directing neurite outgrowth. As yet, however, it is not known how rPTPs act to affect navigation. To gain insight into the mechanisms of rPTP-mediated outgrowth guidance, we have investigated the role of HmLAR2, a Hirudo medicinalis homologue of DLAR, in process outgrowth. HmLAR2 is expressed by, among other cells, a transient neuron-like template cell, the Comb cell. Here we present evidence that HmLAR2 protein becomes concentrated within their growth cones at a stage when C cell processes undergo rapid outgrowth. When antibodies raised against the extracellular domain of HmLAR2 were injected into intact embryos, they bound specifically to the C cell surface at growth cones and along processes and caused the partial internalization of HmLAR2 receptors. Moreover, the C cell processes were found to project aberrantly, to deviate from their normally highly regular trajectories and to extend shorter distances in the presence of the antibodies. We propose that HmLAR2 is required by the C cell for guidance and extension and suggest that it functions via its ectodomain to transduce extracellular guidance cues.

A detached branch stops being recognized as self by other branches of a neuron.

The multiple peripheral projections of a single leech mechanosensory neuron form individual arbors that do not overlap at all with each other, a phenomenon that has been termed "self-avoidance" (Yau, 1976; Kramer and Stent, 1985). This is in marked contrast to the peripheral arbors of adjacent segmental homologues, which partially overlap with each other at their boundaries in target areas of the body wall (Nicholls and Baylor, 1968; Gan and Macagno, 1995). How a neurite differentiates between sibling neurites of the same cell and those of a homologue is not known, but possible mechanisms include the recognition of surface markers of neuronal identity or the detection of cell-specific patterns of activity. In order to test whether this self-recognition requires a neurite to be in direct communication with its soma, we used a laser microbeam to sever a branch of a dye-filled pressure-sensitive (P) neuron in an intact leech embryo. Time-lapse observations of the P cell arbor in the living, unanesthetized, animal for up to 24 h following the surgery showed that the detached branch continued to show dynamic growth behavior throughout the period of observation. However, the detached branch ceased being avoided by the rest of the cell within a few hours, other, attached branches of the neuron overgrowing its territory and directly overlapping with it. Our experiments provide direct evidence for the existence of strong growth-inhibiting interactions between sibling processes, and indicate that self-avoidance by the growing neurites of a cell requires physical continuity between these neurites.

Two receptor tyrosine phosphatases of the LAR family are expressed in the developing leech by specific central neurons as well as select peripheral neurons, muscles, and other cells.

Receptor protein tyrosine phosphatases (rPTPs) are thought to play a crucial role in neuronal development, particularly in pathfinding by growing processes. We have cloned and sequenced two Hirudo medicinalis rPTPs that are homologous to the Drosophila and vertebrate rPTPs of the Leukocyte common antigen-related (LAR) subfamily. These Hirudo rPTPs, HmLAR1 and HmLAR2, are products of different, homologous genes, both containing two tandem intracellular phosphatase domains and ectodomains with three tandem Ig domains and different numbers of tandem fibronectin type III (FIII) domains. They are expressed in distinct patterns during embryogenesis. HmLAR1 mRNA is expressed by a subset of central and peripheral neurons and by several peripheral muscular structures, whereas HmLAR2 mRNA is expressed by a different subset of central neurons and by the peripheral, neuron-like Comb cells. HmLAR1 and HmLAR2 proteins are located on the neurites of central neurons. In addition, HmLAR2 is expressed on the cell body, processes, and growth cones of the Comb cells. Because of their CAM-like ectodomains and homology to proteins known to be involved in pathfinding and because they are expressed by different subsets of neurons, we hypothesize that HmLAR1 and HmLAR2 participate in navigational decisions that distinguish the sets of neurons that express them. Furthermore, we hypothesize that HmLAR2 is also involved in setting up the highly regular array of parallel processes established by the Comb cells. Lastly, we propose that the HmLAR1 ectodomain on peripheral muscle cells plays a role in target recognition via interactions with neuronal receptors, which might include HmLAR1 or HmLAR2.

Cell proliferation in a peripheral target is required for the induction of central neurogenesis in the leech.

Several days after the completion of the early phase of cell proliferation that generates most of the leech central nervous system, the pair of "sex ganglia" in the two reproductive segments of the midbody undergo a second period of neurogenesis that gives rise to several hundred peripherally induced central (PIC) neurons. This proliferative phase, which begins on embryonic day 17 (E17), is induced by the interaction of a few specific neurons in the sex ganglia with a peripheral target, the male genitalia, during a critical period that extends from E13 to E16. The central nervous system (CNS) determines the critical period, since the male genitalia have the capacity to induce PIC neurons beginning on E10 and continuing throughout embryogenesis. Here we first show, by injecting hydroxyurea into staged embryos to ablate dividing cells, that PIC neuron precursors begin to divide at a low rate before E17, during the critical period. Then, through a series of homochronic and heterochronic male organ transplantations combined with hydroxyurea treatment of hosts and/or donors, we show that cell proliferation is required in the target itself for it to be competent to induce PIC neurons. These observations demonstrate that a nerve connection can couple cell proliferation in a peripheral target to cell proliferation in the CNS, providing a novel means for size adjustment of a central neuronal population relative to a peripheral target.

New electrical properties of neurons induced by a homeoprotein.

To explore possible neurogenic functions of the genes of the Hox/HOM complexes, we injected the mRNA from the leech homeobox genes Lox1 and Lox4 into adult neurons that normally do not express them. The ectopic expression of Lox1 induced a specific transformation in the electrical properties of certain identified neurons: action potential amplitude increased about threefold after the injections. This effect of Lox1 expression was restricted, among cell types examined, to the anterior pagoda neurons (APs) and the nut neurons. This effect was also restricted to Lox1 ectopic expression; the action potentials of APs and nut neurons were not enlarged when the mRNAs of either Lox4, another leech Hox/HOM gene, or beta-galactosidase were injected. Lox1 mRNA injection did not affect the resting potential, input resistance, or axonal morphology of the transformed APs, raising the possibility that it acts via the modification of voltage-dependent ion channels. Thus, a specific homeobox gene can transform key neuronal characteristics in a cell-specific manner. We may thus add electrophysiologic properties to other aspects of neuronal identity determined by homeobox gene expression.

Competition among the axonal projections of an identified neuron contributes to the retraction of some of those projections.

AP neurons in the embryonic leech CNS extend lateral projections to peripheral targets through the ganglionic nerve roots and longitudinal projections toward neighboring ganglia through the connective nerves. The lateral projections grow extensively in the periphery; in contrast, the longitudinal projections achieve relatively little growth and eventually retract, the majority having essentially disappeared by the end of embryogenesis. Cutting both nerve roots, which eliminates both lateral projections, however, induces the longitudinal projections of the AP neuron to begin to grow rapidly toward adjacent ganglia within 14 hr after the axotomy. By using a laser microbeam to cut just the lateral projections of the AP cells, we further show that it is indeed the loss of its lateral projections, and not a secondary response to the cutting of other components of the root nerves, that induces the longitudinal projections of the AP cell to grow extensively. In addition, we report that reducing the outgrowth of the lateral projections by: (1) cutting only one lateral projection, or (2) ablating pioneer neurons required by the AP neuron to establish its peripheral arbor, also results in a significant increase in the growth of the longitudinal projections. Finally, we demonstrate that increasing the outgrowth of the longitudinal projections by ablating the AP cells in adjacent ganglia results in a significant reduction in the outgrowth of the lateral projections. Taken together, these results indicate, first, that the longitudinal and lateral projections usually grow at the expense of each other, and second, that normally the extensive outgrowth of its lateral projections is a necessary condition for a developing AP neuron to retract its longitudinal projections.

The establishment of peripheral sensory arbors in the leech: in vivo time-lapse studies reveal a highly dynamic process.

Pressure-sensitive (P) neurons located in the leech CNS form elaborate terminal arbors in the body wall of the animal during mid-embryogenesis. In the experiments discussed here, arbor development in the target region was studied in intact, unanesthetized leech embryos using time-lapse video microscopy of individual, fluorescently stained P neurons. Analysis of time-lapse recordings made over a period of several days revealed that arbor formation is a very dynamic process. At any particular time, most high-order terminal branches were either extending or retracting, in approximately equal numbers and at very similar rates. Many branches underwent several rounds of extension and retraction every hour. Net arbor growth occurred at a much lower rate than the extension and retraction rates of individual branches. Process retraction sometimes resulted in an apparent change in the topological order of processes. Significantly, the initiation of new branches was restricted to a few locations along the parent process, which were termed "hot spots." Moreover, the capacity to generate high-order branches correlated with parent process stability. The target region of the growing P cell arbor in the body wall was subsequently examined using confocal microscopy in fixed preparations. The arbor expanded between the longitudinal and circular muscle layers, a region occupied by small unidentified cells. Simultaneous imaging of the dye-labeled terminal arbor and the surrounding tissue at two different wavelengths suggested that the high-order processes were navigating around these cells, which sometimes forced the growing processes to assume a bent form. These observations suggest that the formation of the P cell arbor can be best described as a "dynamically unstable" process that is constrained by interactions with its environment.

Identified central neurons convey a mitogenic signal from a peripheral target to the CNS.

Regulation of central neurogenesis by a peripheral target has been previously demonstrated in the ventral nerve cord of the leech Hirudo medicinalis (Baptista, C. A., Gershon, T. R. and Macagno, E. R. (1990). Nature 346, 855-858) Specifically, innervation of the male genitalia by the fifth and sixth segmental ganglia (the sex ganglia) was shown to trigger the birth of several hundred central neurons (PIC neurons) in these ganglia. As reported here, removal of the target early during induction shows that PIC neurons can be independently induced in each side of a ganglion, indicating that the inductive signal is both highly localized and conveyed to each hemiganglion independently. Further, since recent observations (Becker, T., Berliner, A. J., Nitabach, M. N., Gan, W.-B. and Macagno, E. R. (1995). Development, 121, 359-369) had indicated that efferent projections are probably involved in this phenomenon, we individually ablated all possible candidates, which led to the identification of two central neurons that appear to play significant roles in conveying the inductive signal to the CNS. Ablation of a single ML neuron reduced cell proliferation in its own hemiganglion by nearly 50%, on the average. In contrast, proliferation on the opposite side of the ganglion increased by about 25%, suggesting the possibility of a compensatory response by the remaining contralateral ML neuron. Simultaneous ablation of both ML neurons in a sex ganglion caused similar reductions in cell proliferation in each hemiganglion. Deletion of a single AL neuron produced a weaker (7%) but nonetheless reproducible reduction. Ablation of the other nine central neurons that might have been involved in PIC neuron induction had no detectable effect. Both ML and AL neurons exhibit ipsilateral peripheral projections, and both arborize mostly in the hemiganglion where they reside. Thus, we conclude that peripheral regulation of central neurogenesis is mediated in the leech by inductive signals conveyed retrogradely to each hemiganglion by specific central neurons that innervate this target and the hemiganglion they affect.

A novel homeobox cluster expressed in repeated structures of the midgut.

A gene cluster (LOX3-C) containing three duplicated homeobox sequences (Lox3A, Lox3B, and Lox3C) was characterized in the leech Hirudo medicinalis. The leech homeoboxes have a limited homology to those of Antennapedia-class genes, but do not have homologs among currently characterized insect genes. The Lox3 genes belong to a new family, named Xlox, that also includes genes from mouse, rat, frog, and a distantly related leech, Helobdella triserialis. All members of the Xlox family are expressed in specific regions of the embryonic gut, where they seem to affect morphogenesis and cell differentiation. The three homeoboxes of LOX3-C described here are contained within nearly identical direct tandem repeats. The LOX3-C region produces at least two transcripts, one present in both embryos and adults and the other only in early embryos. Early Lox3 expression is restricted to specific regions of the midgut primordium, in 12 segmentally repeated, transverse stripes of fusiform cells found at the positions where the midgut will constrict to form the 11 diverticula of the crop. A role in the development of segmentally iterated structures in an initially homogeneous midgut is proposed for LOX3-C.

The leech homeobox gene Lox4 may determine segmental differentiation of identified neurons.

We cloned and characterized a new leech homeobox gene, Lox4, a homolog of the Drosophila genes Ultrabithorax and abdominal-A. Lox4 has a complex and dynamic pattern of expression within a series of segmentally homologous neurons. These include a pair of specialized motor neurons of one segmental ganglion, the rostral penile evertors (RPEs), and their segmental homologs in other midbody ganglia. During gangliogenesis, Lox4 was expressed within this series of neurons in three different temporal patterns: (1) it was never expressed in the RPE homologs of ganglia 1-3; (2) it was expressed in the RPEs during gangliogenesis, but was turned off when these neurons started to differentiate after gangliogenesis; and (3) it was expressed in the RPE homologs of segments 4-5 and 7-21 during gangliogenesis and the subsequent period of axonogenesis. We found that these neurons have three distinct peripheral axonal morphologies that correlate with the three temporal patterns of expression. Our results suggest that the homeobox gene Lox4 may determine neuronal identities within this series of segmental homologs.

Cell- and tissue-specific expression of putative protein kinase mRNAs in the embryonic leech, Hirudo medicinalis.

Protein kinases play important roles in various cellular interactions underlying metazoan development. To complement existing analyses of protein kinase function in the development of members of the three phyla, Chordata, Arthropoda, and Nematoda, we have begun to examine the cell- and tissue-specific localization of protein kinases in another metazoan phylum, the Annelida. For this purpose, we used the polymerase chain reaction to amplify putative protein kinase catalytic domain cDNAs from the medicinal leech, Hirudo medicinalis. This strategy allowed us to identify 11 cytoplasmic and receptor tyrosine kinase catalytic domains, and 2 cytoplasmic serine/threonine kinase catalytic domains. Using these cDNAs as probes for nonradioactive whole-mount in situ hybridization, we examined the embryonic expression pattern of each of the corresponding putative kinase mRNAs. As has been found in other species, most of the Hirudo protein kinase mRNAs were expressed in a highly specific manner in certain embryonic cells and tissues. We found both neuron- and glia-specific kinases within the nervous system, as well as kinases expressed in non-nervous tissues, such as the haemocoelomic, muscular, and excretory systems. These kinase cDNAs encode proteins likely to be critical for proper development, and can be used as cell- and tissue-specific histological probes for the analysis of Hirudo embryogenesis.

Interactions between segmental homologs and between isoneuronal branches guide the formation of sensory terminal fields.

Process outgrowth and peripheral field innervation by an identified mechanosensory neuron were examined in the intact embryonic leech. The dorsal pressure-sensitive (PD) neurons of the leech CNS are found as bilateral pairs in every segmental ganglion, and are amenable to study at early ages in intact embryos. Each PD has one major axonal projection that putatively pioneers the nerve to the dorsal body wall and branches extensively in its own segment, and two minor projections that innervate, via neighboring ganglia, smaller areas in adjacent segments. We found that adjacent embryonic PD cells form overlapping terminal fields in the body wall, but that the extent of overlap was governed by inhibitory interactions among these fields. When one PD neuron was ablated, the adjacent PD cell changed its peripheral arborization by (1) its major axon producing more filopodia and extending longer side branches toward the ablated cell and (2) its minor axon producing a large arbor in the operated segment. Interestingly, although growth was biased toward the side of the ablated neuron, reduced outgrowth of the PD cell was found on the side away from the ablation, while the total extent of arborization of the PD cell kept relatively constant. Further, we found that axotomy of the major PD projection resulted in extensive outgrowth of its minor projections. These results suggest that a single PD neuron has a limited capacity for growth, each of its branches growing at the expense of the others, and that inhibitory interactions between neighboring PD neurons influence the extent and direction of that growth.

Developing neurons use a putative pioneer's peripheral arbor to establish their terminal fields.

Pioneer neurons are known to guide later developing neurons during the initial phases of axonal outgrowth. To determine whether they are also important in the formation of terminal fields by the follower cells, we studied the role of a putative leech pioneer neuron, the pressure-sensitive (PD) neuron, in the establishment of other neurons' peripheral arbors. The PD neuron has a major axon that exits from its segmental ganglion to grow along the dorsal-posterior (DP) nerve to the dorsal body wall, where it arborizes extensively mainly in its own segment. It also has two minor axons that project to the two adjacent segments but branch to a lesser degree. We found that the peripheral projections of several later developing neurons, including the AP motor neuron and the TD sensory neuron, followed, with great precision, the major axon and peripheral arbor of the consegmental PD neuron, up to its fourth-order branches. When a PD neuron was ablated before it had grown to the body wall, the AP and TD axons grew normally toward and reached the target area, but then formed terminal arbors that were greatly reduced in size and abnormal in morphology. Further, if the ablation of a PD neuron was accompanied by the induction, in the same segment, of greater outgrowth of the minor axon of a PD neuron from the adjacent segment, the arbors of the same AP neurons grew along these novel PD neuron branches. These results demonstrate that the peripheral arbor of a PD neuron is a both necessary and sufficient template for the formation of normal terminal fields by certain later growing follower neurons.