We have been studying the molecular and cellular mechanisms underlying the development of the vertebrate central nervous system. It covers all the major events including the patterning of the nervous system, neuronal differentiation, axonal navigation and targeting, synapse formation and plasticity, and neuronal regeneration, mainly in the visual system. The scope of our interests also encompasses various functions of the matured brain, including sensation, behavior, learning and memory.

I. Regional specification in the retina and topographic retinotectal projection

Topographic maps are a fundamental feature of neural networks in the nervous system. Understanding the molecular mechanisms by which topographically ordered neuronal connections are established during development has long been a major challenge in developmental neurobiology. The retinotectal projection of lower vertebrates including birds has been used as a readily accessible model system. In this projection, the temporal (posterior) retina is connected to the rostral (anterior) part of the contralateral optic tectum, the nasal (anterior) retina to the caudal (posterior) tectum, and likewise the dorsal and ventral retina to the ventral and dorsal tectum, respectively. Thus, images received by the retina are precisely projected onto the tectum in a reversed manner.

Since 1992, we have been devoting our efforts to searching for molecules with asymmetrical distribution in the embryonic chick retina, and to characterization of their roles in the topographic retinotectal projection. In 1996, we first identified two winged-helix transcriptional regulators, CBF-1 and CBF-2, expressed in the nasal and temporal retina, respectively. Misexpression experiments in the retina using a retroviral vector showed that these two transcription factors determine the regional specificity of the retinal ganglion cells, namely, the directed axonal projections to the appropriate tectal targets along the anteroposterior axis. Secondly, we identified a novel retinoic acid-generating enzyme, RALDH-3, which is specifically expressed in the ventral region of the retina, together with a dorsal-specific enzyme RALDH-1.

Furthermore, we recently identified a novel secretory protein, Ventroptin, which has BMP-4 neutralizing activity. Ventroptin is expressed in the retina with a ventral high-dorsal low gradient at early stages. This expression pattern is complementary to that of BMP-4. At later stage (E6~), a nasal high-temporal low gradient expression pattern of it is also detected. Ventroptin thus shows a double-gradient expression profile along the dorsoventral and nasotemporal axes. Misexpression of Ventroptin altered expression patterns of several topographic genes and projection of the retinal ganglion-cell axons to the tectum along the both axes.

In this year of 2003, we revealed that misexpression of CBF-1 represses the expression of EphA3 and CBF-2, and induces that of SOHo-1, GH6, ephrin-A2 and ephrin-A5. CBF-1 controls ephrin-A5 by a DNA binding-dependent mechanism, ephrin-A2 by a DNA binding-independent mechanism, and CBF-2, SOHo-1, GH6 and EphA3 by dual mechanisms (Fig. 1A). BMP-2 expression begins double-gradiently in the retina from E5 instead of BMP-4 in a complementary pattern to the Ventroptin expression. Ventroptin antagonizes BMP-2 as well as BMP-4. CBF-1 interferes in BMP-2 signaling and thereby induces expression of ephrin-A2. Our data suggest that CBF-1 is located at the top of the gene cascade for the regional specification along the nasotemporal (NT) axis in the retina (Fig. 1B) and distinct BMP signals play pivotal roles in the topographic projection along both axes.

Fig.1. The molecular mechanisms by which CBF-1 controls the expression of topographic molecules. (A) Schematic representation of modes of actions of CBF-1. You can see that one Eph-ephrin system is controlled as a set by each mode of CBF-1 action.

(B) Expressional regulation of the topographic molecules along the NT axis by CBF-1. EphA3 and ephrins are directly implicated in the control of axon guidance. CBF-1 and Ventroptin repress expression of BMP-2 by inhibiting BMP signaling as an interrupter and antagonist, respectively, and induce ephrin-A2 expression. CBF-1 represses the transcription of negative regulators, X and Y. SOHo-1 and GH6 inhibit the expression of EphA3. CBF-1 also represses CBF-2 expression, however, its downstream target genes have not been identified yet.

Currently, with respect to the other identified molecules, we are conducting misexpression experiments using chick embryos and generating knockout and transgenic mice to elucidate the molecular functions. We expect that our studies will lead to dissection of the molecular mechanism underlying the retinal patterning and topographic retinotectal projection, and ultimately to uncovering the basic principles for establishing complicated but extremely precise neural networks.

II. Axonal morphogenesis and behavior

During development, cells undergo dynamic morphological changes by rearrangements of the cytoskeleton including microtubules. However, molecular mechanisms underlying the microtubule remodeling between orientated and disorientated formations are almost unknown. We found that novel subtypes of collapsin response mediator proteins (CRMP-As) in addition to the originals (CRMP-Bs), which occur from the alternative usage of different first coding exons, are involved in this conversion of microtubule patterns.

Overexpression of CRMP2A and CRMP2B in chick embryonic fibroblasts induced orientated and disorientated patterns of microtubules, respectively. Moreover, sequential overexpression of another subtype overcame the effect of the former expression of the countersubtype. Overexpression experiments in cultured chick retinae showed that CRMP2B promoted axon branching and suppressed axon elongation of ganglion cells, while CRMP2A blocked these effects when co-overexpressed (Fig. 2). Our findings suggest that the opposing activities of CRMP2A and CRMP2B contribute to the cellular morphogenesis including neuronal axonogenesis through remodeling of microtubule organization.

Fig.2. Antagonistic effects of CRMP2A and –2B on axon morphology. Retinae from chick E6 embryos were electroporated with IGG vector to express CRMP2B (A) and CRMP2B + CRMP2A (B), and cultured for 2 days.

III. Physiological roles of protein tyrosine phosphatase receptor type Z (Ptprz)

Protein tyrosine phosphorylation plays crucial roles in various biological aspects including all stages of brain development. In 1994, we found that PTPz/RPTPb/Ptprz, a nervous system-rich receptor-type PTP, is expressed as a chondroitin sulfate proteoglycan in the brain. Ptprz is expressed from the early developmental stage to the adulthood in neurons as well as astrocytes. This suggests that this molecule plays variegated roles in the brain development and brain function.

We found in 1996 that Ptprz binds pleiotrophin (PTN)/HB-GAM and midkine (MK), closely related heparin-binding growth factors which share many biological activities. The chondroitin sulfate portion of Ptprz is essential for the high affinity binding (Kd = ~0.25 nM) to these growth factors, and removal of chondroitin sulfate chains results in a marked decrease of binding affinity (Kd = ~13 nM).

To identify the substrate molecules of Ptprz, we recently developed the yeast substrate-trapping system. This system is based on the yeast two-hybrid system with two essential modifications: Conditional expression of v-src to tyrosine-phosphorylate the prey proteins and screening using a substrate-trap mutant of PTP as bait. Using this system, we successfully isolated a number of candidate clones for substrate molecules (ex. GIT1/Cat-1) and continuously-interacting molecules (ex. PSD-95/SAP90) for Ptprz. We are now continuing efforts to characterize these candidate clones.

In addition, to know the physiological roles of Ptprz in vivo, we generated Ptprz-deficient mice in which the Ptprz gene was replaced with the LacZ gene in 1997. We are currently studying the phenotype of Ptprz-deficient mice biochemically, anatomically, physiologically and ethologically.
We reported this year that mice deficient in Ptprz do not show mucosal damage by VacA, although VacA is incorporated into the gastric epithelial cells to the same extent as in wild-type mice (Fig. 3). The vacuolating cytotoxin VacA produced by Helicobacter pylori causes massive cellular vacuolation in vitro and gastric tissue damage in vivo, leading to gastric ulcers, when administered intragastrically. Primary cultures of gastric epithelial cells from Ptprz^+/+ and Ptprz^-/- mice also showed similar incorporation of VacA, cellular vacuolation and reduction in cellular proliferation, but only Ptprz^+/+ cells showed marked detachment from a reconstituted basement membrane 24 h after treatment with VacA. VacA bound to Ptprz, and the levels of tyrosine phosphorylation of Git1, a Ptprz substrate, were higher after treatment with VacA, indicating that VacA behaves as a ligand for Ptprz. Furthermore, PTN, an endogenous ligand of Ptprz, also induced gastritis specifically in Ptprz^+/+ mice when administered orally. Taken together, these data indicate that erroneous Ptprz signaling induces gastric ulcers.

Fig.3. Pathological analyses of gastric tissues treated with VacA. (A,a) Stereomicroscopic appearance of the inside of stomachs from Ptprz+/+ mice 48 h after administration of VacA (500 mg per kg body weight). (A,b) Diagrammatic representation of a showing the gastric ulcer (red area with arrowheads). (B) Appearance of the inside of stomachs from Ptprz-/- mice treated equally. (C,D) Gastric sections stained with hematoxylin and eosin. Loss of epithelial cells and gastric gland structure (arrowheads) was observed in the mucosal layer in Ptprz+/+ mice (C) but not Ptprz-/- mice (D). The lower panels are enlargements. Scale bars: 500 mm. M, Mucosa; Sm, Submucosa; Mp, Muscularis propia.

IV. Na⁺-level sensing in the brain

Sodium-level sensing system in the brain is essential for the regulation of the sodium and water balance in body fluids. Previously, we demonstrated that Na_x (also called Na_v2/NaG) gene is expressed in the circumventricular organs and Na_x-deficient mice ingest salt in excess. In Na_x-deficient mice, c-fos expression in the subfornical organ (SFO) and organum vasculosum laminae terminalis (OVLT) was markedly elevated as compared with wild-type mice. Stimulation of the SFO/OVLT of wild-type animals by infusion of a hypertonic sodium solution into intracerebroventricule (ICV) leads to avoidance of salt intake. In contrast, Na_x-deficient mice did not show such aversion.

We showed that Na_x is a sodium channel which is sensitive to the increase of extracellular sodium level. Entry of sodium ions occurred in response to a rise of the extracellular sodium concentration (C_1/2= 159 mM). In contrast, these responses were not observed in Na_x-immunonegative cells or neurons of Na_x-deficient mutant origin. Transfection of Na_x cDNA conferred the sodium sensitivity on Na_x-deficient cells. All of the GABA-immunopositive neurons isolated from the SFO responded to the extracellular sodium increase. Based on these findings, we proposed that GABAergic inhibitory neurons expressing Na_x control the activity of the SFO and suppress the salt-intake behavior of animals under thirst conditions.