Phylogenetic conservation of brain microtubule-associated proteins MAP2 and tau

Evolutionary perspective of Big tau structure: 4a exon variants of MAPT

The MAPT gene encoding the microtubule-associated protein tau can generate multiple isoforms by alternative splicing giving rise to proteins which are differentially expressed in specific areas of the nervous system and at different developmental stages. Tau plays important roles in modulating microtubule dynamics, axonal transport, synaptic plasticity, and DNA repair, and has also been associated with neurodegenerative diseases (tauopathies) including Alzheimer's disease and frontotemporal dementia. A unique high-molecular-weight isoform of tau, originally found to be expressed in the peripheral nervous system and projecting neurons, has been termed Big tau and has been shown to uniquely contain the large exon 4a that significantly increases the size and 3D structure of tau. With little progress since the original discovery of Big tau, more than 25 years ago, we have now completed a comprehensive comparative study to analyze the structure of the MAPT gene against available databases with respect to the composition of the tau exons as they evolved from early vertebrates to primates and human. We focused the analysis on the evolution of the 4a exon variants and their homology relative to humans. We discovered that the 4a exon defining Big tau appears to be present early in vertebrate evolution as a large insert that dramatically changed the size of the tau protein with low sequence conservation despite a stable size range of about 250aa, and in some species a larger 4a-L exon of 355aa. We suggest that 4a exon variants evolved independently in different species by an exonization process using new alternative splicing to address the growing complexities of the evolving nervous systems. Thus, the appearance of a significantly larger isoform of tau independently repeated itself multiple times during evolution, accentuating the need across vertebrate species for an elongated domain that likely endows Big tau with novel physiological functions as well as properties related to neurodegeneration.

Reassessment of Neuronal Tau Distribution in Adult Human Brain and Implications for Tau Pathobiology

Tau is a predominantly neuronal, soluble and natively unfolded protein that can bind and stabilize microtubules in the central nervous system. Tau has been extensively studied over several decades, especially in the context of neurodegenerative diseases where it can aberrantly aggregate to form a spectrum of pathological inclusions. The presence of tau inclusions in the form of neurofibrillary tangles, neuropil threads and dystrophic neurites within senile plaques are essential and defining features of Alzheimer’s disease. The current dogma favors the notion that tau is predominantly an axonal protein, and that in Alzheimer’s disease there is a redistribution of tau towards the neuronal soma that is associated with the formation of pathological inclusions such as neurofibrillary tangles and neuropil threads. Using novel as well as previously established highly specific tau antibodies, we demonstrate that contrary to this overwhelmingly accepted fact, as asserted in numerous articles and reviews, in adult human brain, tau is more abundant in cortical gray matter that is enriched in neuronal soma and dendrites compared to white matter that is predominantly rich in neuronal axons. Additionally, in Alzheimer’s disease tau pathology is significantly more abundant in the brain cortical gray matter of affected brain regions compared to the adjacent white matter regions. These findings have important implications for the biological function of tau as well as the mechanisms involved in the progressive spread of tau associated with the insidious nature of Alzheimer’s disease.

Ectopic Expression Induces Abnormal Somatodendritic Distribution of Tau in the Mouse Brain

Tau is a microtubule (MT)-associated protein that is localized to the axon. In Alzheimer's disease, the distribution of tau undergoes a remarkable alteration, leading to the formation of tau inclusions in the somatodendritic compartment. To investigate how this mislocalization occurs, we recently developed immunohistochemical tools that can separately detect endogenous mouse and exogenous human tau with high sensitivity, which allows us to visualize not only the pathological but also the pre-aggregated tau in mouse brain tissues of both sexes. Using these antibodies, we found that in tau-transgenic mouse brains, exogenous human tau was abundant in dendrites and somata even in the presymptomatic period, whereas the axonal localization of endogenous mouse tau was unaffected. In stark contrast, exogenous tau was properly localized to the axon in human tau knock-in mice. We tracked this difference to the temporal expression patterns of tau. Endogenous mouse tau and exogenous human tau in human tau knock-in mice exhibited high expression levels during the neonatal period and strong suppression into the adulthood. However, human tau in transgenic mice was expressed continuously and at high levels in adult animals. These results indicated the uncontrolled expression of exogenous tau beyond the developmental period as a cause of mislocalization in the transgenic mice. Superresolution microscopic and biochemical analyses also indicated that the interaction between MTs and exogenous tau was impaired only in the tau-transgenic mice, but not in knock-in mice. Thus, the ectopic expression of tau may be critical for its somatodendritic mislocalization, a key step of the tauopathy.SIGNIFICANCE STATEMENT Somatodendritic localization of tau may be an early step leading to the neuronal degeneration in tauopathies. However, the mechanisms of the normal axonal distribution of tau and the mislocalization of pathological tau remain obscure. Our immunohistochemical and biochemical analyses demonstrated that the endogenous mouse tau is transiently expressed in neonatal brains, that exogenous human tau expressed corresponding to such tau expression profile can distribute into the axon, and that the constitutive expression of tau into adulthood (e.g., human tau in transgenic mice) results in abnormal somatodendritic localization. Thus, the expression profile of tau is tightly associated with the localization of tau, and the ectopic expression of tau in matured neurons may be involved in the pathogenesis of tauopathy.

Distribution of endogenous normal tau in the mouse brain

Tau is a microtubule‐associated protein (MAP) that is localized to the axon. In Alzheimer's disease (AD), the distribution of tau undergoes a remarkable alteration, leading to the formation of tau inclusions in the somatodendritic compartment. While the abnormal aggregated tau has been extensively studied in human patient tissues and animal models of AD, how normal tau localizes to the axon, which would be the foundation to understand how the mis‐localization occurs, has not been well studied due to the poor detectability of normal unaggregated tau in vivo. Therefore, we developed immunohistochemical techniques that can detect normal mouse and human tau in brain tissues with high sensitivity. Using these techniques, we demonstrate the global distribution of tau in the mouse brain and confirmed that normal tau is exclusively localized to the axonal compartment in vivo. Interestingly, tau antibodies strongly labeled nonmyelinated axons such as hippocampal mossy fibers, while white matters generally exhibited low levels of immunoreactivity. Furthermore, mouse tau is highly expressed not only in neurons but also in oligodendrocytes. With super resolution imaging using the stimulated‐depletion microscopy, axonal tau appeared punctate rather than fibrous, indicating that tau decorates microtubules sparsely. Co‐labeling with presynaptic and postsynaptic markers revealed that normal tau is not localized to synapses but sparsely distributes in the axon. Taken together, this study reports novel antibodies to investigate the localization and mis‐localization of tau in vivo and novel findings of normal tau localization in the mouse brain.

Microtubule-associated protein tau promotes neuronal class II β-tubulin microtubule formation and axon elongation in embryonic Xenopus laevis

Compared with its roles in neurodegeneration, much less is known about microtubule-associated protein tau's normal functions in vivo, especially during development. The external development and ease of manipulating gene expression of Xenopus laevis embryos make them especially useful for studying gene function during early development. To study tau's functions in axon outgrowth, we characterized the most prominent tau isoforms of Xenopus embryos and manipulated their expression. None of these four isoforms were strictly analogous to those commonly studied in mammals, as all constitutively contained exon 10, which is preferentially removed from mammalian fetal tau isoforms, as well as exon 8, which in mammals is rare. Nonetheless, like mammalian tau, Xenopus tau exhibited alternative splicing of exon 4a, which in mammals distinguishes 'big' tau of peripheral neurons, and exon 6. Strongly suppressing tau expression with antisense morpholino oligonucleotides only modestly compromised peripheral nerve outgrowth of intact tadpoles, but severely disrupted neuronal microtubules containing class II β-tubulins while leaving other microtubules largely unperturbed. Thus, the relatively mild dependence of axon development on tau likely resulted from having only a single class of microtubules disrupted by its loss. Also, consistent with its greater expression in long peripheral axons, boosting expression of 'big' tau increased neurite outgrowth significantly and enhanced tubulin acetylation more so than did the smaller isoform. These data demonstrate the utility of Xenopus as a tool to gain new insights into tau's functions in vivo.

Divergence of RNA localization between rat and mouse neurons reveals the potential for rapid brain evolution

Background

Neurons display a highly polarized architecture. Their ability to modify their features under intracellular and extracellular stimuli, known as synaptic plasticity, is a key component of the neurochemical basis of learning and memory. A key feature of synaptic plasticity involves the delivery of mRNAs to distinct sub-cellular domains where they are locally translated. Regulatory coordination of these spatio-temporal events is critical for synaptogenesis and synaptic plasticity as defects in these processes can lead to neurological diseases. In this work, using microdissected dendrites from primary cultures of hippocampal neurons of two mouse strains (C57BL/6 and Balb/c) and one rat strain (Sprague–Dawley), we investigate via microarrays, subcellular localization of mRNAs in dendrites of neurons to assay the evolutionary differences in subcellular dendritic transcripts localization.

Results

Our microarray analysis highlighted significantly greater evolutionary diversification of RNA localization in the dendritic transcriptomes (81% gene identity difference among the top 5% highly expressed genes) compared to the transcriptomes of 11 different central nervous system (CNS) and non-CNS tissues (average of 44% gene identity difference among the top 5% highly expressed genes). Differentially localized genes include many genes involved in CNS function.

Conclusions

Species differences in sub-cellular localization may reflect non-functional neutral drift. However, the functional categories of mRNA showing differential localization suggest that at least part of the divergence may reflect activity-dependent functional differences of neurons, mediated by species-specific RNA subcellular localization mechanisms.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-883) contains supplementary material, which is available to authorized users.

Neural differentiation on synthetic scaffold materials

The potential of stem cells to differentiate into a variety of subgroups of neural cells makes stem cell differentiation and transplantation a promising candidate for neurodegenerative disorder therapies. However, selective differentiation of stem cells to neurons while preventing glial scar formation is a complex process. Mimicking the natural environment of neural tissue is pivotal, thus various synthetic materials have been developed for this purpose. The synthetic scaffolds can direct stem cells into a neural lineage by including extracellular factors that act on cell fate, which are mainly soluble signals, extracellular matrix proteins and physical factors (e.g. elasticity and topography). This article reviews synthetic materials developed for neural regeneration in terms of their extracellular matrix mimicking properties. Functionalization of synthetic materials by addition of bioactive chemical groups and adjustment of physical properties such as topography, electroactivity and elasticity are discussed.

Semaphorin 3A induces CaV2.3 channel-dependent conversion of axons to dendrites

Polarized neurites (axons and dendrites) form the functional circuitry of the nervous system. Secreted guidance cues often control the polarity of neuron migration and neurite outgrowth by regulating ion channels. Here, we show that secreted semaphorin 3A (Sema3A) induces the neurite identity of Xenopus spinal commissural interneurons (xSCINs) by activating Ca(V)2.3 channels (Ca(V)2.3). Sema3A treatment converted the identity of axons of cultured xSCINs to that of dendrites by recruiting functional Ca(V)2.3. Inhibition of Sema3A signalling prevented both the expression of Ca(V)2.3 and acquisition of the dendrite identity, and inhibition of Ca(V)2.3 function resulted in multiple axon-like neurites of xSCINs in the spinal cord. Furthermore, Sema3A-triggered cGMP production and PKG activity induced, respectively, the expression of functional Ca(V)2.3 and the dendrite identity. These results reveal a mechanism by which a guidance cue controls the identity of neurites during nervous system development.

Contrasting expression of Kv4.3, an A-type K+ channel, in migrating Purkinje cells and other post-migratory cerebellar neurons

Kv4.3, an A-type K+ channel, is the only channel molecule showing anterior-posterior (A-P) compartmentalization in the granular layer of mammalian cerebellum known so far. Kv4.3 mRNA has been detected from the posterior but not anterior granular layer in adult rat cerebellum. To characterize this A-P compartmentalization further, we examined Kv4.3 protein expression in rat cerebellum by immunohistochemistry at the embryonic, early postnatal and adult stages. Specificity of the Kv4.3 antibody was confirmed by both Western blot and immunoprecipitation analysis. In adulthood, Kv4.3 was detected from the somatodendritic domain of posterior granule cells, with a restriction boundary in the vermal lobule VI extending laterally to the hemispheric crus 1 ansiform lobules. At the early postnatal stage, this A-P pattern first appeared on postnatal day 8, when significant numbers of granule cells had migrated into the posterior granular layer and started to express Kv4.3. Similar Kv4.3 expression in the somatodendritic domain of post-migratory neurons in the cerebellum was also observed in basket cells, stellate cells, a subset of GABAergic deep neurons, Lugaro cells and, probably, deep Lugaro cells. However, none of them showed A-P compartmentalization. Strikingly, we found Kv4.3 in several clusters of migrating Purkinje cells with mediolateral compartmentalization. These Purkinje cells no longer expressed Kv4.3 after completing the migration. By contrasting the expression in migrating and post-migratory neurons, our results suggest that Kv4.3 may play an important role in the development of cerebellum, as well as in the mature cerebellum.

Tau and axonopathy in neurodegenerative disorders

The microtubule (MT)-associated protein (MAP) tau in neurons has been implicated as a significant factor in the axonal growth, development of neuronal polarity, and the maintenance of MT dynamics. Tau is localized to the axon, and is known to promote MT assembly and to stabilize axonal MTs. These functions of tau are primarily regulated by the activities of protein kinases and phosphatases. In Alzheimer's disease and other neurodegenerative disorders, abundant filamentous tau inclusions are found to be major neuropathological characteristics of these diseases. Both somato-dendritic and axonal tau lesions appear to be closely associated with axonal disruption. Furthermore, recent discoveries of pathogenic mutations on the tau gene suggest that abnormalities of tau alone are causative of neurodegeneration. Finally, analyses of transgenic mice that express human tau proteins have enabled in vivo quantitative assessments of axonal functions and have provided information about mechanistic relationships between pathological alteration of tau and axonal degeneration.

MAP2 Isoforms in Developing Cat Cerebral Cortex and Corpus Callosum

The microtubule-associated protein MAP2 was studied in the developing cat visual cortex and corpus callosum. Biochemically, no MAP2a was detectable in either structure during the first postnatal month; adult cortex revealed small amounts of MAP2a. MAP2b was abundant in cortical tissue during the first postnatal month and decreased in concentration towards adulthood; it was barely detectable in corpus callosum at all ages. MAP2c was present in cortex and corpus callosum at birth; in cortex it consisted of three proteins of similar molecular weights between 65 and 70 kD. The two larger, phosphorylated forms disappeared after postnatal day 28, the smaller form after day 39. In corpus callosum, MAP2c changed from a phosphorylated to an unphosphorylated variant during the first postnatal month and then disappeared. Immunocytochemical experiments revealed MAP2 in cell bodies and dendrites of neurons in all cortical layers, from birth onwards. In corpus callosum, in the first month after birth, a little MAP2, possibly MAP2c, was detectable in axons. The present data indicate that MAP2 isoforms differ in their cellular distribution, temporal appearance and structural association, and that their composition undergoes profound changes during the period of axonal stabilization and dendritic maturation.

Molecular cloning of XTP, a tau-like microtubule-associated protein from Xenopus laevis tadpoles

The microtubules of the mammalian nervous system are stabilised by several microtubule-associated proteins (MAPs), including the tau and MAP-2 protein families. The most prominent feature of mammalian tau and MAP-2 proteins is a common and highly homologous microtubule-binding region consisting of three or four imperfect tandem repeats. In this paper we report the cloning and characterisation of a Xenopus laevis tau-like protein (XTP) from tadpole tails. This protein encompasses two isoforms of 673 or 644 amino acids with four tandem repeats that are highly homologous to mammalian tau repeats. Both isoforms share a common amino terminal half, whereas the carboxyl terminus downstream of the repeat region is unique for each isoform. Northern blot analysis revealed that both isoforms are preferentially expressed in the tail of X. laevis tadpoles, whereas a shorter version of XTP is expressed in the head. Recombinant proteins of both XTP isoforms were able to bind microtubules. The longest isoform, however, was more effective at promoting tubulin polymerisation, indicating that sequences downstream of the repeat region affect the microtubule assembling capacity. These results demonstrate that tau-like proteins are found in non-mammalian vertebrate species, where they may support the stability of microtubules.

MAP2 phosphorylation and visual plasticity in Xenopus

Microtubule-associated protein 2 (MAP2) has been implicated in activity-dependent structural changes in dendrites. MAP2 regulates the assembly of cytoskeletal proteins such as microtubules and actin, and its function is phosphorylation-dependent. In hippocampus, MAP2 has been reported to be dephosphorylated by activation of the NMDA-type glutamate receptor, a key player in synaptic plasticity. In this work, we used a phospho-specific MAP2 antibody (Ab 305) that recognizes epitopes close to the microtubule-binding domain to investigate the possible role of MAP2 in the Xenopus visual system. The binocular system in Xenopus exhibits activity-dependent synapse rearrangement during a critical period of development. We have found that, in critical period animals, NMDA receptor activation leads to the dephosphorylation of MAP2 at sites recognized by Ab 305 in a dose-dependent manner. We compared the responses of MAP2 to NMDA treatment in animals with high binocular plasticity (critical period juveniles and dark-reared adults) and low plasticity (normal adults). Our results show that, in all groups, NMDA treatment induces the dephosphorylation of MAP2. Tecta from frogs with different degrees of plasticity show no differences in the baseline level of MAP2 phosphorylation or in the NMDA-induced MAP2 dephosphorylation response. These results suggest that activity may modify dendrite structure via the NMDA receptor--MAP2-cytoskeletal protein pathway, but this pathway does not seem to be a determinant of the degree of plasticity.

Effect of antenatal betamethasone treatment on microtubule-associated proteins MAP1B and MAP2 in fetal sheep

Betamethasone has been used extensively to accelerate fetal lung maturation, yet little is known of its effects on neuronal morphogenesis in the developing fetus. Microtubule-associated proteins (MAPs) are a diverse family of cytoskeletal proteins that are important for brain development and the maintenance of neuroarchitecture. Vehicle (n = 7) or betamethasone (10 ug h-1, n = 7) was infused I.V. to fetal sheep over 48 h beginning at 0.87 of gestation (128 days of gestation), producing fetal plasma betamethasone concentrations resembling those to which the human fetus is exposed during antenatal glucocorticoid therapy. Paraffin sections of the left hemisphere were stained with monoclonal antibodies against MAP1B and the MAP2 isoforms MAP2a,b,c and MAP2a,b. The level of the juvenile isoform MAP2c was determined by comparison of the two MAP2 immunostainings. We were able to detect MAP1B and MAP2 immunoreactivity (IR) in the fetal sheep brain. MAP2c was the major MAP2, constituting 90.2 % of the total MAPBetamethasone exposure diminished MAP1B IR in the frontal cortex and caudate putamen (P < 0.05) but not in the hippocampus. A decrease of MAP2 IR was found in the frontal cortex, hippocampus and caudate putamen (P < 0.05). Loss of MAP2 IR was mainly due to the loss of MAP2c IR. Haematoxylin-eosin staining did not demonstrate irreversible neuronal damage. Regional cerebral blood flow determined using coloured microspheres was significantly decreased by 28 % in the frontal cortex and by 36 % in the caudate putamen but not in the hippocampus 24 h after the onset of betamethasone exposure (P < 0.05). The loss of MAP1B and MAP2a,b,c IR showed a significant correlation to the cerebral blood flow decrease only in the frontal cortex (P < 0.05). These data suggest that mechanisms other than metabolic insufficiency caused by the decreased cerebral blood flow may contribute to the loss of MAPs. The results suggest that clinical doses of betamethasone may have acute effects on cytoskeletal proteins in the fetal brain.

Immunocytochemical changes of cytoskeleton components and calmodulin in the frog cerebellum and optic tectum during hibernation

During hibernation, variation in the metabolism of nerve cells occurs. Since the cytoskeleton plays an important role in nerve cell function, we have analyzed the immunocytochemical expression of two cytoskeleton components, i.e. phosphorylated 200 kDa neurofilament protein, and microtubule-associated protein 2 in the cerebellum and optic tectum of hibernating frogs (Rana esculenta) in comparison with active animals. In addition, we have considered the immunocytochemical expression of calmodulin, which is known to be involved in neurofilament phosphorylation. In hibernating animals, there was a decrease in the immunoreactivity for phosphorylated 200 kDa neurofilament protein and microtubule-associated protein 2 of fibers in both the cerebellum and in the optic tectum. In contrast, in the large neurons of the cerebellum, i.e. Purkinje neurons, there was an increase in the immunoreactivity for microtubule-associated protein 2. The changes in the cytoskeleton components were accompanied by a decrease in calmodulin immunoreactivity in the cytoplasm of nerve cells of the cerebellum. All the changes observed are consistent with a low neuronal activity during hibernation, as also indicated by previous microdensitometric and microfluorometric data. This shows a higher degree of chromatin condensation in hibernating animals and suggests that hibernation represents a simple form of neuronal plasticity.

Tau-like proteins in the nervous system of goldfish

This report describes the presence of a group of tau-like proteins (TLPs) in goldfish central nervous system. The TLPs were immunoreactive with antibodies that recognized the carboxy-terminal domain of mammalian tau, but not with antibodies that recognized the amino-terminus. The TLPs of goldfish exhibited the basic properties of tau proteins including neuronal specificity, structural heterogeneity, heat stability and the ability to co-assemble with tubulin. We propose that TLPs may represent a precursor of tau, that share the microtubule binding domain and the carboxy-terminal domain with mammalian tau proteins. In contrast the amino-terminus of the TLPs is much shorter and may represent a more variable domain of tau proteins.

PTL-1, a microtubule-associated protein with tau-like repeats from the nematode Caenorhabditis elegans

Tau, MAP2 and MAP4 are structural microtubule-associated proteins (MAPs) that promote the assembly and stability of microtubules. They share three or four imperfect tandem repeats of an amino acid motif, which is involved in the binding to microtubules. All sequences to data containing this motif are of mammalian origin. We report here the cloning and functional characterisation of a new member of this family of proteins from the nematode Caenorhabditis elegans. This protein exists as two isoforms of 413 and 453 amino acids with four or five tandem repeats that are 50% identical to the tau/MAP2/MAP4 repeats. Both isoforms bind to microtubules and promote microtubule assembly, with the five-repeat isoform being more effective at promoting assembly than the four-repeat isoform. When expressed in COS cells, the five-repeat isoform co-localises with microtubules and induces the formation of microtubule bundles, whereas its expression in Sf9 cells leads to the extension of long unipolar processes. In view of its length, amino acid sequence and functional characteristics, we have named this invertebrate structural MAP ‘Protein with Tau-Like Repeats’ (PTL-1). In C. elegans PTL-1 is expressed in two places known to require microtubule function. It is first seen in the embryonic epidermis, when circumferentially oriented microtubules help to distribute forces generated during elongation. Later, it is found in mechanosensory neurons which contain unusual 15 protofilament microtubules required for the response to touch. These findings indicate that MAPs of the tau/MAP2/MAP4 family are found throughout much of the animal kingdom, where they may play a role in specialised processes requiring microtubules.

Cytoplasmic mechanisms of axonal and dendritic growth in neurons

The structural mechanisms responsible for the gradual elaboration of the cytoplasmic elongation of neurons are reviewed. In addition to discussing recent work, important older work is included to inform newcomers to the field how the current perspective arose. The highly specialized axon and the less exaggerated dendrite both result from the advance of the motile growth cone. In the area of physiology, studies in the last decade have directly confirmed the classic model of the growth cone pulling forward and the axon elongating from this tension. Particularly in the case of the axon, cytoplasmic elongation is closely linked to the formation of an axial microtubule bundle from behind the advancing growth cone. Substantial progress has been made in understanding the expression of microtubule-associated proteins during neuronal differentiation to stiffen and stabilize axonal microtubules, providing specialized structural support. Studies of membrane organelle transport along the axonal microtubules produced an explosion of knowledge about ATPase molecules serving as motors driving material along microtubule rails. However, most aspects of the cytoplasmic mechanisms responsible for neurogenesis remain poorly understood. There is little agreement on mechanisms for the addition of new plasma membrane or the addition of new cytoskeletal filaments in the growing axon. Also poorly understood are the mechanisms that couple the promiscuous motility of the growth cone to the addition of cytoplasmic elements.

The Xenopus laevis homologue to the neuronal cyclin-dependent kinase (cdk5) is expressed in embryos by gastrulation

Phosphorylation of the neuronal cytoskeletal proteins NF-H, NF-M and tau is important for normal axonal development, and is involved in axonal injury and neurodegenerative diseases. In mammalian neurons, one kinase that phosphorylates these axonal cytoskeletal proteins is cyclin-dependent kinase 5 (cdk5). Cdk5 is a member of the family of cyclin-dependent kinases (cdks), whose other family members regulate mitosis. Unlike the other cdks, cdk5 is abundant in differentiated neurons. Embryos of the clawed frog Xenopus laevis have proved useful for studying other cyclin-dependent kinases, neurofilament proteins and tau during development. As a first step in studying the role of cdk5 and its effects on neurofilaments during Xenopus neural development, four cDNA clones were isolated by screening a frog brain cDNA library at lowered stringency with a cDNA probe to rat cdk5. The frog cdk5 clones encoded a protein of 292 amino acids that was 97% identical to rat cdk5. In situ hybridization demonstrated that the Xenopus cdk5 transcript, like that of mammals, was expressed in differentiated post-mitotic neurons. The high degree of sequence homology and shared neuronal expression suggests that the role of cdk5 in neurons is highly conserved between mammals and amphibians. Northern blot analysis indicated that during Xenopus development, cdk5 mRNA was first expressed between the midblastula transition and gastrulation, which both occur long before neuronal differentiation. These stages mark the transition from synchronous to asynchronous cell division and are the earliest stages of zygotic gene expression. This early expression of Xenopus cdk5 mRNA implies a role for cdk5 during embryogenesis that is separate from its role as an axonal cytoskeletal protein kinase. These observations provide the foundation for exploiting X. laevis embryos to study the role of cdk5 both in the early stages of axonal differentiation and also in early embryogenesis.

Changes in the microtubule proteins in the developing and transected spinal cords of the bullfrog tadpole: induction of microtubule-associated protein 2c and enhanced levels of Tau and tubulin in regenerating central axons

The distribution of tubulin, microtubule-associated protein 2 and Tau in the spinal cords of bullfrog tadpoles during development and after transection was studied. alpha-Tubulin or beta-tubulin immunoreactivity was present in the axons, neuronal perikarya and dendrites, as revealed by immunocytochemistry. The axonal staining intensity of the tubulins in the tadpoles was significantly stronger than that in the adult bullfrog. Microtubule-associated protein 2 immunoreactivity was localized largely to dendrites and expanded from distal to proximal dendrites with time; a high-molecular-weight microtubule-associated protein 2 was seen on the immunoblots of cord homogenates throughout development Tau1 stained mainly the axons. Two-dimensional gel immunoblotting disclosed that the tadpole contained a greater number of isoforms of Tau than the frog. Complete transection of the spinal cords of stage IV tadpoles was followed by regeneration of the damaged cord region. The levels of tubulin and Tau immunoreactivity in the regenerating axons of the ventral fasciculi were generally increased. Strikingly, microtubule-associated protein 2 immunoreactivity appeared in the regenerating axons and the chromatolytic cell bodies of axotomized motor neurons, paralleling the induction of microtubule-associated protein 2c in the regenerating cord segment shown by immunoblotting. The chromatolytic cell bodies were also markedly labeled by Tau1, whereas the high-molecular-weight microtubule-associated protein 2 diminished on the immunoblots, in accordance with the reduced level of staining for the dendrites. It is apparent that the changes in the cytoskeletal proteins in the regenerating axons mostly recapitulated their developmental patterns. Moreover, the data indicate a close relationship between tubulin and microtubule-associated proteins in axonal growth as well as providing evidence for similar molecular mechanisms underlying successful regeneration for central and peripheral axons.

Tau-beta-galactosidase, an axon-targeted fusion protein

The most commonly used enzymatic reporter molecule, Escherichia coli beta-galactosidase (beta-gal; beta-D-galactoside galactohydrolase, EC 3.2.1.23), fails to readily diffuse into axons; consequently, the morphologies of beta-gal-labeled neurons cannot directly be determined. For analysis of neuronal pathfinding and synaptic connectivity, this information is essential. We have constructed an axon-targeted beta-gal reporter by fusing the cDNA encoding the bovine microtubule-binding protein, tau, to lacZ, the E. coli gene encoding beta-gal. This reporter labels cell bodies and axons when expressed by developing and adult Drosophila neurons. It also reveals the entire cellular extent of nonneuronal cells such as muscle fibers and glia. To generate neuronal markers for studies of Drosophila neural development, we constructed a tau-beta-gal enhancer-trap transposon. From 1500 independent lines generated by mobilization of this transposon, we have isolated a set of useful markers for specific subsets of neurons, glia, and muscles. Since the tau cDNA-lacZ reporter utilizes bovine tau, it may also effectively target beta-gal in vertebrate neurons and prove to be a useful reagent for the analysis of vertebrate nervous systems.

Purification of microtubule proteins from Xenopus egg extracts: identification of a 230K MAP4-like protein

We describe the purification of microtubule proteins from Xenopus egg extracts by temperature-dependent assembly and disassembly in the presence of dimethyl sulfoxide and identify a number of presumptive microtubule-associated proteins (MAPs). One of these proteins has a molecular weight of 230 kDa and is immunologically related to HeLa MAP4. We show that this MAP is heat stable and phosphorylated, and that it promotes elongation of microtubules from axonemes.

Attenuation of microtubule-associated protein 1B expression by antisense oligodeoxynucleotides inhibits initiation of neurite outgrowth

Microtubule-associated protein 1B, formerly also known as microtubule-associated protein 5, is the first structural microtubule accessory protein to appear in outgrowing axons. In PC12 pheochromocytoma cells microtubule-associated protein 1B levels increase several-fold after the addition of nerve growth factor and this increase is correlated with the initiation of process formation. To determine whether microtubule-associated protein 1B is essential for neurite outgrowth, we used antisense oligodeoxynucleotides to inhibit its expression in nerve growth factor-treated PC12 cells in the rat. The application of several different antisense oligodeoxynucleotides to the microtubule-associated protein 1B mRNA sequence inhibited both microtubule-associated protein 1B expression and neurite extension. Specificity was shown by the lack of effect of control sense oligonucleotides and by the lack of effect of the microtubule-associated protein 1B antisense oligodeoxynucleotides on the expression of either tubulin or microtubule-associated protein 3, another microtubule-associated protein whose synthesis is stimulated by nerve growth factor treatment of PC12 cells. After removal of the antisense oligodeoxynucleotides, microtubule-associated protein 1B expression recovered to normal levels and the cells grew normal neurites with the timing and morphological characteristics of normal nerve growth factor-induced outgrowth, indicating that the blockade was not because of non-specific toxic effects. These results indicate that microtubule-associated protein 1B is an essential component of the molecular mechanism underlying the formation of neuronal processes.

Reorganisation of the microtubular cytoskeleton by embryonic microtubule-associated protein 2 (MAP2c)

Microtubule-associated protein 2c (MAP2c) is one of a set of embryonic MAP forms that are expressed during neuronal differentiation in the developing nervous system. We have investigated its mode of action by expressing recombinant protein in non-neuronal cell lines using cell cDNA transfection techniques. At every level of expression, all the MAP2c was bound to cellular microtubules. At low MAP2c levels, the microtubules retained their normal arrangement, radiating from the centrosomal microtubule-organising centre (MTOC) but at higher levels an increasing proportion of microtubules occurred independently of the MTOC. In most cells, radially oriented microtubules still attached to the MTOC co-existed with detached microtubules, suggesting that the primary effect of MAP2 is to increase the probability that tubulin polymerisation will occur independently of the MTOC. The MTOC-independent microtubules formed bundles whose distribution depended on their length in relation to the diameter of the transfected cell. Short bundles were attached to the cell cortex at one end and followed a straight course through the cytoplasm, whereas longer bundles followed a curved path around the periphery of the cell. By comparing these patterns to those produced by two chemical agents that stabilise microtubules, taxol and dimethyl sulphoxide, we conclude that effects of MAP2c arise from two sources. It stabilises microtubules without providing assembly initiation sites and as a result produces relatively few, long microtubule bundles. These bend only when they encounter the restraining influence of the cortical cytoskeleton of the cell, indicating that MAP2c also imparts stiffness to them. By conferring these properties of stability and stiffness to neuronal microtubules MAP2c contributes to supporting the structure of developing neurites.

Microtubule-associated protein 2 (MAP2)-immunoreactive neurons in the retina of Bufo marinus: colocalisation with tyrosine hydroxylase and serotonin in amacrine cells

Neuron populations in the retina of the toad, Bufo marinus, were labelled with a monoclonal antibody raised against microtubule-associated protein 2 (MAP2). A subpopulation of cones, probably corresponding to the blue-sensitive small single cones, large diameter amacrine cells in the most proximal row of the inner nuclear layer and some large ganglion cells in the ganglion cell layer were labelled. Double labelling experiments were carried out to establish the colocalisation of MAP2 with known putative transmitter substances of the anuran amacrine cells. MAP2 was colocalised in a subpopulation of serotonin-immunoreactive and in all tyrosine hydroxylase-immunoreactive amacrine cells. The results indicate, that the MAP2 content in the neurons of the anuran retina can be correlated with other well-defined neurochemical and/or physiological properties.

Abnormal expression of two microtubule-associated proteins (MAP2 and MAP5) in specific subfields of the hippocampal formation in schizophrenia

A variety of cytoarchitectural disturbances have been described in limbic regions in postmortem studies of schizophrenia, many of which suggest a developmental disturbance of normal neuronal geometry. This geometry is established and maintained by elements of the neuronal cytoskeleton. Immunohistochemistry with a panel of 15 monoclonal antibodies was used to monitor the presence of neuronal cytoskeletal proteins in the hippocampal formations of six patients with schizophrenia, six normal controls, and six with neurodegenerative disorders. In five of the six subjects with schizophrenia, prominent and specific alterations were found in the distribution of two microtubule-associated proteins, MAP2 and MAP5, which were anatomically selective for the subiculum and entorhinal cortex. In contrast, the immunoreactivity of other cytoskeletal proteins (i.e., tau, tubulins, and selected neurofilament protein phosphoisoforms) was similar for all subjects. Defects in the expression of MAP2 and MAP5, two proteins that contribute to the establishment and maintenance of neuronal polarity, could underlie some of the cytoarchitectural abnormalities described in schizophrenia and impair signal transduction in the affected dendrites. The subiculum and entorhinal cortex interconnect the hippocampal formation with widespread cortices and subcortical nuclei and play important roles in higher cognitive functions. Hence, pathologic lesions that distort the polarized geometry of neurons could play a role in the emergence of aberrant behavior in schizophrenia.

Differential Distribution of Tau Proteins in Developing Cat Cerebral Cortex and Corpus Callosum

During the postnatal development of cat visual cortex and corpus callosum the molecular composition of tau proteins varied with age. In both structures, they changed between postnatal days 19 and 39 from a set of two juvenile forms to a set of at least two adult variants with higher molecular weights. During the first postnatal week, tau proteins were detectable with TAU-1 antibody in axons of corpus callosum and visual cortex, and in some perikarya and dendrites in the visual cortex. At later ages, tau proteins were located exclusively within axons in all cortical layers and in the corpus callosum. Dephosphorylation of postnatal day 11 cortical tissue by alkaline phosphatase strongly increased tau protein immunoreactivity on Western blots and in numerous perikarya and dendrites in all cortical layers, in sections, suggesting that some tau forms had been unmasked. During postnatal development the intensity of this phosphate-dependent somatodendritic staining decreased, but remained in a few neurons in cortical layers II and III. On blots, the immunoreactivity of adult tau to TAU-1 was only marginally increased by dephosphorylation. Other tau antibodies (TAU-2, B19 and BR133) recognized two juvenile and two adult cat tau proteins on blots, and localized tau in axons or perikarya and dendrites in tissue untreated with alkaline phosphatase. Tau proteins in mature tissue were soluble and not associated with detergent-resistant structures. Furthermore, dephosphorylation by alkaline phosphatase resulted in the appearance of more tau proteins in soluble fractions. Therefore tau proteins seem to alter their degree of phosphorylation during development. This could affect microtubule stability as well as influence axonal and dendritic differentiation.

Neuronal polarity: targeting of microtubule components into axons and dendrites

The functional polarity of nerve cells depends on the outgrowth of both axons and dendrites. These processes, which were distinguished by morphological and physiological criteria, have been shown in recent years to differ in molecular composition, including their cytoskeleton. The asymmetric distribution of cytoskeletal elements and, particularly, the segregation of microtubule-associated proteins by their differential transport, may play an important role in the assembly of distinct microtubules in the two neuronal domains. An additional mechanism to achieve this subcellular localization is the transport of specific mRNAs to allow the local synthesis of specific proteins close to their functional site. This may endow the cell with a rapid mechanism for the regulation of synthesis under special conditions, which may be important during neuronal development and plasticity.

Developmental changes in phosphorylation of MAP-2 and synapsin I in cytosol and taxol polymerised microtubules from chicken brain

In cytosol, cyclic AMP stimulated phosphorylation of microtubule associated protein-2 (MAP-2) increased from 2 days to adult in proportion to the increase in the concentration of MAP-2. By contrast, the calmodulin stimulated phosphorylation of MAP-2 decreased in proportion to the decrease in the concentration of calmodulin stimulated protein kinase II (CMK II). Similarly, the cAMP stimulated phosphorylation of the site on synapsin I labeled by the cAMP stimulated protein kinase (PKA) changed little during development whereas the calcium/calmodulin stimulated phosphorylation of the CMK II site decreased dramatically in proportion to the decrease in the concentration of CMK II. The decrease in the concentration of CMK II which occurs in cytosol during synapse maturation was also observed in taxol polymerised microtubules and the effects of the change in the relative concentrations of CMK II and PKA on the phosphorylation of MAP-2 and synapsin I in this fraction were similar to that observed in the cytosol. These results are consistent with the hypothesis that the developmental changes in phosphorylation of endogenous substrates by PKA is controlled largely by changes in the concentration of those substrates, whereas the concentration of CMK II is limiting so that the developmental changes in the phosphorylation of endogenous substrates by CMK II are a function of the concentration of CMK II itself as well as the concentration of endogenous substrates. Some possible functional consequences of this during synapse maturation are discussed.

Tau in Alzheimer neurofibrillary tangles. N- and C-terminal regions are differentially associated with paired helical filaments and the location of a putative abnormal phosphorylation site

To investigate the extent to which whole tau proteins, structurally abnormal tau and fragments of tau are incorporated into neurofibrillary tangles in Alzheimer's disease, an immunocytochemical mapping study using a panel of antibodies to several synthetic human tau peptides has been performed. Neurofibrillary tangles were immunolabelled in situ, and paired helical filaments (PHF), the principal structural component of tangles, were immunolabelled after isolation and Pronase treatment. N-Terminal and C-terminal domains of tau were found to be present in tangles in situ. SDS-treated PHF were found to contain most of the C-terminal half of tau and were also labelled by antibodies to ubiquitin. Only some of these PHF were labelled by antisera to tau sequences towards the N-terminus, and this enabled the identification of a region of tau in which proteolytic cleavage may occur. The ultrastructural appearance of the immunolabelling suggested that both the N- and C-terminal domains of tau extend outwards from the axis of PHF. After Pronase treatment. PHF were strongly labelled only by an antiserum to PHF and by the antiserum to the most C-terminal tau synthetic peptide. The latter antiserum also strongly labelled extracellular tangles in situ, whereas these extracellular tangles were poorly labelled by the antisera to the other synthetic peptides. One anti-(tau peptide) serum labelled a population of neurofibrillary tangles in situ only after alkaline phosphatase pretreatment of tissue sections. Our results show that, although peptides along the length of the tau molecule are associated with neurofibrillary tangles in situ, only the C-terminal one-third of the molecule is tightly associated with PHF, since this region of tau is resistant to SDS treatment of PHF. We also report the existence in PHF in situ of a masked tau epitope which is partially unmasked by dephosphorylation. These results are indicative of post-translational changes in tangle-associated tau in degenerating neurons in Alzheimer's disease.

Microtubule-associated protein-2 and neurofilament immunoreactivity in neurons and small, intensely fluorescent cells of an amphibian cardiac ganglion

The localization of two cytoskeletal proteins was analysed in the cell bodies and processes of ganglionic neurons and small, intensely fluorescent cells of the parasympathetic cardiac ganglion of Necturus maculosus (mudpuppy). Antibodies against microtubule-associated protein-2 and against the highly phosphorylated isoforms of high and middle molecular weight neurofilament subunits were used as somatodendritic and axonal markers, respectively. The ganglionic neurons, which usually have only one major process, and small, intensely fluorescent cells, which have several processes, showed distinctly different staining patterns with the two antibodies. In control and denervated ganglia, the ganglionic cell bodies and several hundred micrometers of the proximal processes were labeled with the antibody against microtubule-associated protein-2, whereas small, intensely fluorescent cells and processes showed a paucity of immunoreactivity. The neurofilament antibody labeled numerous axons in the ganglion but did not label the proximal part of the postganglionic process or small, intensely fluorescent cell processes. Denervation resulted in the presence of phosphorylated neurofilament subunit immunoreactivity in the soma and proximal process of the ganglionic neuron. These data suggest that (i) small, intensely fluorescent cells and ganglionic neurons in the mudpuppy cardiac ganglion contain distinctly different cytoskeletal proteins, (ii) the proximal part of postganglionic "axons" contains dendrite-like and not axon-like cytoskeletal proteins, and (iii) deafferentation promotes the localization of phosphorylated forms of neurofilament subunits in the soma and proximal process of parasympathetic ganglionic neurons.

Identification of nuclear tau isoforms in human neuroblastoma cells

The tau proteins have been reported only in association with microtubules and with ribosomes in situ, in the normal central nervous system. In addition, tau has been shown to be an integral component of paired helical filaments, the principal constituent of the neurofibrillary tangles found in brains of patients with Alzheimer disease and of most aged individuals with Down syndrome (trisomy 21). We report here the localization of the well-characterized Tau-1 monoclonal antibody to the nucleolar organizer regions of the acrocentric chromosomes and to their interphase counterpart, the fibrillar component of the nucleolus, in human neuroblastoma cells. Similar localization to the nucleolar organizer regions was also observed in other human cell lines and in one monkey kidney cell line but was not seen in non-primate species. Immunochemically, we further demonstrate the existence of the entire tau molecule in the isolated nuclei of neuroblastoma cells. Nuclear tau proteins, like the tau proteins of the paired helical filaments, cannot be extracted in standard SDS-containing electrophoresis sample buffer but require pretreatment with formic acid prior to immunoblot analysis. This work indicates that tau may function in processes not directly associated with microtubules and that highly insoluble complexes of tau may also play a role in normal cellular physiology.

Microtubule-associated protein 2 (MAP2) in Purkinje cell dendrites: evidence that factors other than binding to microtubules are involved in determining its cytoplasmic distribution

We have studied the distribution of microtubule-associated protein 2 (MAP2) in the Purkinje cell dendrites of rats whose cerebella were exposed to X-irradiation during the second postnatal week. The Purkinje cells of such animals have abnormally elongated apical primary processes that branch in the other molecular layer rather than close to the cell body as in normal tissue. The results show that in these distorted dendrites the MAP2 distribution is "shifted" distally relative to the normal pattern, in which MAP2 is distributed evenly throughout the dendritic tree. Tubulin and other microtubule-associated proteins, such as MAP1, are not affected and remain evenly distributed throughout the dendritic tree despite the anatomical distortion. We conclude that the distribution of MAP2 in Purkinje cells is not determined solely by its binding to tubulin. Other factors must be involved and these appear to be related to dendritic morphology and possibly to branching.

The roles of microtubule-associated proteins in brain morphogenesis: a review

Microtubule-associated proteins (MAPs) are a diverse family of cytoskeletal proteins that copurify with tubulin in vitro. Recently a number of novel approaches have been used to learn more about the functions of MAPs during brain development, including: localization of MAPs and their mRNA in the developing brain, comparisons of MAPs between species to learn potential fundamental characteristics, biochemical analysis of changes in MAPs in process-bearing cell lines, and sequence analysis of MAP cDNAs and cDNA transfection studies. Taken together, these data allow us to assign roles to MAPs which are abundant in the developing brain, and to develop models for future studies. Four MAPs are particularly abundant in the developing brain: MAP1B, the high and low-molecular weight forms of MAP2, and juvenile tau. MAP1B is the only MAP to be found consistently in extending processes in both the developing and adult brain, making it a likely regulator of neurite outgrowth. High-molecular weight MAP2 and tau crosslink microtubules in dendrites and axons, respectively. Low-molecular weight MAP2 may be able to regulate MAP2-mediated crosslinking to make processes more labile during development and in adult brain regions where synaptogenesis is active. Tau-mediated crosslinking may be regulated by temporal regulation of the expression of tau forms with different binding affinities to tubulin. High-molecular weight MAP2 is sequestered into dendrites by the selective transport of its mRNA. This allows rapid and local regulation of MAP2 synthesis.

Effect of excitatory amino acids on microtubule-associated proteins in cultured cortical and spinal neurones

The effect of excitatory amino acid stimulation on the cytoskeleton of cultured spinal cord and cortical neurons was monitored with antibodies against microtubule-associated proteins tau and MAP2. In unstimulated cultures tau-1 immunoreactivity was restricted to axon-like processes. Stimulation with glutamate (0.1-1 mM) or N-methyl-D-aspartate (NMDA) (0.1 mM) resulted in a dramatic increase in the intensity of tau labelling in axons and the appearance of staining within a proportion of neuronal cell bodies and dendrites. Quisqualate or kainate stimulation resulted only in an increase in tau immunoreactivity within axons. The NMDA mediated events were calcium dependent and the effects of all excitatory amino acids could be blocked by specific antagonists. In contrast, following stimulation with excitatory amino acids, MAP2-immunoreactivity was associated with filaments which formed a complex network within the cell body. This suggests that the different excitatory amino acid receptor subtypes can have differential effects on the neuronal cytoskeleton.

The expression of phosphorylated and non-phosphorylated forms of MAP5 in the amphibian CNS

MAP5 is a microtubule-associated protein that in rat and quail is more abundant in the developing than in the adult brain. Previous studies in our laboratory have shown that MAP5 can be resolved into two forms by SDS-PAGE, MAP5a and MAP5b (Mr 300,000-320,000 Da) with MAP5a representing a highly phosphorylated form of this protein. In the present study, the relationship between MAP5 expression and neuronal growth and plasticity was investigated by assessing the amount and distribution of MAP5a and MAP5b in both the developing Xenopus brain and in different regions of the adult brain where neurons of varying growth potential and plasticity are present. In the larval and metamorphic Xenopus brain, like the neonatal rat brain, MAP5 is present in the highly phosphorylated form, MAP5a, and in concentrated in neuronal processes. In the adult Xenopus brain, MAP5a remains high in the optic tectum but, like the situation in the adult rat brain, is undetectable in the telencephalon. Immunohistochemistry showed that MAP5 was concentrated in the outer layer of the tectum, where ingrowing and plastic retinal ganglion cell axons are found. The correlation between MAP5 expression and phosphorylation and growth potential suggests that this molecule plays an important role in the regulation and organization of the neuronal cytoskeleton during neurite outgrowth and plasticity.

Hyperammonemia induces brain tubulin

We have developed an animal model of hyperammonemia consisting of feeding rats a diet containing ammonium acetate. Using this model we have found that hyperammonemia induces tubulin synthesis in brain. Initially tubulin accumulates rapidly (28% after 2 days on diet) and continues increasing but at a slower rate, reaching a 50% increase after 100 days on the diet. The effect is reversible, rats fed the ammonium diet return to normal levels of tubulin two days after withdrawal of the ammonium diet. In contrast to the effect on brain, hyperammonemia did not increase tubulin content in liver or kidney. Moreover, the effect on brain is selective, with maximum increases of tubulin content in hippocampus, septum and reticular formation while other areas such as locus coeruleus and mammillary nucleus are not affected at all. The results presented show that the induction of tubulin is a consequence of an increased polymerization of microtubules which in turn is due to an altered phosphorylation of microtubule-associated proteins.

Monoclonal antibody markers for amphibian oligodendrocytes and neurons

Few immunocytochemical probes have been developed for cold-blooded vertebrates, thus hampering analyses of cellular processes in these species. Those developed from mammalian and avian tissue often fail either to react or to show similar specificities in poikilotherms. Therefore, we have begun raising monoclonal antibodies (mabs) in mice against frog and tadpole brain tissue. The following analyses of two of these mabs suggest that these antibodies represent specific probes for frog axons and oligodendrocytes. Mab Olig recognizes all the myelinated axon tracts of the mature frog brain and spinal cord, as well as the tracts of the developing tadpole CNS once they have become myelinated. Axons cut in cross section show characteristic o-shaped staining around individual axons when processed with this antibody. Particularly easy to visualize in the tadpole are immunoreactive cell bodies and processes, seen in continuity with the myelin sheath. Occasionally, in this developing tissue, cells with highly branched processes characteristic of immature oligodendrocytes are observed. No other cells or processes within the brain or spinal cord react with this antibody. Mab Linc stains numerous filaments in all axonal projections. Occasionally, a thin rim of filamentous staining is observed in cell somata, but many regions rich in neuronal somata or dendrites are unreactive to this antibody. This in vivo staining pattern suggests that the Linc antigen is differentially distributed within neurons and exhibits a high concentration in axons. Linc immunoreactivity is robust in the processes of a subpopulation of dissociated tectal cells in culture. These Linc-positive cells are characterized as neurons on morphological criteria. Also, intense Linc immunoreactivity is present in the bundles of retinal axons that extend from retinal explants. Olig immunoreactivity, however, has not been detected in tectal cultures or retinal explants. Improved staining following Triton X-100 treatment of tissue sections suggests that neither of the mabs recognizes lipid antigens and that both are probably localized within the cell cytoplasm. Only the Linc mab reacts on Western blots of denatured brain protein. Linc consistently recognizes two Triton X-100-insoluble proteins with apparent molecular weights of 56 and 58 kD. The epitopes recognized by the Olig and Linc mabs have been surveyed in terms of their resistance to optic nerve crush and their consequent value in studies requiring such procedures. Possible homologies to known cell-type-specific molecules are discussed.

Embryonic MAP2 lacks the cross-linking sidearm sequences and dendritic targeting signal of adult MAP2

The most prominent microtubule-associated protein of the neuronal cytoskeleton is MAP2. In the brain it exists as a pair of high-molecular weight proteins, MAP2a and MAP2b, and a smaller form, MAP2c, which is particularly abundant in the developing brain. High-molecular weight MAP2 is expressed in dendrites, where its messenger RNA is also located, but is not found in axons; it has been shown to be present in fine filaments that crosslink dendritic microtubules. This correlates with the primary structure of high-molecular weight MAP2, which consists of a short carboxy-terminal tubulin-binding domain and a long amino-terminal arm, which forms a filamentous sidearm on reconstituted microtubules. Here we report that the high- and low-molecular weight forms of MAP2 are generated by alternative splicing and share the entire C-terminal tubulin-binding domain as well as a short N-terminal sequence. In contrast to high molecular weight MAP2, embryonic brain MAP2c lacks 1,342 amino acids from the filamentous sidearm domain. Furthermore, the mRNA for low molecular weight MAP2c is not present in dendrites, indicating that the dendritic targeting signal is specific for the high-molecular weight form.