Search

Comparative study of CSF levels of tau and AD7C-neuronal thread protein (NTP) in patients with AD and control subjects.AD is characterized by neurofibrillary tangles composed of the abnormally hyperphosphorylated microtubule-associated protein tau. AD7C-NTP is a proposed AD marker expressed at early stages of neurofibrillary degeneration.Enzyme-linked immunosorbent assays specific for tau and AD7C-NTP. CSF samples were obtained from 35 demented patients (25 with antemortem clinical diagnosis of probable AD, 5 with neuropathologic diagnosis of definite AD, 5 with Lewy body pathology), 29 nondemented patients with PD, and 16 elderly healthy control subjects. Receiver operating characteristics (ROC) and multivariate discriminant analysis for AD versus controls. Correlational analysis of CSF tau and AD7C-NTP and of each marker with Mini-Mental State Examination (MMSE) scores was performed.Levels of both tau and AD7C-NTP were significantly elevated in the AD patients compared with control subjects. ROC analysis showed that CSF tau distinguished between patients with AD and nondemented control subjects with 63% sensitivity and 89% specificity, AD7C-NTP with 70% sensitivity and 87% specificity. Combined evaluation of both markers with discriminant analysis raised the specificity to 93% at a 63% sensitivity level. Both markers positively correlated with each other within the AD group, but not among control subjects. CSF levels of AD7C-NTP, but not of tau, showed a small but significant inverse correlation (r = -0.43) with MMSE scores of AD patients.CSF levels of tau and AD7C-NTP may be useful biomarkers for AD.

The role of the low affinity nerve growth factor receptor (p75(NGFR)) in NGF-mediated signaling is not yet understood. Here we show by co-immunoprecipitation that NGF activates a protein kinase that is directly associated with p75(NGFR) in dorsal root ganglion (DRG) cells and PC12 cells in culture. Two proteins of 120 and 104 kDa constitute the majority of this activity. In PC12 cells, TrkA activation was necessary to elicit p75(NGFR)-associated kinase activity. Although NGF binding to p75(NGFR) was not necessary for kinase activation, it accelerated the activation of the kinase at low NGF concentrations. Deletion analysis showed that a 43 amino acid region in the cytoplasmic domain of p75(NGFR) was responsible for this effect. These findings show that p75(NGFR) accelerates TrkA-mediated signaling and, in addition, demonstrate that p75NGFR and TrkA collaborate to activate a previously undescribed p75(NGFR)-associated protein kinase.

Mutations affecting the peripheral myelin protein-22 (PMP22) gene have been shown to be associated with inherited peripheral neuropathies. To provide the molecular basis for the analysis of such mutations, we have cloned and characterized the human PMP22 gene. It spans approximately 40 kilobases and contains four coding exons. Detailed analysis of its 5‘-flanking region suggested the presence of two alternatively transcribed, but untranslated exons. Mapping of separate PMP22 mRNA transcription initiation sites to each of these exons indicates that PMP22 expression is regulated by two alternatively used promoters. In support of this hypothesis, both putative promoter sequences demonstrated the ability to drive expression of reporter genes in transfection experiments. Furthermore, the structures of the 5‘-portions of the PMP22 genes appear to be identical in rat and human, supporting the biological significance of the observed arrangement of regulatory regions. The relative expression of the alternative PMP22 transcripts is tissue-specific, and high levels of the exon 1A-containing transcript are tightly coupled to myelin formation. In contrast, exon 1B-containing transcripts are predominant in non-neural tissues and in growth-arrested primary fibroblasts. Interestingly, although a strong upregulation of PMP22 mRNA was observed in cultured Schwann cells in the presence of the adenylate cyclase activator forskolin under various culture conditions, the regulation of the different PMP22 mRNA species did not mimic the regulation that occurs during myelin formation in vivo. The observed regulation of the PMP22 gene by a complex molecular mechanism is consistent with the proposed dual role of PMP22 in neural and non-neural tissue.

Peripheral myelin protein, 22 kDa (PMP22), is a myelin molecule associated with Schwann cells in peripheral nerves (Snipes, G. J., Suter, U., Welcher, A. A., and Shooter, E. M. (1992) J. Cell Biol. 117, 225-238). Mutations affecting the PMP22 gene have been implicated in the trembler mutation in mice (Suter, U., Welcher, A. A., Ozcelik, T., Snipes, G. J., Kosaras, B., Francke, U., Billings-Gagliardi, S., Sidman, R. L., and Shooter, E. M. (1992) Nature 356, 241-244; Suter, U., Moskow, J. J., Welcher, A. A., Snipes, G. J., Kosaras, B., Sidman, R. L., Buchberg, A. M., and Shooter, E. M. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 4382-4386) and Charcot-Marie-Tooth Disease in humans (Patel, P. I., Roa, B. B., Welcher, A. A., Schoener-Scott, R., Trask, B. J., Pentao, L., Snipes, G. J., Garcia, C. A., Francke, U., Shooter, E. M., Lupski, J. R., and Suter, U. (1992) Nature genet. 1, 159-165). In this report, we have studied PMP22 production in cultured rat Schwann cells. Schwann cells contain a 1.8-kilobase mRNA transcript coding for PMP22, and its production is up-regulated in vitro by forskolin. Metabolic labeling combined with immunoprecipitation methods using antibodies raised against synthetic peptides of PMP22 reveal that Schwann cells generate the protein from an 18-kDa precursor form which is post-translationally modified by N-linked glycosylation. A second molecule (molecular mass, 48 kDa) that reacted with PMP22 antibodies was also detected in Schwann cells but is not related chemically to PMP22 as determined by pulse-chase labeling. Metabolic labeling of rat sciatic nerve and Western blot analyses of purified rat sciatic nerve myelin reveal that deglycosylation of PMP22 gives rise to an 18-kDa protein similar in size to that in Schwann cells. These results indicate that cultured Schwann cells may provide a good model in which to investigate the production and function of PMP22 and to establish the cellular basis for the protein's involvement in inherited peripheral neuropathies.

Neurotrophins (NTs) play an important role in the modulation of synaptic transmission and in morphological changes in synaptic structures. Although there is agreement that brain-derived neurotrophic factor (BDNF) is sorted to large dense-core vesicles (LDCVs) and released via the regulated secretory pathway, there has been some dispute regarding the mode of secretion of nerve growth factor (NGF) and neurotrophin-3 (NT-3), two structurally related members of the NT family. In this study, we examined the subcellular localization and release characteristics of NGF, BDNF, and NT-3 in adenovirus-infected primary cortical neurons. We found that all members of the NT family colocalized with markers for the endoplasmic reticulum and Golgi within cell bodies and in a punctate manner with a marker for LDCVs within processes. Moreover, their release was triggered by depolarization, indicating that NGF, BDNF, and NT-3 are released via the regulated secretory pathway. When neurons were coinfected with two separate adenoviruses coding for NGF or BDNF, both NTs showed almost complete vesicular colocalization within single cells, suggesting that different NTs might be packaged into shared vesicles. We also examined whether the two splice variants of NGF, the short and long precursors, differ in their release characteristics. We found that neurons infected with viruses coding for either splice variant released NGF in a regulated way. Overall, our study supports the notion that all members of the NT family undergo activity-dependent regulated release from neurons, enabling them to act as "synaptotrophins" on electrically active neurons.

Nerve growth factor (NGF) regulates the activity of the transcription factor NF-kappaB (nuclear factor-kappaB) through its low affinity receptor, p75. In the present study we found that NGF binding to p75 induces nuclear translocation of p65 and increases NF-kappaB binding activity in a cell line overexpressing p75, but only after the cells have been subjected to a previous stress. Under physiological conditions, in the absence of stress, NGF is unable to alter p65 nuclear levels. Tumor necrosis factor-alpha (TNF-alpha) induces a down-regulation of IkappaB-alpha, -beta and -epsilon both in physiological and in stress, i.e. serum-free, conditions. In contrast, NGF only induces the specific degradation of IkappaB-beta after serum withdrawal, without affecting IkappaB-alpha or -epsilon either in the presence or in the absence of stress. IkappaB-beta consists of several isoforms, whose relative abundance is regulated by serum withdrawal. NGF does not target all the IkappaB-beta isoforms with the same potency, being more effective in reducing the levels of the isoforms up-regulated by serum withdrawal. TRAF-6 is expressed at the same level under both physiological and stress conditions. These results indicate that NGF is able to induce NF-kappaB nuclear translocation by a mechanism that involves specific IkappaB-beta degradation only after the cells have been subjected to a severe stress.

The short- and long-term effects of nerve growth factor (NGF) were studied on fibroblast cell lines stably expressing both TrkA and either wild-type p75 or a mutant that lacks the palmitoylation site of p75. The lack of palmitoylation had no effect on the ability of p75 to enhance the short-term NGF-induced tyrosine phosphorylation of TrkA over a wide range of NGF concentrations. Long-term treatment of the cell lines with NGF led to loss of cell adhesion to the culture dishes that increased with increasing concentrations of NGF and increased expression of TrkA. Treatment of the cell lines with mutant NGFs that bound selectively to TrkA or p75 alone revealed that cell detachment was mediated solely through TrkA. Increased cell detachment correlated with a decrease in the expression levels of fibronectin and cadherin, cell surface molecules involved in cell adhesion. The loss of cell adhesion with the cell line expressing the palmitoylation-deficient p75 were identical to those expressing wild type, as was anticipated from the lack of involvement of p75 in this process.

Peripheral myelin protein 22 (PMP22) was initially described as a minor component of peripheral myelin. Mutations affecting the PMP22 gene cause demyelinating neuropathies, supporting a role for the protein in PNS myelination. Furthermore, PMP22 carries the L2/HNK-1 carbohydrate epitope suggesting an adhesion/recognition function. Despite advances in characterizing the PMP22 gene, the specific role(s) of the protein in myelin remains unknown. In this study we determined the temporal expression pattern of PMP22 in comparison to galactocerebroside (GalC) and myelin associated glycoprotein (MAG), early constituents of PNS myelin, and to protein zero (P0) and myelin basic protein (MBP), late components of myelin. In sciatic nerve lysates, PMP22 was detected at postnatal day 3, after MAG, but before MBP expression. The same results were obtained in cocultures of dorsal root ganglion neurons and Schwann cells (SCs). Low levels of PMP22 were found in early, anti-MAG and anti-GalC immunoreactive, myelinating cocultures. However, PMP22 could only be detected in the SC plasma membrane after basal lamina formation. In long-term myelinating cocultures PMP22 levels continued to increase and the protein was found in anti-P0 and anti-MBP immunoreactive myelin segments. Furthermore, PMP22, MBP, and P0 protein levels were greatly enhanced by progesterone treatment of the cocultures. The highest levels of PMP22 expression were associated with late stages of myelination; however the presence of the protein in nonmyelinating SCs and in SCs commencing myelination supports multiple roles for PMP22 in peripheral nerve biology.

The role of the low affinity neurotrophin receptor p75LNTR in neurotrophin signal transduction remains open. Recent reports show that this receptor generates intracellular signals independent of Trk activity, and others imply that it collaborates with Trk(s) to enhance cellular responses to low neurotrophin concentrations. We have used the Cytosensor microphysiometer as a direct marker of intracellular metabolic activity to address the physiologic role of p75LNTR in nerve growth factor (NGF) signal transduction. NGF treatment of PC12 or TrkA-transfected Chinese hamster ovary (CHO) cells results in a rapid, transient increase in the extracellular acidification rate as measured by the Cytosensor; in both cell types, p75LNTR enhances this response. p75LNTR affects both the magnitude of and the duration of the extracellular acidification response to NGF. Moreover, it is not merely the presence of p75LNTR, but also the ratio of p75LNTR:TrkA which determines cellular responsiveness to NGF. In transiently transfected CHO cells, a 5:1 ratio of p75LNTR:trkA cDNAs produced the greatest change in NGF-induced acid secretion. Pretreatment of PC12 cells with anti-p75LNTR antibodies decreased the responsiveness to NGF. However, long-term NGF exposure to PC12 cells in which p75LNTR expression was decreased to approximately 10% of wild-type levels showed a longer duration of acid secretion compared to wild-type PC12 cells. Together, these data suggest that p75LNTR may play a dual role in modulating NGF signal transduction by enhancing and extending cellular responses to short-term ligand exposures while attenuating the metabolic response to long-term ligand exposures. With regard to potential Trk-independent p75LNTR signal transduction mechanisms, we detected no change in extracellular acidification response in 75LNTR-transfected CHO cells, PCNA-15 fibroblasts, or Schwann cells, all of which express large amounts of p75LNTR and no Trk. Thus, p75LNTR cannot produce any signal detected by microphysiometry in the absence of TrkA.

A growing body of evidence suggests that neurotrophins (NTs) play a critical role in synaptic plasticity and other activity dependent processes in the CNS. Release of these growth factors by neurons and neuroendocrine cells was recently shown to occur via the regulated secretory pathway, representing a possible mechanism for preferentially supplying NTs locally to active synapses. However, the identity and characteristics of the intracellular storage compartment for NTs undergoing stimulus-coupled secretion remains controversial. As a step towards addressing these issues we have investigated the subcellular localization of epitope-tagged nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and neurotrophin-3 (NT-3) in neuroendocrine cells. Placement of the myc-epitope tag at the neurotrophin carboxy terminus did not affect essential properties of the NTs such as their ability to induce Trk tyrosine phosphorylation or their sorting into the regulated secretory pathway in PC12 and AtT-20 neuroendocrine cells. Epitope-tagged NTs colocalize with dense core vesicle (DCV)-markers at the light microscopic level in both cell lines investigated. Furthermore, at an EM level immunoreactivity (IR) for myc-tagged NGF was found over dense core granules (DCGs) in PC12 cells. These data provide evidence that NTs can be stored in DCVs in neuronal model cell lines and, potentially, in neurons as well.

In order to obtain the basic knowledge necessary to develop therapeutical intervention for blindness due to the damaged retina and optic nerve, the mechanism of retinal degeneration and regeneration in an amphibian model, Cynops pyrrhogaster, was studied. In the retinal degenerative process following enucleation and reimplantation of the eye ball, evidence was found for active cell death of neural retinal cells. As the degeneration proceeded, Musashi, an ribonucleic acid (RNA)-binding protein, started its expression in the daughter cells of proliferating retinal pigment epithelium (RPE) cells, messenger RNA (mRNA) expression of proneural genes with basic helix-loop-helix motif was then detected in the newly developing retina. These results suggest that transdifferentiation of RPE cells to neural retina involves at least partial cascade, if not entirely, of neural induction from uncommitted ectodermal tissue. Search for genes that are required for transdifferentiation of RPE cells to neural retinal cells, in addition to those mentioned above, will provide the basic knowledge for successful retinal transplantation and retinal regeneration in higher vertebrates.

The neurotrophin receptor TrkB plays a key role in promoting cell survival and differentiation in the nervous system. Two adhesive motifs in the extracellular domain of TrkB have been proposed based on its predicted secondary structure. To investigate the potential adhesive function of trkB, a full length trkB cDNA was stably transfected into NIH 3T3 cells and TrkB-expressing clones isolated. Transfectant clones producing different levels of TrkB protein were subjected to a homotypic aggregation assay. Results showed that parental cells were non-adhesive during the assay while TrkB-expressing cells displayed varying degrees of aggregation depending on the amount of TrkB protein expressed. The observed adhesion was Ca(2+)-, Mg(2+)-, and temperature-dependent, characteristics shared by the cadherin family of adhesion molecules. The transfected cell lines also expressed cadherin in proportion to TrkB expression and both molecules were required for cell adhesion. Double immunofluorescence staining studies showed that TrkB was colocalized with cadherin and catenin at cell-cell contact sites. Whether TrkB and cadherin mediate adhesion separately or synergistically remains to be determined.

Peripheral nerve trauma induces the expression of genes presumed to be involved in the process of nerve degeneration and repair. In the present study, an in vivo paradigm was employed to identify molecules which may have important roles in these processes. A cDNA library was constructed with RNA extracted from rat dorsal root ganglia (DRG) 3 days after a sciatic nerve crush. After differential hybridization to this library, several cDNAs were identified that encoded mRNAs that were upregulated in the DRG ipsilateral to the crush injury, as opposed to the contralateral or naive DRG. Approximately 0.15% of all the clones screened were found to be induced. This report presents the types of induced sequences identified and characterizes one of them, DA11. The 0.7 kb DA11 full length cDNA clone contains a 405 nucleotide open reading frame that encodes a putative protein of 15.2 kDa (135 amino acid residues) and is a member of the family of fatty acid binding proteins (FABP). The DA11 protein differs by one amino acid residue from the sequence of the C-FAPB protein and by eight residues from the sequence of mal1, proteins found in rat and mouse skin, respectively. Northern and Western blot analyses showed that the DA11 mRNA and protein were induced in the injured DRG. Furthermore, studies using antibodies generated against DA11 found that the DA11-like immunoreactivity was more pronounced in the nuclei of neurons located in the DRG ipsilateral to the sciatic cut than those located in the contralateral DRG. The induction of DA11 mRNA and protein in DRG neurons suggests, for the first time, the involvement of a neuronal FABP in the process of degeneration and repair in the nervous system.

The low affinity neurotrophin receptor, termed p75LNTR, plays a role in increasing the amount of nerve growth factor that becomes bound to the tyrosine kinase receptor, trkA (Barker, P. A., and Shooter, E. M. (1994) Neuron 13: 203-215), possibly by increasing the nerve growth factor concentration in the microenvironment surrounding the trkA receptor. Because protein acylation may be a means by which cell surface receptor distribution may be regulated, we have determined the acylation status of p75LNTR. We find that p75LNTR expressed in PC12, PCNA, or transfected COS cells is metabolically labeled with [3H]palmitic acid. This modification occurs post-translationally, and the incorporated fatty acid is removed by hydroxylamine treatment at pH 7 or 11 and by sulfhydryl reducing agents, suggesting a thioester linkage to palmitic acid. Consistent with this, p75LNTR in which the juxtamembrane cysteine present at position 279 is substituted with alanine is expressed but cannot be metabolically labeled with [3H]palmitic acid. Substitution of other cysteine residues present in the transmembrane or intracellular domain of the receptor has no effect on protein acylation, suggesting that only Cys279 is esterified to palmitate.

The binding domain of the low affinity nerve growth factor receptor (p75NGFR) is built from four "cysteine repeats," with almost identical patterns of half-cystine residues. To study the pattern of disulfide bridging within each cysteine repeat, we have mutated pairs of cysteine residues, primarily in the fourth repeat, and have tested the relative ability of each mutant to be cross-linked to the nerve growth factor after transient expression in COS 7 cells. The results give an indication of the relative importance of different disulfide bonds for the stability of the active conformation. Immunofluorescent staining shows that all of these proteins, even those that may be partially misfolded, are expressed to varying degrees on the surface of COS cells. They can also be detected by the monoclonal antibody MC192 on a Western blot following electrophoresis on a nondenaturing gel. Some cannot, however, be detected by MC192 on a Western blot after heat denaturation in SDS. The ability to refold under these conditions correlates in general with the ability of the mutant protein to bind nerve growth factor on the surface of COS cells.

p21ras is believed to be involved in the neuronal differentiation of cells responsive to nerve growth factor (NGF). We show that NGF stimulates the activation of p21ras in embryonic sensory neurons and in PC12 cells. In the initial 5 min of exposure to NGF, the activation is concentration-dependent. In the sensory neurons and PC12 cells, the apparent maximal activation was reached at 50 and 10 ng/ml, respectively, with half-maximal activation at approximately 5 and 2-3 ng/ml, respectively. Kinetic analysis at low concentrations of NGF showed that p21ras activation slowly increases with time in both types of cells, while high concentrations result in rapid activation within 5 min. These results indicate that NGF regulates the activation state of p21ras in these cells and provides evidence suggesting that activation of p21ras is involved in NGF signal transduction. Treatment of PC12 cells with brain-derived neurotrophic factor or neurotrophin-3 (NT-3) failed to activate p21ras, suggesting that binding alone to p75LNGFR is insufficient for ras activation. Treatment with the kinase inhibitor, K252a, which inhibits the NGF tyrosine kinase receptor p140trk, abolished ras activation, suggesting that p140trk is the major mediator of p21ras activation by NGF.

A series of mouse nerve growth factor (NGF) deletion mutants have been constructed using in vitro mutagenesis to define domains of the protein essential for its activity. Deletions of the amino or carboxyl termini of mature NGF or of an internal domain, which normally produces a surface-exposed reverse turn, have been analyzed. Mutants with deletions in the amino terminus or in the reverse turn retain significant biological activity, whereas, in contrast, a mutant NGF lacking the seven most carboxyl-terminal amino acids is appropriately synthesized but shows no measurable biological activity. These results suggests that the flexible carboxyl tail of NGF, and perhaps other neurotrophins, plays a crucial role in mediating receptor recognition and/or ligand binding.

Molecular genetic studies have established that mutations in the gene encoding the 22-kDa peripheral myelin protein (PMP-22) are responsible for hereditary peripheral neuropathies in the trembler mouse and in a subset of humans with Charcot-Marie-Tooth disease, type 1a. The function of the PMP-22 protein remains unknown. Several studies on myelin proteins in the PNS have indicated that the L2/HNK-1 epitope, which is believed to be both a ligand for cellular adhesion and a target for autoimmune monoclonal IgM neuritis, may be found on heretofore unidentified proteins with a molecular mass of 19-28 kDa. In this report, we provide immunological evidence that at least one of these proteins is PMP-22.

The trkA receptor functions as a signal transducing receptor for nerve growth factor. In this report, we show that alternative splicing results in the production of two distinct trkA isoforms in both rats and humans. These isoforms differ by virtue of a 6-amino acid insertion in their extracellular domain, the placement of which corresponds exactly with the breakpoint found in several human trkA oncogenes. When tested in fibroblasts, the presence (trkAII) or absence (trkAI) of the 6-amino acid insert does not affect the receptor's ligand binding specificity or its ability to transduce functional signals in response to nerve growth factor. In rats and humans, trkAII is the only isoform expressed within neuronal tissues at appreciable levels whereas trkAI, the form of trkA originally cloned, appears to be expressed mainly in non-neuronal tissues.

Peripheral myelin protein, 22 kDa (PMP22), is a myelin molecule associated with Schwann cells in peripheral nerves (Snipes, G. J., Suter, U., Welcher, A. A., and Shooter, E. M. (1992) J. Cell Biol. 117, 225-238). Mutations affecting the PMP22 gene have been implicated in the trembler mutation in mice (Suter, U., Welcher, A. A., Ozcelik, T., Snipes, G. J., Kosaras, B., Francke, U., Billings-Gagliardi, S., Sidman, R. L., and Shooter, E. M. (1992) Nature 356, 241-244; Suter, U., Moskow, J. J., Welcher, A. A., Snipes, G. J., Kosaras, B., Sidman, R. L., Buchberg, A. M., and Shooter, E. M. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 4382-4386) and Charcot-Marie-Tooth Disease in humans (Patel, P. I., Roa, B. B., Welcher, A. A., Schoener-Scott, R., Trask, B. J., Pentao, L., Snipes, G. J., Garcia, C. A., Francke, U., Shooter, E. M., Lupski, J. R., and Suter, U. (1992) Nature genet. 1, 159-165). In this report, we have studied PMP22 production in cultured rat Schwann cells. Schwann cells contain a 1.8-kilobase mRNA transcript coding for PMP22, and its production is up-regulated in vitro by forskolin. Metabolic labeling combined with immunoprecipitation methods using antibodies raised against synthetic peptides of PMP22 reveal that Schwann cells generate the protein from an 18-kDa precursor form which is post-translationally modified by N-linked glycosylation. A second molecule (molecular mass, 48 kDa) that reacted with PMP22 antibodies was also detected in Schwann cells but is not related chemically to PMP22 as determined by pulse-chase labeling. Metabolic labeling of rat sciatic nerve and Western blot analyses of purified rat sciatic nerve myelin reveal that deglycosylation of PMP22 gives rise to an 18-kDa protein similar in size to that in Schwann cells. These results indicate that cultured Schwann cells may provide a good model in which to investigate the production and function of PMP22 and to establish the cellular basis for the protein's involvement in inherited peripheral neuropathies.

Human neuroblastoma SH-SY5Y cells differentiate terminally in culture upon exposure to nerve growth factor (NGF) for 4-5 weeks. The neuronal phenotypic properties acquired in response to prolonged NGF treatment include morphological differentiation, cessation of mitotic activity, neuronal marker expression, increased membrane electrical potentials, and a survival dependence upon NGF for trophic support (Jensen, L.M. (1987) Dev. Biol. 120, 56-64). Thus, differentiated cultures survive indefinitely in the continued presence of NGF, however, withdrawal of NGF from differentiated cultures effects the loss of cellular viability within 3-6 days. Here, we show that death of differentiated SH-SY5Y cells caused by NGF deprivation is characteristic of apoptosis. To compare the differentiation promoting and the neurotrophic properties of NGF, whole SH-SY5Y cell extracts were analyzed by two-dimensional polyacrylamide gel electrophoresis using isoelectric focusing and nonequilibrium pH gradient electrophoresis gels in the first dimension. Steady-state levels of polypeptides extracted from whole-cell lysates of naive (untreated) cells, terminally differentiated cells, and NGF-deprived differentiated cells were compared. Over 1,000 spots from each were analyzed using computer-aided spot matching and densitometry. We noted 25 polypeptides that decreased during differentiation, including 15 that decreased by a factor of 10 or more. The levels of five polypeptides were induced from very low or undetectable levels in naive cells. Withdrawal of NGF from terminally differentiated cells produced alterations in steady-state protein patterns substantially distinct from those occurring during differentiation. While levels of most proteins do not appear affected early after NGF withdrawal, others rapidly return to levels comparable with those of the naive state and some changes occurring with differentiation are enhanced further upon NGF withdrawal. Three polypeptides were regulated uniquely by NGF withdrawal, including two that were induced, on average, 20- and 28-fold and another that was depressed more than 7-fold after NGF deprivation, before cell death. These data indicate that NGF elicits both constitutive and nonconstitutive changes in gene expression and suggest that the differentiation promoting and the neurotrophic properties of NGF correlate with the regulation of different gene products.

The binding domain of p75NGFR contains four "repeats" of a 6-cysteine pattern. To test whether these repeats have any structural or functional independence, each repeat has been separately deleted. In each case, deletion led to the loss of most nerve growth factor (NGF) binding activity. The epitopes of two monoclonal antibodies, MC192 and 271c, could be distinguished, however. Repeat IV was found to be unnecessary for binding MC192, whereas Repeat I was not required for binding 217c. This suggests that either terminal repeat can be removed without loss of native-like structure in the remaining repeats. Trp155 in the fourth repeat forms an essential part of the 217c epitope but is not required for either MC192 or NGF binding. Deletion of the linker region between the membrane-spanning domain and the cysteine-rich domain does not affect the binding of NGF or MC192, and has only a slight, if any, effect on 217c binding. Cyclic permutation of the four repeats failed to yield protein capable of binding NGF or MC192.

The changes in the levels of S100 beta (a protein that stimulates neurite extension and neuronal survival) and 42A and 42C (S100-like proteins whose mRNAs are induced in PC12 cells by nerve growth factor) during development and after rat sciatic nerve lesions were analyzed. S100 beta, 42A, and 42C mRNAs showed differential regulation during development. S100 beta mRNA was present both in sciatic nerve and brain, and increased more than 11-fold during the first 3 wk of nerve postnatal development. 42A and 42C mRNAs were essentially restricted to sciatic nerve, with little found in either embryonic or adult brain. The levels of 42C and 42A mRNAs in sciatic nerve increased 4- and 14-fold, respectively, by postnatal day 23 compared to postnatal day 2. 42A, 42C, and S100 beta mRNAs also showed a differential regulation during sciatic nerve degeneration and regeneration. Axotomized and control sciatic nerves were examined by Northern blots at various times after a crush or cut injury. 42A and 42C mRNA levels increased rapidly in the distal segment of axotomized nerve, remained two- to five-fold higher than controls at day 14 after injury but returned to control levels by 40 days. In contrast, S100 beta mRNA showed a three-fold decrease in the axotomized nerve between days 1 and 3 after injury, and slowly returned towards control levels over the next few weeks. The decrease in S100 beta mRNA was reflected by a corresponding decrease in S100 beta protein levels. The induction of 42A and 42C mRNAs and repression of S100 beta mRNA remained if nerve regeneration was prevented.(ABSTRACT TRUNCATED AT 250 WORDS)

Cross-linking of 125I-nerve growth factor (NGF) to PC12 cells with the photoreactive heterobifunctional agent N-hydroxysuccinimidyl-4-azidobenzoate results in the labeling of two major bands with Mr 158,000 and 100,000 and a minor band with Mr 225,000 as determined by polyacrylamide gel electrophoresis under denaturing and reducing conditions. Binding of 125I-NGF to and cross-linking into all these species is abolished in the presence of excess unlabeled NGF but not in the presence of unlabeled epidermal growth factor, insulin, or bovine pancreatic trypsin inhibitor. When PC12 cells with bound 125I-NGF are incubated in excess unlabeled NGF at 0 degree C prior to cross-linking, only the Mr 158,000 species remains. In addition, binding of 125I-NGF to the Mr 158,000 complex is trypsin-resistant, whereas binding to the Mr 100,000 complex is not. These experiments identify the Mr 158,000 species as the slow NGF-receptor complex (chase stable at 0 degree C) and the smaller Mr 100,000 species as the fast NGF-receptor complex (trypsin sensitive). Furthermore, 125I-NGF bound to the former but not to the latter species is displaced by very-low concentrations of NGF, showing that at least a significant fraction of the high-molecular-weight slow receptor is also a high-affinity receptor. This identification is supported by the finding that chick sensory neurons which possess both high- and low-affinity receptors exhibit two major labeled bands with Mr 145,000 and 105,000 as a result of cross-linking with 125I-NGF, whereas a cell population enriched in non-neuronal cells, which possess only low-affinity receptors, exhibits only the Mr 105,000 component. A shift in molecular weight of both species after pretreatment with neuraminidase indicates that both complexes contain sialoglycoproteins and rules out the possibility that differences in sialic acid content are responsible for the difference in molecular weight of the two complexes. The relative amount of the labeling of these two complexes is not affected by the presence of protease inhibitors nor by a variation of 5000-fold in cross-linker concentration. These results place some limits on possible models for the NGF receptors and their interconversion.

Nerve growth factor (NGF) regulates the microtubule-dependent extension and maintenance of axons by some peripheral neurons. We show here that one effect of NGF is to promote microtubule assembly during neurite outgrowth in PC12 cells. Though NGF causes an increase in total tubulin levels, the formation of neurites and the assembly of microtubules follow a time course completely distinct from that of the tubulin induction. The increases in microtubule mass and neurite extension closely parallel 10- and 20-fold inductions of tau and MAP1, proteins shown previously to promote microtubule assembly in vitro. When NGF is removed from PC12 cells, neurites disappear, microtubule mass decreases, and both microtubule-associated proteins return to undifferentiated levels. These data suggest that the induction of tau and MAP1 in response to NGF promotes microtubule assembly and that these factors are therefore key regulators of neurite outgrowth.

The arginine esteropeptidase gamma subunit of the 7 S nerve growth factor complex contains six different molecular species which arise by post-translational modifications. An endopeptidase cleavage of the larger of the two peptide chains in two-chain gamma species produces the three-chain species and introduces a new COOH-terminal lysine residue. The charge heterogeneity of the various species arises from the removal, by a carboxypeptidase B-like enzyme in the submaxillary gland, of this lysine residue and/or the original COOH-terminal arginine residue of the shorter Mr = 10,000 chain. These latter interconversions of the gamma species have been reproduced in vitro using carboxypeptidase B. The significant decrease in the proportion of two-chain gamma species in saliva compared to submaxillary gamma subunit suggests that the endopeptidase cleavage can occur during secretion of saliva; however, the subcellular localization of the enzyme has not been ascertained.

The biosynthesis of beta nerve growth factor (betaNGF) was studied in isolated mouse submaxillary glands incubated with L-[35S]cystine. Sodium dodecyl sulfate gels of anti-betaNGF immunoprecipitates from labeled gland homogenates showed a single major peak of radioactivity, which comigrated with purified betaNGF. This species was nearly completely precipitated by the addition of equivalent amounts of anti-betaNGF, but was absent from immunoprecipitates obtained by the addition of ferritin plus anti-ferritin. The cystine-containing tryptic peptides of the labeled species appeared identical with those of purified betaNGF. In submaxillary glands from adult male mice, labeling of betaNGF represented approximately 0.2% of the trichloroacetic acid-precipitable radioactivity. Castration reduced this value to one-third, while testosterone treatment of castrated animals restored the relative betaNGF synthesis to normal or more. No betaNGF synthesis could be detected in glands from female animals. Several tissues were examined for their ability to synthesize betaNGF in culture. Only submaxillary gland incorporated detectable amounts of radioactivity into betaNGF. Labeling of betaNGF could also be obtained by direct injection of isotope into the submaxillary gland in vivo. The results are discussed in terms of the integration of betaNGF synthesis into neuronal development and maintenance.

The nerve growth factor (NGF) and the epidermal growth factor (EGF) of the male mouse submaxillary gland are found in association with arginine esteropeptidases. These enzymes, the gamma subunits of 7 S NGF and the EGF-binding protein, respectively, have similar molecular weights, amino acid compositions, and substrate specificities (Greene, L. A., Shooter, E. M., and Varon, S. (1969) Biochemistry 8, 3735-3741; Taylor, J.M., Mitchell, W. M., and Cohen, S. (1974) J. Biol. Chem. 249, 21-8-2194). Although on the basis of molecular weight, the EGF-binding protein has a peptide chain composition similar to that of the gamma3 subunit, it does not have the same isoelectric point nor electrophoretic properties as the subunit. In urea, its isoelectric point differs from those of the gamma1, gamma2, and gamma3 subunits. The EGF-binding protein has antigenic determinants not shared by the gamma subunits and vice versa. It also fails to replace the gamma subunits in the formation of 7 S NGF points to the specificity of this complex and supports the hypothesis that the nerve and epidermal growth factors are generated from larger precursors by the proteolytic action of their respective arginine esteropeptidases.

The electrophoretic analyses of fractions prepared from saline extracts of barley grain by precipitation in aqueous ethanol solutions at low temperatures give a clearer picture of the protein composition of these extracts than can be obtained from analyses of fractions precipitated with ammonium sulphate. At least six major and eight minor components may be detected, as well as a group of inhomogeneous components which are concentrated in one fraction. All the major components appear to be proteins. The analyses suggest that these components are individual components and are not interaction products. The carbohydrate material derived from the barley gums is again concentrated in the last fraction. Comparison with previous work suggests that these numerous components which have been identified by electrophoresis can probably be assigned to one or other of the already known five major sedimenting groups of barley compounds.

Approximately 10% of the total nitrogen content of barley grain may be extracted from ground barley (Spratt-Archer) with saline solutions under conditions chosen to minimize proteinase activity. Electrophoretic analysis of arbitrary fractions produced from these extracts by ammonium sulphate show that they contain many non-dialysable compounds. At least two more major components are revealed by the electrophoretic analysis than may be accounted for by Quensel's four globulins and the barley albumin of Danielsson and Sandegren. Of the numerous minor components, one contains nucleic acid whilst a second probably belongs to the general class of plant proteinases. The non-protein material is derived mainly from the barley gums and is chiefly precipitated in the fraction precipitated between 0.7-1.0 saturation with ammonium sulphate.

The literature concerned with the composition of dialysed aqueous and saline extracts of barley grain has been reviewed, in particular in relation to the electrophoretic and ultracentrifugal analyses of the many fractions prepared from the extracts by precipitation with ammonium sulphate. The evidence suggests that at least the α-globulin originally described by Quensel1 is a mixture. Similarly, certain albumin preparations which sediment as single boundaries in the ultracentrifuge are shown by electrophoresis to be complex. On the other hand, the β- and γ-globulins appear to be more homogeneous. No complete separation of albumins and globulins is apparently achieved by dialysing the extracts against water. It would appear that a group of non-dialysable polypeptides which are not precipitated by ammonium sulphate constitute an important part of the nitrogenous fractions in the barley extracts.

The electrophoretic analysis of arbitrary ammonium sulphate fractions from a standard unhopped wort suggests that wort contains fewer individual complex nitrogen compounds than does barley. Although the form of the moving boundaries and the distribution of nitrogen amongst the fractions both emphasize the degradation of nitrogen compounds in malting and mashing, there is also evidence for the presence of unchanged barley nitrogen compounds in wort. Dextrins and pentosans are precipitated with the nitrogen compounds, the former at all levels of saturation but the latter chiefly above 0.4 saturation.

In the presence of nerve growth factor (NGF) PC12 pheochromocytoma cells develop properties of sympathetic neurons and extend long microtubule-containing neurites. PC12 cell microtubule protein was isolated using an assembly and disassembly procedure, either directly from an unlabeled cell extract or by copolymerization of a [35S]-methionine-labeled cell extract with rat brain microtubule proteins. Microtubule proteins of PC12 cells treated and untreated with NGF did not reveal any differences as analyzed by SDS-PAGE. The major constituents were the tubulin dimer and microtubule-associated proteins, mainly one high molecular weight component (HMW2) (Mr = 290,000), which resembles by several criteria the high molecular weight protein MAP2 of rat brain microtubule protein. PC12 cell microtubule protein contained intrinsic cAMP-dependent and cAMP-independent protein kinase activity. In vitro phosphorylation experiments indicated that, unlike in rat brain microtubule preparations, both subunits of the tubulin dimer were substrate for the endogenous kinase activity. The addition of cAMP to the incubation medium exerted a slight increase in phosphorylation of the HMW2 protein and an 80,000 polypeptide, and most effectively the phosphorylation of a Mr = 62,000 peptide was increased. This component could be identified as tyrosine hydroxylase by indirect immunoprecipitation procedures. Thus, the in vitro phosphorylation pattern of PC12 cell microtubule protein differed substantially from the one obtained from rat brain microtubule preparations.

The epidermal growth factor (EGF) can be isolated from the submaxillary gland of the adult male mouse as part of a high molecular weight complex (HMW-EGF). This complex can be reversibly dissociated into its subunits, EGF AND EGF-binding protein, an arginine esteropeptidase (Taylor, J. M., Cohen, S., and Mitchell, W. M. (1970) Proc. Natl. Acad. Sci. U. S. A. 67, 164-171). The COOH-terminal arginine residue of EGF was quantitatively removed by digestion with carboxypeptidase B...

Cross-linking of 125I-nerve growth factor (NGF) to PC12 cells with the photoreactive heterobifunctional agent N-hydroxysuccinimidyl-4-azidobenzoate results in the labeling of two major bands with Mr 158,000 and 100,000 and a minor band with Mr 225,000 as determined by polyacrylamide gel electrophoresis under denaturing and reducing conditions. Binding of 125I-NGF to and cross-linking into all these species is abolished in the presence of excess unlabeled NGF but not in the presence of unlabeled epidermal growth factor, insulin, or bovine pancreatic trypsin inhibitor. When PC12 cells with bound 125I-NGF are incubated in excess unlabeled NGF at 0 degree C prior to cross-linking, only the Mr 158,000 species remains. In addition, binding of 125I-NGF to the Mr 158,000 complex is trypsin-resistant, whereas binding to the Mr 100,000 complex is not. These experiments identify the Mr 158,000 species as the slow NGF-receptor complex (chase stable at 0 degree C) and the smaller Mr 100,000 species as the fast NGF-receptor complex (trypsin sensitive). Furthermore, 125I-NGF bound to the former but not to the latter species is displaced by very-low concentrations of NGF, showing that at least a significant fraction of the high-molecular-weight slow receptor is also a high-affinity receptor. This identification is supported by the finding that chick sensory neurons which possess both high- and low-affinity receptors exhibit two major labeled bands with Mr 145,000 and 105,000 as a result of cross-linking with 125I-NGF, whereas a cell population enriched in non-neuronal cells, which possess only low-affinity receptors, exhibits only the Mr 105,000 component. A shift in molecular weight of both species after pretreatment with neuraminidase indicates that both complexes contain sialoglycoproteins and rules out the possibility that differences in sialic acid content are responsible for the difference in molecular weight of the two complexes. The relative amount of the labeling of these two complexes is not affected by the presence of protease inhibitors nor by a variation of 5000-fold in cross-linker concentration. These results place some limits on possible models for the NGF receptors and their interconversion.

The internalization and subsequent fate of the two populations of nerve growth factor (NGF) receptors on pheochromocytoma PC12 cells were explored either by identifying the relative amounts and sizes of the receptors, after incubation of cells with [125I]NGF, by cross-linking with a photoreactive heterobifunctional reagent or by following the topological distribution of the cross-linked receptors with time. The ratio of the slow, high-affinity to the fast, low-affinity NGF receptor decreased over a 5-h incubation with [125I]NGF in a process which did not involve proteolytic conversion of the slow to the fast receptor. During this period the cross-linked slow receptor moved from a trypsin-labile to a trypsin-stable site suggestive of internalization. In contrast, the cross-linked fast NGF receptor remained trypsin sensitive for at least 2 h of incubation, indicative of a constant cell surface localization. The internalized [125I]NGF in the cross-linked slow NGF receptor was not degraded, indicating that cross-linking, by preventing the acid pH-induced dissociation of the NGF-receptor complex in the endosomes, blocks normal sorting of [125I]NGF to the lysosomes. The cross-linked receptor was not recycled to the cell surface. If this reflects the properties of the unmodified receptor then another process, possibly receptor conversion, is required to replenish slow NGF receptors in the cell surface.

An epidermal growth factor (EGF) receptor-interactive monoclonal antibody (151-IgG) that inhibits EGF binding to PC12 rat pheochromocytoma cells and to various other cell types has been produced. The hybridoma clone was obtained by fusing Sp2/O-Ag14 myeloma cells with splenocytes from Balb/C mice which had been immunized with n-octyl glucoside-solubilized protein from isolated PC12 cell plasma membranes. The antibody is an IgG which binds to protein A. 151-IgG did not bind EGF. At 0.5 degrees C 151-IgG was directly competitive for EGF binding to PC12 cells. It also inhibited EGF binding to bovine corneal endothelial cells, rabbit corneal fibroblasts, human foreskin fibroblasts, and normal rat kidney cells, and it slightly enchanced EGF binding to SW 3T3 cells. PC12 cells have the same number of binding sites for 151-IgG as for EGF (approximately 27,000 sites/cell). 151-IgG inhibited the photoactivatable cross-linking of EGF to a protein of Mr 170,000 in PC12 cells. 151-IgG inhibited the EGF-stimulated incorporation of [3H]thymidine into quiescent bovine corneal endothelial cells, rabbit corneal endothelial cells, epithelial normal rat kidney cells, and SW 3T3 cells while it enhanced the EGF-stimulated [3H]thymidine incorporation into quiescent human foreskin fibroblasts. 151-IgG by itself possessed intrinsic EGF-like activity for human fibroblasts but not for the other cells tested. This suggests that there is a difference in EGF receptors and/or processing in these normal cell types.

The nerve growth factor (NGF) and the epidermal growth factor (EGF) of the male mouse submaxillary gland are found in association with arginine esteropeptidases. These enzymes, the gamma subunits of 7 S NGF and the EGF-binding protein, respectively, have similar molecular weights, amino acid compositions, and substrate specificities (Greene, L. A., Shooter, E. M., and Varon, S. (1969) Biochemistry 8, 3735-3741; Taylor, J.M., Mitchell, W. M., and Cohen, S. (1974) J. Biol. Chem. 249, 21-8-2194). Although on the basis of molecular weight, the EGF-binding protein has a peptide chain composition similar to that of the gamma3 subunit, it does not have the same isoelectric point nor electrophoretic properties as the subunit. In urea, its isoelectric point differs from those of the gamma1, gamma2, and gamma3 subunits. The EGF-binding protein has antigenic determinants not shared by the gamma subunits and vice versa. It also fails to replace the gamma subunits in the formation of 7 S NGF points to the specificity of this complex and supports the hypothesis that the nerve and epidermal growth factors are generated from larger precursors by the proteolytic action of their respective arginine esteropeptidases.