Elevation of the terminal complement activation products C5a and C5b-9 in ALS patient blood.

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease, characterized by the progressive loss of motor neurons within the central nervous system. Neural degeneration and inflammatory processes, including activation of the complement system are hallmarks of this pathology. Our past work in ALS animal models (hSOD1 G93A rodents) has revealed that blockade of the receptor for complement activation fragment C5a (C5aR), improves ALS-like symptoms and extends survival. We now show that the levels of C5a and C5b-9, but not C3a nor C4a, are significantly elevated in plasma from ALS patients compared to healthy controls. C5a was also elevated within leukocytes from ALS patients suggesting heightened C5a receptor interaction. Overall, these findings indicate that there is enhanced peripheral immune complement terminal pathway activation in ALS, which may have relevance in the disease process.

Maternal smoking is associated with impaired neonatal toll-like-receptor-mediated immune responses.

Infants of smokers have much higher rates of respiratory infection, asthma and airway disease. The current study assessed the effects of maternal smoking in pregnancy on neonatal toll-like-receptor (TLR)-mediated immune responses as a possible contributing factor to the elevated rates of respiratory illness. In a prospective birth cohort, the cord blood immune responses of neonates of smoking and nonsmoking mothers were compared. Maternal and cord serum cotinine were measured to confirm the level of cigarette smoke exposure. Neonatal cytokine responses were assessed to optimal doses of TLR2, TLR3, TLR4 and TLR9 ligands. Cotinine levels confirmed maternal reporting of cigarette smoking in pregnancy, with significantly higher cotinine levels in maternal and cord blood compared with the nonsmoking group. Infants of smoking mothers showed significantly attenuated innate TLR-mediated responses compared with infants of nonsmokers. The current findings indicate that in addition to effects on developing airways, maternal smoking also has significant immunological effects in pregnancy, which could contribute to the well recognised, subsequent increased risk of respiratory infections and asthma. These effects appear to be mediated through effects on toll-like receptor-mediated innate response pathways, which also promote regulatory pathways in the inhibition of allergic immune responses in the postnatal period, suggesting that other environmental interactions are highly relevant to the "hygiene hypothesis".

Maternal reactivity to fetal alloantigens is related to newborn immune responses and subsequent allergic disease.

BACKGROUND: Maternal allergy confers stronger allergy risk (than paternal allergy) suggesting that maternal patterns of immune response can directly influence immune development in offspring. Women prone to allergic immune responses to allergens may also have altered immune responses to other antigens including fetal antigens. OBJECTIVE: This exploratory study examines relationships between maternal immune responses to fetal antigens and the subsequent risk of allergy. METHODS: Mononuclear cells (MNC) were collected from 36 mother-infant pairs to compare maternal (and fetal) cellular immune responses to alloantigens (fetal, maternal or unrelated donor [URD]), and allergens in allergic (18 pairs) and non-allergic (18 pairs) mothers. Thirty children had documented allergic outcomes at 6 years of age. RESULTS: In this population, allergic outcomes in the offspring were associated more strongly with materno-fetal immune interactions than with a maternal family history of allergy. Specifically, allergic disease at 6 years of age was associated with significantly higher maternal responses to fetal alloantigens (lymphoproliferation, P=0.008; IL-13, P=0.02 and IFN-gamma, P=0.015), whereas associations with maternal allergy did not reach significance (P=0.07). Fetal IFN-gamma alloantigen responses were significantly correlated with the degree of human lymphocyte antigen (HLA) mismatch (maternal HLA class II antibodies) (tau=0.3, P=0.03). The capacity of the fetus to produce IL-13 (tau=0.4, P=0.001) and IL-10 (tau=0.3, P=0.029) was directly related to the level of these cytokines produced by the mother in response to fetal antigens. Allergic mothers showed a non-significant trend for stronger lymphoproliferation to fetal alloantigens. The number of previous pregnancies (gravidity) was associated with stronger maternal responses to fetal alloantigens, as shown by lymphoproliferation (Kendall tau=0.3, P=0.04) and IFN-gamma (tau=0.3, P=0.04) synthesis, but did not affect fetal responses to the various stimuli. CONCLUSIONS: Maternal responses to fetal antigens were related to fetal immune responses and subsequent allergy. This novel observation suggests that events at the materno-fetal interface have an important influence on early immune development and should be investigated further.

P2X7-like receptor subunits enhance excitatory synaptic transmission at central synapses by presynaptic mechanisms.

Recent studies demonstrate that P2X7 receptor subunits (P2X7RS) are present at central and peripheral synapses and suggest that P2X7RS can regulate transmitter release. In brainstem slices from 15 to 26 day old pentobarbitone-anesthetized mice, we examined the effect of P2X7RS activation on excitatory postsynaptic currents (EPSCs) recorded from hypoglossal motoneurons using whole-cell patch clamp techniques. After blockade of most P2X receptors with suramin (which is inactive at P2X7RS) and of adenosine receptors with 8-phenyltheophylline (8PT), bath application of the P2X receptor agonist 3'-0-(4-benzoyl)ATP (BzATP) elicited a 40.5+/-16.0% (mean+/-S.E.M., n = 8, P = 0.039) increase in evoked EPSC amplitude and significantly reduced paired pulse facilitation of evoked EPSCs. This response to BzATP (with suramin and 8PT present) was completely blocked by prior application of Brilliant Blue G (200 nM or 2 microM), a P2X7RS antagonist. In contrast, BzATP application with suramin and 8PT present did not alter miniature EPSC frequency or amplitude when action potentials were blocked with tetrodotoxin. These electrophysiological results suggest that P2X7RS activation increases central excitatory transmitter release via presynaptic mechanisms, confirming previous indirect measures of enhanced transmitter release. We suggest that possible presynaptic mechanisms underlying enhancement of evoked transmitter release by P2X7RS activation are modulation of action potential width or an increase in presynaptic terminal excitability, due to subthreshold membrane depolarization which increases the number of terminals releasing transmitter in response to stimulation.

Maternal smoking in pregnancy alters neonatal cytokine responses.

BACKGROUND: Maternal cigarette smoking in pregnancy is an important, common and avoidable exposure that has been linked with elevated cord blood (CB) immunoglobulin E levels and subsequent asthma and allergic disease in childhood. Despite this, there is still very little information about the immunological effects of maternal smoking on the fetus. METHODS: This aim of this study was to compare cord blood mononuclear cell (CBMC) cytokine responses to allergens [ovalbumin (OVA) or house dust mite (HDM)] and mitogens [concanavalin A (ConA) or phytohemaglutinen (PHA)] in neonates whose mothers smoked throughout pregnancy (n = 17) with responses of neonates whose mothers never smoked (n = 40). Cell cultures were stimulated for 24 h and supernatants collected for cytokine detection by enzyme-linked immunosorbent assay [interleukin (IL)-13, IL-6, interferon (IFN)gamma and IL-10]. Cell pellets were also collected for cytokine mRNA detection (IL-5, IL-9, IFNgamma). RESULTS: Maternal smoking in pregnancy was associated with significantly higher neonatal T helper type 2 (IL-13 protein) responses to both HDM (P = 0.01) and OVA (P = 0.035). These effects remained statistically significant after allowing for confounding factors, including the effects of maternal atopy. Similar trends were also seen for IL-9mRNA, IL-5mRNA and IL-6 responses, although these were not statistically significant. Although IFNgamma mRNA responses to PHA (P = 0.015) and ConA (P = 0.025) were lower if mothers smoked in pregnancy, there were no differences in neonatal (Th1) IFNgamma protein responses to allergens or mitogens. CONCLUSIONS: These findings indicate that maternal cigarette smoking can modify aspects of fetal immune function and highlight the need for further studies in this area.

Alterations in ciliary neurotrophic factor signaling in rapsyn deficient mice.

Rapsyn is a key molecule involved in the formation of postsynaptic specializations at the neuromuscular junction, in its absence there are both pre- and post-synaptic deficits including failure to cluster acetylcholine receptors. Recently we have documented increases in both nerve-muscle branching and numbers of motoneurons, suggesting alterations in skeletal muscle derived trophic support for motoneurons. The aim of the present study was to evaluate the contribution of target derived trophic factors to increases in motoneuron branching and number, in rapsyn deficient mice that had their postsynaptic specializations disrupted. We have used reverse transcription-polymerase chain reaction and Western blot to document the expression of known trophic factors and their receptors in muscle, during the period of synapse formation in rapsyn deficient mouse embryos. We found that the mRNA levels for ciliary neurotrophic factor (CNTF) was decreased in the rapsyn deficient muscles compared with litter mate controls although those for NGF, BDNF, NT-3 and TGF-beta2 did not differ. We found that both the mRNA and the protein expression for suppressor of cytokine signaling 3 (SOCS3) decreased although janus kinase 2 (JAK2) did not change in the rapsyn deficient muscles compared with litter mate controls. These results suggest that failure to form postsynaptic specializations in rapsyn deficient mice has altered the CNTF cytokine signaling pathway within skeletal muscle, the target for motoneurons. This alteration may in part, account for the increased muscle nerve branching and motoneuron survival seen in rapsyn deficient mice.

Transport of endosomal early antigen 1 in the rat sciatic nerve and location in cultured neurons.

Early endosomal antigen 1 (EEA1) is known to be a marker of early endosomes and in cultured hippocampal neurons it preferentially localizes to the dendritic but not the axonal compartment. We show in cultured dorsal root ganglia and superior cervical ganglia neurons that EEA1 localizes to the cell bodies and the neurites of both sensory and sympathetic neurons. We then show in vivo using a ligated rat sciatic nerve that EEA1 significantly accumulates on the proximal side and not on the distal side of the ligation. This suggests that EEA1 is transported in the anterograde direction in axons either as part of the homeostatic process or to the nerve ligation site in response to nerve injury.

The mode of toxic action of the pesticide gliftor: the metabolism of 1,3-difluoroacetone to (-)-erythro-fluorocitrate.

The biochemical toxicology of 1,3-difluoroacetone, a known metabolite of the major ingredient of the pesticide Gliftor (1,3-difluoro-2-propanol), was investigated in vivo and in vitro. Rat kidney homogenates supplemented with coenzyme A, ATP, oxaloacetate, and Mg2+ converted 1,3-difluoroacetone to (-)-erythro-fluorocitrate in vitro. Administration of 1,3-difluoroacetone (100 mg kg(-1) body weight) to rats in vivo resulted in (-)-erythro-fluorocitrate synthesis in the kidney, which was preceded by an elevation in fluoride levels and followed by citrate accumulation. Animals dosed with 1,3-difluoroacetone did not display the 2-3 hour lag phase in either (-)-erythro-fluorocitrate synthesis or in citrate and fluoride accumulation characteristic of animals dosed with 1,3-difluoro-2-propanol. We demonstrate that the conversion of 1,3-difluoro-2-propanol to 1,3-difluoroacetone by an NAD+-dependent oxidation is the rate-limiting step in the synthesis of the toxic product, (-)-erythro-fluorocitrate from 1,3-difluoro-2-propanol. Prior administration of 4-methylpyrazole (90 mg kg(-1) body weight) was shown to prevent the conversion of 1,3-difluoro-2-propanol (100 mg kg(-1) body weight) to (-)-erythro-fluorocitrate in vivo and to eliminate the fluoride and citrate elevations seen in 1,3-difluoro-2-propanol-intoxicated animals. However, administration of 4-methylpyrazole (90 mg kg(-1) body weight) to rats 2 hours prior to 1,3-difluoroacetone (100 mg kg(-1) body weight) was ineffective in preventing (-)-erythro-fluorocitrate synthesis and did not diminish fluoride or citrate accumulation in vivo. We conclude that the prophylactic and antidotal properties of 4-methylpyrazole seen in animals treated with 1,3-difluoro-2-propanol derive from its capacity to inhibit the NAD+-dependent oxidation responsible for converting 1,3-difluoro-2-propanol to 1,3-difluoroacetone in the committed step of the toxic pathway.

Promotion of motoneuron survival and branching in rapsyn-deficient mice.

Inhibition of programmed cell death of motoneurons during embryonic development requires the presence of their target muscle and coincides with the initial stages of synaptogenesis. To evaluate the role of synapse formation on motoneuron survival during embryonic development, we counted the number of motoneurons in rapsyn-deficient mice. Rapsyn is a 43 kDa protein needed for the formation of postsynaptic specialisations at vertebrate neuromuscular synapses. Here we show that the rapsyn-deficient mice have a significant increase in the number of motoneurons in the brachial lateral motor column during the period of naturally occurring programmed cell death compared to their wild-type littermates. In addition, we observed an increase in intramuscular axonal branching in the rapsyn-deficient diaphragms compared to their wild-type littermates at embryonic day 18.5. These results suggest that deficits in the formation of the postsynaptic specialisation at the neuromuscular synapse, brought about by the absence of rapsyn, are sufficient to induce increases in both axonal branching and the survival of the innervating motoneuron. Moreover, these results support the idea that skeletal muscle activity through effective synaptic transmission and intramuscular axonal branching are major mechanisms that regulate motoneuron survival during development.

Expression and localisation of dynamin and syntaxin during neural development and neuromuscular synapse formation.

The expression and subcellular localisation of dynamin and syntaxin were examined during the periods of motor neuron development and neuromuscular synaptogenesis in the mouse embryo. Both dynamin and syntaxin could be detected by immunoblotting in the spinal cord at embryonic day 10 (E10; 2 days before axon outgrowth) and at all subsequent ages examined. Reverse transcription and polymerase chain reaction (RT-PCR) identified low levels of all three carboxy-terminal splicing forms of dynamin I in spinal cord from as early as E10. During the period of maturation of spinal neurons, from E10 to the first postnatal day (P0), the short carboxy-terminal splicing form of dynamin I (dynamin I*b) was up-regulated, as was dynamin III, relative to dynamin II mRNA. Syntaxin immunostaining became colocalized with the synaptic vesicle protein, SV2, at neuromuscular synapses within 12 hours of the commencement of synapse formation and throughout subsequent development. In contrast, dynamin, which is important for activity-dependent synaptic vesicle recycling and, thus, sustained neurotransmission, could not be detected at most newly formed synapses until several days after synapse formation. The delayed appearance of dynamin at the synapse, thus, heralds the neonatal development of robust synaptic transmission at the neuromuscular junction.

Overexpression of rapsyn inhibits agrin-induced acetylcholine receptor clustering in muscle cells.

Rapsyn is a protein on the cytoplasmic face of the postsynaptic membrane of skeletal muscle that is essential for clustering acetylcholine receptors (AChR). Here we show that transfection of rapsyn cDNA can restore AChR clustering function to muscle cells cultured from rapsyn deficient (KORAP) mice. KORAP myotubes displayed no AChR aggregates before or after treatment with neural agrin. After transfection with rapsyn expression plasmid, some KORAP myotubes expressed rapsyn at physiological levels. These formed large AChR-rapsyn clusters in response to agrin, just like wild-type myotubes. KORAP myotubes that overexpressed rapsyn formed only scattered AChR-rapsyn microaggregates, irrespective of agrin treatment. KORAP cells were then transfected with mutant forms of rapsyn. A deletion mutant lacking residues 16-254 formed rapsyn microaggregates, but failed to aggregate AChRs. Substitution mutation to the C-terminal serine phosphorylation site of rapsyn (M43(D405,D406)) did not impair the response to agrin, showing that differential phosphorylation of this site is unlikely to mediate agrin-induced clustering. The results indicate that rapsyn expression is essential for agrin-induced AChR clustering but that its overexpression inhibits this pathway. The approach of using rapsyn-deficient muscle cells opens the way for defining the role of rapsyn in agrin-induced AChR clustering.

Development of the neuromuscular junction: genetic analysis in mice.

Formation of the skeletal neuromuscular junction is a multi-step process that requires communication between the nerve and muscle. Studies in many laboratories have led to identification of factors that seem likely to mediate these interactions. 'Knock-out' mice have now been generated with mutations in several genes that encode candidate transsynaptic messengers and components of their effector mechanisms. Using these mice, it is possible to test hypotheses about the control of synaptogenesis. Here, we review our studies on neuromuscular development in mutant mice lacking agrin alpha CGRP, rapsyn, MuSK, dystrophin, dystrobrevin, utrophin, laminin alpha 5, laminin beta 2, collagen alpha 3 (IV), the acetylcholine receptor epsilon subunit, the collagenous tail of acetylcholinesterase, fibroblast growth factor-5, the neural cell adhesion molecule, and tenascin-C.

The biochemical toxicology of 1,3-difluoro-2-propanol, the major ingredient of the pesticide gliftor: the potential of 4-methylpyrazole as an antidote.

Administration to rats of 1,3-difluoro-2-propanol (100 mg kg-1 body weight), the major ingredient of the pesticide gliftor, resulted in accumulation of citrate in the kidney after a 3 hour lag phase. 1,3-Difluoro-2-propanol was found to be metabolized to 1,3-difluoroacetone and ultimately to the aconitate hydratase inhibitor (-) erythrofluorocitrate and free fluoride. The conversion of 1,3-difluoro-2-propanol to 1,3-difluoroacetone was found to be catalyzed by an NAD(+)-dependent alcohol dehydrogenase, while the defluorination was attributed to microsomal monooxygenase activity induced by phenobarbitone and inhibited by piperonyl butoxide. 4-Methylpyrazole was found to inhibit both of these processes in vitro and when administered (90 mg kg-1 body weight) to rats, 2 hours prior to 1,3-difluoro-2-propanol, eliminated signs of poisoning, prevented (-) erythrofluorocitrate synthesis, and markedly decreased citrate and fluoride accumulation in vivo. 4-Methylpyrazole also appeared to diminish (-) erythrofluorocitrate synthesis from fluoroacetate in vivo, and this was attributed to its capacity to inhibit malate dehydrogenase activity. The antidotal potential of 4-methylpyrazole and the potential for 1,3-difluoro-2-propanol to replace fluoroacetate (compound 1080) as a vertebrate pesticide is discussed.

Rapsyn and agrin slow the metabolic degradation of the acetylcholine receptor.

Rapsyn is a 43-kDa cytoplasmic protein that clusters nicotinic acetylcholine receptors (AChR) in the postsynaptic membrane. Here we examine the effect of rapsynmediated AChR clustering on the metabolic stability of the AChR. When transfected into QT-6 fibroblasts, cell surface AChRs (alpha, beta, epsilon, and delta subunit combination) pulse labeled with 125I-alpha-bungarotoxin were degraded with a half-life of 16.4 +/- 1.1 h (mean +/- SEM). Cotransfection of rapsyn with AChR caused extensive AChR clustering and increased AChR half-life to 20.5 +/- 1.0 h. Anti-AChR antibodies such as mab 35 cause an increased AChR degradation often associated with myasthenia gravis: 80.8 +/- 2.5% of AChRs labeled at zero time were degraded over a 12-h period. Contransfection of rapsyn reduced this AChR loss to 66.4 +/- 3.8%. Rapsyn also reduced normal AChR degradation, from 53.2 +/- 2.1 to 44.2 +/- 2.2%. Muscle cell lines from wild-type myotubes displayed few AChR clusters, but treatment with neural agrin increased the number of AChR clusters 30-fold. Clustering was accompanied by reductions in AChR degradation (both in the presence and absence of mab 35) similar in magnitude to those produced by overexpression of rapsyn in QT-6 cells. In rapsyn-deficient myotubes, treatment with neural agrin neither caused AChR clustering nor reduced AChR degradation. Thus neural agrin may slow AChR degradation by inducing the rapsyn-dependent clustering of AChRs.

Defective neuromuscular synaptogenesis in agrin-deficient mutant mice.

During neuromuscular synapse formation, motor axons induce clustering of acetylcholine receptors (AChRs) in the muscle fiber membrane. The protein agrin, originally isolated from the basal lamina of the synaptic cleft, is synthesized and secreted by motoneurons and triggers formation of AChR clusters on cultured myotubes. We show here postsynaptic AChR aggregates are markedly reduced in number, size, and density in muscles of agrin-deficient mutant mice. These results support the hypothesis that agrin is a critical organizer of postsynaptic differentiation does occur in the mutant, suggesting the existence of a second-nerve-derived synaptic organizing signal. In addition, we show that intramuscular nerve branching and presynaptic differentiation are abnormal in the mutant, phenotypes which may reflect either a distinct effect of agrin or impaired retrograde signaling from a defective postsynaptic apparatus.

Synapse-associated expression of an acetylcholine receptor-inducing protein, ARIA/heregulin, and its putative receptors, ErbB2 and ErbB3, in developing mammalian muscle.

Developing motor axons induce synaptic specializations in muscle fibers, including preferential transcription of acetylcholine receptor (AChR) subunit genes by subsynaptic nuclei. One candidate nerve-derived signaling molecule is AChR-inducing activity (ARIA)/heregulin, a ligand of the erbB family of receptor tyrosine kinases. Here, we asked whether ARIA and erbB kinases are expressed in patterns compatible with their proposed signaling roles. In developing muscle, ARIA was present not only at synaptic sites, but also in extrasynaptic regions of the muscle fiber. ARIA was synthesized, rather than merely taken up, by muscle cells, as indicated by the presence of ARIA mRNA in muscle and of ARIA protein in a clonal muscle cell line. ARIA-responsive myotubes expressed both erbB2 and erbB3, but little EGFR/erbB1 or erbB4. In adults, erbB2 and erbB3 were localized to the postsynaptic membrane. ErbB3 was restricted to the postsynaptic membrane perinatally, at a time when ARIA was still broadly distributed. Thus, our data are consistent with a model in which ARIA interacts with erbB kinases on the muscle cell surface to provide a local signal that induces synaptic expression of AChR genes. However, much of the ARIA is produced by muscle, not nerve, and the spatially restricted response may result from the localization of erbB kinases as well as of ARIA. Finally, we show that erbB3 is not concentrated at synaptic sites in mutant mice that lack rapsyn, a cytoskeletal protein required for AChR clustering, suggesting that pathways for synaptic AChR expression and clustering interact.

Failure of postsynaptic specialization to develop at neuromuscular junctions of rapsyn-deficient mice.

Of numerous synaptic components that have been identified, perhaps the best-studied are the nicotinic acetylcholine receptors (AChRs) of the vertebrate neuromuscular junction. AChRs are diffusely distributed on embryonic myotubes, but become highly concentrated (approximately 10,000 microns-2) in the postsynaptic membrane as development proceeds. At least two distinct processes contribute to this accumulation. One is local synthesis: subsynaptic muscle nuclei transcribe AChR subunit genes at higher rates than extra-synaptic nuclei, so AChR messenger RNA is concentrated near synaptic sites. Second, once AChRs have been inserted in the membrane, they form high-density clusters by tethering to a subsynaptic cytoskeletal complex. A key component of this complex is rapsyn, a peripheral membrane protein of relative molecular mass 43K (refs 4, 5), which is precisely colocalized with AChRs at synaptic sites from the earliest stages of neuromuscular synaptogenesis. In heterologous systems, expression of recombinant rapsyn leads to clustering of diffusely distributed AChRs, suggesting that rapsyn may control formation of clusters. To assess the role of rapsyn in vivo, we generated and characterized mutant mice with a targeted disruption of the Rapsyn gene. We report that rapsyn is essential for the formation of AChR clusters, but that synapse-specific transcription of AChR subunit genes can proceed in its absence.

The renal glomerulus of mice lacking s-laminin/laminin beta 2: nephrosis despite molecular compensation by laminin beta 1.

S-laminin/laminin beta 2, a homologue of the widely distributed laminin B1/beta 1 chain, is a major component of adult renal glomerular basement membrane (GBM). Immature GBM bears beta 1, which is replaced by beta 2 as development proceeds. In mutant mice that lack beta 2, the GBM remains rich in beta 1, suggesting that a feedback mechanism normally regulates GBM maturation. The beta 2-deficient GBM is structurally intact and contains normal complements of several collagenous and noncollagenous glycoproteins. However, mutant mice develop massive proteinuria due to failure of the glomerular filtration barrier. These results support the idea that laminin beta chains are functionally distinct although they assemble to form similar structures. Laminin beta 2-deficient mice may provide a model for human congenital or idiopathic nephrotic syndromes.

Aberrant differentiation of neuromuscular junctions in mice lacking s-laminin/laminin beta 2.

Synapse formation requires a complex interchange of information between the pre- and postsynaptic partners. At the skeletal neuromuscular junction, some of this information is contained in the basal lamina (BL), which runs through the synaptic cleft between the motor nerve terminal and the muscle fibre. During regeneration following injury, components of synaptic BL can trigger several features of postsynaptic differentiation in the absence of the nerve terminal, and of presynaptic differentiation in the absence of the muscle fibre. One nerve-derived component of synaptic BL, agrin, is known to affect postsynaptic differentiation, but no muscle-derived components have yet been shown to influence motor nerve terminals. A candidate for such a role is s-laminin (also called laminin beta 2), a homologue of the B1 (beta 1) chain of the widely distributed BL glycoprotein, laminin. s-Laminin is synthesized by muscle cells and concentrated in synaptic BL. In vitro, recombinant s-laminin fragments are selectively adhesive for motor neuron-like cells, inhibit neurite outgrowth promoted by other matrix molecules, and act as a 'stop signal' for growing neurites. By generating and characterizing mice with a targeted mutation of the s-laminin gene, we show here that s-laminin regulates formation of motor nerve terminals.

A framework for improved training of Sigma-Pi networks.

This paper proposes and demonstrates a framework for Sigma-Pi networks such that the combinatorial increase in product terms is avoided. This is achieved by only implementing a subset of the possible product terms (sub-net Sigma-Pi). Application of a dynamic weight pruning algorithm enables redundant weights to be removed and replaced during the learning process, hence permitting access to a larger weight space than employed at network initialization. More than one learning rate is applied to ensure that the inclusion of higher order descriptors does not result in over description of the training set (memorization). The application of such a framework is tested using a problem requiring significant generalization ability. Performance of the resulting sub-net Sigma-Pi network is compared to that returned by optimal multi-layer perceptrons and general Sigma-Pi solutions.