A p85 subunit-independent p110alpha PI 3-kinase colocalizes with p70 S6 kinase on actin stress fibers and regulates thrombin-stimulated stress fiber formation in swiss 3T3 cells.

The signaling pathways linking receptor activation to actin stress fiber rearrangements during growth factor-induced cell shape change are still to be determined. Recently our laboratory demonstrated the involvement of p70 S6 kinase (p70(s6k)) activation in thrombin-induced stress fiber formation in Swiss 3T3 cells. The present work shows that thrombin-induced p70(s6k) activation is inhibited by the PI 3-kinase inhibitors wortmannin and LY-294002. These inhibitors also significantly reduced thrombin-induced stress fiber formation, demonstrating a role for PI 3-kinase activity in this process, most likely upstream of p70(s6k). Furthermore, the p110alpha form of PI 3-kinase was localized to actin stress fibers, as was previously shown for p70(s6k), as well as to a golgi-like distribution. In contrast, PI 3-kinase p110gamma colocalized with microtubules. The PI 3-kinase p85 subunit, known to be capable of association with p110alpha, was present in a predominantly golgi-like distribution with no presence on actin filaments, suggesting the existence of distinctly localized PI 3-kinase pools. Immunodepletion of p85 from cell lysates resulted in only partial depletion of p110alpha and p110alpha-associated PI 3-kinase activity, confirming the presence of a p85-free p110alpha pool located on the actin stress fibers. Our data, therefore, point to the importance of subcellular localization of PI 3-kinase in signal transduction and to a novel action of p85 subunit-independent PI 3-kinase p110alpha in the stimulation by thrombin of p70(s6k) activation and actin stress fiber formation.

Signal transduction from membrane to nucleus: the special case for neurons.

Neurons have a unique problem with signal transduction from the membrane in the region of their terminals back to the cell body and nucleus. This distance may be several meters in some nerves in some species, so there is a requirement for some mechanism to stabilize the signal. This review examines two complementary mechanisms for this signal transduction, either by the retrograde axonal transport of the neurotrophic factor together with its receptor, or the transport of a stable activated second messenger molecule. Extrapolation of studies on the fibroblast signal transduction pathway, where it has been shown that G1 can translocate from the membrane to the nucleus, has led to the demonstration of the retrograde axonal transport of several putative signaling molecules. The alpha subunits of both G1 and Gz are retrogradely transported and Gz alpha or possibly the intact heterotrimeric Gz subsequently accumulates in dorsal root ganglia nuclei. Thus Gz1 Gi1 and potentially other G-proteins and distinct signaling molecules may provide additional signal transduction pathways to that of the neurotrophins from terminal to nucleus.

Retrograde axonal transport of the alpha subunit of the GTP-binding protein Gz to the nucleus of sensory neurons.

Nerve cells are exquisitely sensitive to target tissue derived factors and the discovery that nerve growth factor could be retrogradely transported in axons suggested that the physical translocation of proteins along the axon could be a mechanism to convey this signal. This message is not due to the neurotrophic factor itself but rather due to second messengers generated by interaction with receptors. We have previously demonstrated the retrograde axonal transport of the alpha subunits of two putative second messenger molecules Gi and Gz. We have investigated more thoroughly the transport of the alpha subunit of Gz (Gz alpha) and in order to be more certain that the immunoreactivity seen is due to Gz alpha, we have made antibodies to peptides from both the N- and C-terminal regions of Gz alpha, which recognise the same 41 kDa band on Western blots of brain and sciatic nerve extracts. This band is eliminated when the antibodies are previously incubated with the specific peptide to which they were made. Using these antibodies for immunohistochemical localisation for Gz alpha, we now report that the GTP-binding protein Gz, is not only retrogradely transported in axons but that it translocates to the neuronal nucleus. Furthermore, the levels seen in the nuclear compartment decline after axotomy or ligation of the mice under ether anaesthetic, suggesting it is the retrogradely transported Gz alpha that is accumulating in the nucleus after activation at the nerve terminal.

Developmental signalling.

1. In investigating the communication paths between target tissue and neurons we have been led to propose two classes of neurotrophic factors. One comprises the factors which transport themselves, the other factors relying on the transport of a second messenger. The former may have labile second messenger systems necessitating the translocation of agonist and receptor from the nerve terminal to the cell body and the latter must possess a stable second messenger system that itself is sufficiently robust to survive the transport to the cell body. 2. One such class of stable messengers may be the GTP-binding protein family and it has been shown that the alpha subunits of both Gi alpha and Gz alpha can be retrogradely transported in the mouse sciatic nerve. 3. Examination of the cell bodies in the dorsal root ganglia revealed that Gz alpha accumulated in the nucleus of cells with intact axons but that 24 h after axonal ligation this immunoreactivity decreased. 4. It is suggested that Gz is activated at the nerve terminal and it, or at least its alpha subunit, undergoes retrograde transport to the cell body where it accumulates in the nucleus.

Retrograde axonal transport of signal transduction proteins in rat sciatic nerve.

Neurons require a mechanism to transmit stable signals over the large distance from the nerve growth cone or terminal to the cell body, in order that information from the target tissue can be relayed to the cell body where it is required. Nerve growth factor (NGF), a target-derived neurotrophic factor, is thought to signal over this distance by receptor mediated internalization of NGF, followed by retrograde axonal transport of the NGF-receptor complex. In this paper we show, by immunohistochemistry of rat sciatic nerve, accumulation of phosphotyrosine immunoreactivity only on the distal side of a nerve crush, suggesting axonal transport of tyrosine kinases and/or tyrosine phosphorylated proteins primarily in a retrograde direction. Furthermore, we also show retrograde axonal transport of phosphoinositide 3-kinase, ERK, MEK and MEK kinase, of which all but MEK kinase are known to be activated downstream of tyrosine receptor kinase activation. The retrograde transport of these proteins suggests that they may be involved in transmission of signals along the axon, relaying neurotrophic factor receptor activation at the nerve terminal to the nerve cell body.

Glutamate decarboxylase solubilized from the rat cerebral cortex by two different concentrations of Triton X-100: effects of glutamate analogues and analysis by SDS-PAGE/western blotting using GAD6 and K2 antibodies.

Analysis of two preparations (containing 0.1% and 0.5% Triton X-100) of glutamate decarboxylase (GAD) by Western blotting using GAD6 and K2 antibodies specifically recognizing two GAD isoenzymes, GAD65 and GAD67, respectively, indicated that the higher concentration of Triton X-100 at best only moderately favoured solubilization of GAD67. Several glutamate analogues were found to be either equally potent or equally inactive as inhibitors of glutamate decarboxylase activities in the two preparations. Among typical ligands for glutamate receptors and transporters, only quinolinic and L-cysteine sulphinic acids were weak inhibitors of GAD. Kainate, alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionate (AMPA), 3-((RS)-carboxypiperazin-4-yl)-propyl-1-phosphonic acid (CPP), L-threo-3-hydroxy-aspartate, L-trans-pyrrolidine-2,4-dicarboxylate, dihydrokainate, kynurenic acid and N-methyl-D-aspartate were inactive. Even though the activity of glutamate decarboxylase in homogenates of rat cerebral cortex is higher at 0.5% than at 0.1% Triton X-100, structural requirements of the enzyme active site appear to be independent of Triton X-100 concentration. Furthermore, since the less soluble component of the enzyme activity contains about the same ratio of GAD65 to GAD67 as the more soluble one, it does not seem that the fractionation with Triton X-100 can be easily used to separate the two isoenzymes from each other.