Viviane Hoppe

Flies lacking all synapsins are unexpectedly healthy but are impaired in complex behaviour

Vertebrate synapsins are abundant synaptic vesicle phosphoproteins that have been proposed to fine‐regulate neurotransmitter release by phosphorylation‐dependent control of synaptic vesicle motility. However, the consequences of a total lack of all synapsin isoforms due to a knock‐out of all three mouse synapsin genes have not yet been investigated. In Drosophila a single synapsin gene encodes several isoforms and is expressed in most synaptic terminals. Thus the targeted deletion of the synapsin gene of Drosophila eliminates the possibility of functional knock‐out complementation by other isoforms. Unexpectedly, synapsin null mutant flies show no obvious defects in brain morphology, and no striking qualitative changes in behaviour are observed. Ultrastructural analysis of an identified ‘model’ synapse of the larval nerve muscle preparation revealed no difference between wild‐type and mutant, and spontaneous or evoked excitatory junction potentials at this synapse were normal up to a stimulus frequency of 5 Hz. However, when several behavioural responses were analysed quantitatively, specific differences between mutant and wild‐type flies are noted. Adult locomotor activity, optomotor responses at high pattern velocities, wing beat frequency, and visual pattern preference are modified. Synapsin mutant flies show faster habituation of an olfactory jump response, enhanced ethanol tolerance, and significant defects in learning and memory as measured using three different paradigms. Larval behavioural defects are described in a separate paper. We conclude that Drosophila synapsins play a significant role in nervous system function, which is subtle at the cellular level but manifests itself in complex behaviour.

The induction of early response genes in rat smooth muscle cells by PDGF-AA is not sufficient to stimulate DNA-synthesis

The effect of the three platelet‐derived growth factor (PDGF) isoforms AA, AB and BB on the induction of the early growth response genes c‐fos, egr‐1 and c‐myc mRNA in vascular smooth muscle cells from rat was compared with their respective mitogenic potency. The three PDGF isoforms strongly stimulated the induction to a similar extent. In contrast, PDGF‐AB and ‐BB provoked a marked DNA synthesis whereas PDGF‐AA exerted only a poor mitogenic effect in smooth muscle cells. PDGF‐AA‐stimulated receptor autophosphorylation was not detectable in comparison with the strong effect elicited by PDGF‐AB or ‐BB and correlated with its low mitogenicity but not with the almost equal induction of the early response genes. It is discussed that no or only very low receptor phosphorylation is required to link receptor activation to the induction of c‐fos, egr‐1 or c‐myc. Furthermore the induction of the investigated gene does not seem to be sufficient for an optimal mitogenic response.

Effect of growth factors on the activation of human Tenon’s capsule fibroblasts

Purpose. To investigate stimulatory effects of PDGF-AA, PDGF-AB, PDGF-BB, bFGF, IL-1ß, TGF-ß1 and TGF-ß2 on the proliferation and myofibroblast transformation of cultured human Tenon’s capsule fibroblasts and to characterize expression of PDGF- and TGF-ß-receptors in these cells. Methods. To determine cell proliferation, cell number of 2nd passage cultured human Tenon’s capsule fibroblasts was measured before and after addition of growth factors using a computer-based cell counter system. Immunoblotting was used to detect and quantitate a-smooth-muscle actin (a-SMA) expression. Expression of PDGF- and TGF-ß-receptor mRNA was detected by RT-PCR, expression of the corresponding protein was demonstrated using Western blot. Results. A significant increase in proliferation (p = 0.05) was detected after exogenous stimulation with PDGF-AA (10ng/ml and 100ng/ml), PDGF-AB (10ng/ml and 100ng/ml), PDGF-BB (10ng/ml and 100ng/ml), bFGF (100 ng/ml), IL-1ß (1 ng/ml and 10 ng/ml), TGF-ß1 (0.5 ng/ml) and TGF-ß2 (0.5 ng/ml). Both TGF-ß1 and TGF-ß2 stimulated expression of a-SMA in a dose dependent manner with peak activity at a concentration of 50 ng/ml (TGF-ß1) and 500 ng/ml (TGF-ß2). Protein and mRNA of PDGF-receptor type a and type ß and TGF-ß-receptors type I, II and III are expressed in cultured human Tenon’s capsule fibroblasts. Conclusions. The present investigation strongly supports the hypothesis that PDGF-isoforms are major stimulators of proliferation of Tenon’s capsule fibroblasts after glaucoma filtering surgery while TGF-ß-isoforms are essential for the transformation of Tenon’s capsule fibroblasts into myofibroblasts.

Formation of caspase-3 complexes and fragmentation of caspase-12 during anisomycin-induced apoptosis in AKR-2B cells without aggregation of Apaf-1.

Treatment of AKR-2B fibroblasts with anisomycin (10 microM) led to a rapid disintegration of the cells (t1/2 = 5 h) which was complete after 24 h. Cell death was associated with typical hallmarks of apoptosis like membrane blebbing, exposure of phophatidylserine on the cell surface, nuclear condensation and specific cleavage of rRNA. However, there was no dissipation of the mitochondrial potential and no intranucleosomal fragmentation. By affinity labeling with YVK(-bio)D.aomk in combination with immunostaining against activated caspase-3 analyzed by 2-D gel electrophoresis it was shown that caspase-3 is the dominant executioner caspase. Gel filtration experiments of cytosolic extract analyzed by Western blotting revealed the formation of high-molecular-weight complexes of caspase-3 (600 kDa and 250 kDa, respectively), but there was no complex formation of Apaf-1. Anisomycin treatment led to a strong activation of the stress kinases p38 kinases and the jun kinases, that was not sufficient for the activation of caspase-3 which required much higher concentrations. By using the selective inhibitors SB 203580 for p38 kinases and SP 600125 for c-jun kinases, respectively, it is shown that activation of these kinases is not necessary for cell death induced by anisomycin in AKR-2B cells. Furthermore, we disclose the activation of caspase-12 in AKR-2B cells following the addition of anisomycin. Caspase-12 zymogen present as a cytosolic complex (> 600 kDa) is activated by anisomycin leading to an uncomplexed cleaved enzyme. Since anisomycin treatment did neither lead to stress of the endoplasmic reticulum nor to a breakdown of intracellular Ca(2+)-stores, alternative pathways involved in the activation of caspases are discussed.

Mutations dislocate caspase-12 from the endoplasmatic reticulum to the cytosol

Mouse AKR‐2B cells express two forms of caspase‐12: the full‐length form coding for a protein of 47.8 kDa and a new splice variant of 40.2 kDa which is devoid of the CARD domain. In addition, three point mutations were disclosed: I/L‐15, E/D‐46 and P/L‐105. A major portion of the two protein variants was found in the cytosol. Immunofluorescence studies showed an even distribution of caspase‐12 within the cell, indicative for a cytoplasmatic localization. Transfection of AKR‐2B cells with wild‐type caspase‐12 showed a colocalization of this protein with the endoplasmic reticulum (ER). Unlike mouse embryonal fibroblasts (MEF) which contain wild‐type caspase‐12, AKR‐2B cells were largely resistant against treatment with the endoplasmatic reticulum stressing reagents brefeldin and tunicamycin. In AKR‐2B cells, cytoplasmatic caspase‐12 is bound to high molecular weight complexes of >1000 kDa [Cell Death Differ. 9 (2001) 125] and serum depletion leads to cleavage and detachment of caspase‐12 from this high molecular weight complex. Cleavage of caspase‐12 and ‐3 occurred almost simultaneously reaching a maximum 3–5 h after serum deprivation at which time also maximum apoptosis is found. Analysis of caspase‐12 cleavage in vitro in comparison with fragmentation in vivo suggests that during death in AKR‐2B cells induced by starvation, cleavage was brought about by caspase‐3 at positions D24 and D94. Thus, mutated caspase‐12 is differently integrated in signaling pathways of cell death and has lost its function as initiator caspase upon ER‐stress. Instead, it is turned into a substrate of effector caspases. The implication of these findings in the pathological phenotype of ARK‐2B mice is discussed.

ATP and adenosine prevent via different pathways the activation of caspases in apoptotic AKR-2B fibroblasts.

Confluent AKR-2B fibroblasts rapidly disintegrate after serum deprivation. ATP or adenosine added immediately after serum removal afforded substantial protection against cell death even for a long period of 24 h. ED50 values were 14 and 110 μM for ATP and adenosine, respectively. In the presence of 5 μg/ml cycloheximide the protective effect of both substances was suppressed, indicating that protein synthesis is required. The protective effect of ATP was highly specific since among numerous tested derivatives only ATP-[γ-S] exhibited a substantial protective effect.The ability of ATP and adenosine to modulate cell division was analyzed. Both substances did not exhibit any mitogenic effect. Adenosine completely blocked PDGF-BB induced cell division, whereas ATP had no effect. Unlike adenosine, ATP strongly stimulated Ca2+-release from intracellular stores. On the other hand, adenosine stimulated an increase in the intracellular concentration of cAMP from 0.4–1.5 μM, whereas ATP decreased the content below 0.1 μM. ATP stimulated the phosphorylation of MAP-kinase, RSK and p70S6-kinase; adenosine was inactive. After complexation of [Ca2+]i the protective effect of ATP was greatly lost while adenosine was still active. Surprisingly neither ATP nor adenosine caused an activation of PKC-isoforms. After incubation with pertussis toxin, the protection by ATP was reduced indicating an involvement of Gi-proteins in the signal transduction induced by ATP. Our results indicate that ATP as well as adenosine are potent inhibitors of cell death caused by serum deprivation and that this protective effect apparently occurs via distinct pathways. However, both pathways must converge at the point of caspase activation, since the stimulation of DEVDase- and VEIDase-activities, respectively, are suppressed by either ATP or adenosine.