Vitellogenesis in the fruit fly, Drosophila melanogaster: antagonists demonstrate that the PLC, IP3/DAG, PK-C pathway is triggered by calmodulin.

In Drosophila melanogaster M. (Diptera: Drosophilidae), a phospholipase-C to proteininase-C signal cascade leads to the endocytic uptake of yolk precursor molecules. The data suggest that D. melanogaster has a phospholipase-C/proteinkinase-C signaling pathway similar to that previously shown to be required for vitellogenesis in the milkweed bug, Oncopeltus fasciatus Dallas (Hemiptera: Lygaeidae). Calmodulin, derived from epithelial cells and transported to the oocytes via gap junctions, may trigger this pathway. To investigate this, a series of known antagonists to various elements of the pathway were used. W-7 (which prevents calmodulin binding to phospholipase-C), U-73122 (which prevents activation of phospholipase-C), verapamil (which blocks Ca(2+) release by IP3), HAG (which blocks diacylglycerol), and staurosporine (which inactivates proteinkinase-C) were each shown to inhibit endocytosis, thereby blocking formation of nascent yolk spheres.

Role of adherens junction proteins in differential herpes simplex virus type 2 infectivity in communication-competent and -deficient cell lines.

BACKGROUND: Gap junctional intercellular communication decreases with HSV-2 infection. To determine the importance of functional gap junctions for infectivity, we compared HSV-2 growth in communication-competent and -deficient cell lines. METHODS: HSV-2 infectivity was tested in five cell lines: WB rat liver epithelial cells (communication-competent), WB-aB1 (communication-deficient), WB-a/32-10 (communication-rescued), HeLa (communication-deficient), and Cx43-transfected HeLa (communication-rescued) cells. HSV-2 growth curves and indirect immunofluorescence labeling of viral and cell proteins were performed in wild-type and mutant WB cells. RESULTS: Although wild-type WB cells were highly permissive for HSV-2 infection, virus production was significantly attenuated in communication-deficient and -rescued mutant WB cells. HeLa exhibited no difference in infectivity between communication-competent and -deficient cell lines. Tight and adherens junction proteins, including zonula occludens-1 and nectin-1, were not different in the WB cell lines. However, E-cadherin levels were elevated and ß-catenin was found to co-localize with glycoprotein E, a viral glycoprotein associated with cell-to-cell spread, in the mutant WB cells. CONCLUSIONS: These results suggest that attenuated viral production in mutant WB cells is due to viral protein co-localization with adherens junction proteins rather than the loss or restoration of functional gap junctions.

Vitellogenesis in Oncopeltus fasciatus: PLC/IP(3), DAG/PK-C pathway triggered by CaM.

In Oncopeltus fasciatus, evidence shown here indicates it is calmodulin (CaM) that activates phospholipase-C (PLC), beginning a signalling pathway necessary for endocytic uptake of yolk precursor molecules. Epithelial cell-produced CaM, transported to oocytes via gap junctions, has been shown to be required for receptor-mediated endocytic uptake of vitellogenins (Vgs, the protein precursors of yolk). To determine if CaM was directly or indirectly stimulating the phospholipase-C (PLC) signalling cascade and thus controlling Vg endocytosis we used a series of molecules known to inactivate various elements of the pathway. W-7 prevents CaM from interacting with other molecules. Neomycin isolates PIP(2) from PLC. U-73122 directly inactivates PLC. 2-APB blocks IP(3) receptors which would otherwise cause release of Ca(2+). Verapamil and CdCl(2) block Ca(2+) release channels. Staurosporin and calphostin are inhibitors of PK-C. 1-Hexadecyl-2-acetyl glycerol (HAG) binds to diacylglycerol (DAG). Through the use of these antagonists we show here that: (1) the activation of phospholipase-C in this system requires CaM. (2) Stimulated phospholipase-C converts PIP(2) into IP(3) and DAG. (3) IP(3) causes increase in cytosolic Ca(2+). (4) DAG and Ca(2+) each stimulate phosphokinase-C, resulting in endocytosis of Vgs.

Fractional contribution of major ions to the membrane potential of Drosophila melanogaster oocytes.

In ovarian follicles of Drosophila melanogaster, ion substitution experiments revealed that K(+) is the greatest contributor (68%) in setting oocyte steady-state potential (E(m)), while Mg(2+) and a metabolic component account for the rest. Because of the intense use made of Drosophila ovarian follicles in many lines of research, it is important to know how changes in the surrounding medium, particularly in major diffusible ions, may affect the physiology of the cells. The contributions made to the Drosophila oocyte membrane potential (E(m)) by [Na(+)](o), [K(+)](o), [Mg(2+)](o), [Ca(2+)](o), [Cl(-)](o), and pH (protons) were determined by substitutions made to the composition of the incubation medium. Only K(+) and Mg(2+) were found to participate in setting the level of E(m). In follicles subjected to changes in external pH from the normal 7.3 to either pH 6 or pH 8, E(m) changed rapidly by about 6 mV, but within 8 min had returned to the original E(m). Approximately half of all follicles exposed to reduced [Cl(-)](o) showed no change in E(m), and these all had input resistances of 330 kOmega or greater. The remaining follicles had smaller input resistances, and these first depolarized by about 5 mV. Over several minutes, their input resistances increased and they repolarized to a value more electronegative than their value prior to reduction in [Cl(-)](o). Together, K(+) and Mg(2+) accounted for up to 87% of measured steady-state potential. Treatment with sodium azide, ammonium vanadate, or chilling revealed a metabolically driven component that could account for the remaining 13%.

Importance of molecular configuration in gap junctional permeability.

Gap junctions between insect oocytes and follicular epithelial cells allow transit of elongate Ca(2+)-binding proteins Calmodulin (CaM, 17 kDa) and Troponin-C (Trop-C, 18 kDa), but not multi-branched dextran (10 kDa) nor the Ca(2+)-binding protein Osteocalcin (Osteo, 6 kDa). By microinjection of fluorescently labeled versions of each of these molecules we were able to obtain visual evidence that, despite their lesser molecular weight, molecules with greater cross-sections were unable to transit these gap junctions, while heavier but elongate molecules could. While CaM had previously been shown to pass through gap junctions from oocytes to their surrounding epithelial cells, the ability of CaM and Trop-C to transit the gap junctions between adjacent epithelial cells had not been demonstrated. Evidence shown here demonstrates that the homologous gap junctions among epithelial cells, like the heterologous gap junctions between epithelial cells and the oocyte they surround, allow transit of elongate molecules up to at least 18 kDa. Furthermore, the evidence for four different molecules of differing molecular weights and configurations supports the hypothesis that it is molecular configuration, not chemical activity, that primarily determines the observed permeability of gap junctions to molecules 5-6 times larger than the molecular weight limit previously acknowledged.

Herpes simplex virus-type 2 infectivity and agents that block gap junctional intercellular communication.

In rat liver epithelial (WB) cells, the protein kinase C inhibitor H7 blocked gap junctional intercellular communication (GJIC) and reduced virus infectivity. Octanol, 18-beta-glycyrrhetinic acid, and staurosporine, agents that reduce GJIC, had no effect upon virus infectivity. Previous studies demonstrated that herpes simplex virus-type 2 (HSV-2) infection was accompanied by attenuated GJIC. Of agents tested, only H7 reduced plaque forming unit (pfu) ability in a dose-dependent manner with 100% plaque reduction at 40 microM without evidence of cytotoxicity. Dye transfer indicated that H7 decreased GJIC, although Western blotting revealed that it did not alter phosphorylation of the gap junction protein, connexin 43 (Cx43). Using indirect immunofluorescence, Cx43 was found to localize in membrane plaques in uninfected cells and H7 did not alter this distribution. However, Cx43 was lost from the membrane at 24h in both H7-treated and untreated cells infected with HSV-2. Viral infection increased serine phosphorylation, particularly in the nuclear region, and this effect was reduced following H7 treatment. Thus, H7 attenuated both GJIC and infectivity of HSV-2 in WB cells but the anti-viral effects were due to reduced nuclear protein phosphorylation rather than alterations in phosphorylation or localization of Cx43.

Calmodulin transit via gap junctions is reduced in the absence of an electric field.

Gap junctional transport of Calmodulin (CaM) from epithelial cells to insect oocytes is enhanced by alignment of the molecules via an electric field. It has recently been shown that CaM is needed for uptake of vitellogenins, is produced in the epithelial cells and reaches oocytes via gap junctions. For CaM to transit the gap junctions something must align these elongated molecules with the lumina of the gap junctions. This might be accomplished by the electric field that exists at the membrane of any cell with an Em of >0 mV. Fluorescently labeled CaM was injected into oocytes. At t=0, the epithelial cell/oocyte "fluorescence" ratio showed epithelial cells to be 24%+/-1.5% as bright as the injected oocyte. In follicles which maintained an electric field for one hour the epithelial cell/oocyte fluorescence ratio had risen to 79%+/-1.4%, while for follicles in which the field was cancelled by holding Em at 0 mV the ratio was only 45%+/-1.7%. After termination of the holding current follicles regained their original Em and their original electric field. At the end of a second hour of incubation the ratio had risen to 76%+/-1.2%, very close to what was observed in the untreated control follicles.

Calmodulin transmitted through gap junctions stimulates endocytic incorporation of yolk precursors in insect oocytes.

In ovarian follicles of Oncopeltus fasciatus, and of Xylocopa virginica, calmodulin (CaM) of epithelial cell origin is required by oocytes for endocytic uptake of yolk precursor molecules. Furthermore, this 17-19 kDa protein is normally transported to the oocytes via gap junctions. Downregulation of gap junctions by treatment with 1 mM octanol or separation of the epithelial cells from their oocytes terminated precursor uptake, and this activity could be rescued by microinjection of 60 microM CaM, but not by injections of incubation medium, nor solutions of other molecular species tested. That endogenous CaM is required was confirmed by incubating otherwise untreated follicles in physiological salt solution (PSS) containing either calmidazolium or W-7, both known antagonists of CaM. By radioimmunoprecipitation, we show that the epithelial cells surrounding an oocyte synthesized 15 times as much calmodulin as did the oocytes they encircled. Neither octanol-treated follicles nor denuded oocytes incubated in medium containing calmodulin were able to resume endocytosis, arguing against an extracellular route. However, fluorescently labeled calmodulin microinjected into oocytes is shown to have crossed through gap junctions, making epithelial cells distinctly fluorescent.

Antiviral agents alter ability of HSV-2 to disrupt gap junctional intercellular communication between mammalian cells in vitro.

In cultured mammalian cells (Vero), different antiviral agents change to differing degrees the ability of HSV2 to down-regulate gap junctions, each agent having a specific effect. Measured by intracellular electrodes, control cell populations showed 49-51% coupling, uninfected populations treated with acyclovir or SDS averaged 43-51% coupling while populations infected with HSV2 had coupling reduced to 8%. The antiviral agent acyclovir (1 microg/ml), which suppresses viral replication, failed to prevent this down regulation (final coupling ratio of 11%). A plant extract (250 microg/ml) from Pilostigma thonningii offered slightly more protection (final coupling ratio of 22%), while sodium dodecyl sulfate (SDS) (50 microM) afforded nearly complete protection (final coupling ratio of 40%). With SDS there was an initial down regulation to only 16% coupling by 3 h post infection, followed by a recovery of intercellular communication to near control levels by 24 h. While SDS was originally believed to alter the viral coat and prevent entry into the cell, our data are in agreement with recent studies which indicate that SDS treated viruses can enter into host cells, but in a severely diminished condition. Our data also suggest that the gap junction antagonist is brought into the cells as part of the entering virus.

Ion physiology of vitellogenic follicles.

The ion physiology of vitellogenic follicles from a lepidopteran (Hyalophora cecropia) and a hemipteran (Rhodnius prolixus) are compared. Similarities that can be expected to occur in vitellogenic follicles of many other insects include: (1) gap junctions, which unite the cells of a follicle into an integrated electrical system, (2) transmembrane K(+) and H(+) gradients that account for over 60% of follicular membrane potentials, (3) absence of a Cl(-) potential, (but the opening of channels to this anion when vitellogenesis terminates in H. cecropia), (4) an electrogenic proton pump that supplements follicular membrane potentials, (5) Ca(2+) action potentials evoked by injecting depolarizing currents into oocytes, and (6) the use of osmotic pressure to control epithelial patency. Differences include: a Na(+)/K(+)-ATPase that accounts for about 20% of the follicular resting potential in R. prolixus but is absent from H. cecropia, and an intrafollicular Ca(2+) current that moves from oocyte to nurse cells through cytoplasmic bridges in H. cecropia. Evidence is also summarized for two promising mechanisms that require further substantiation: (1) transmission via gap junctions of a follicle cell product that promotes endocytosis in the oocyte; and (2) transport of the proton pump back and forth between cell surface and endosomes as the membrane that carries it recycles through successive rounds of vitellogenin uptake.

For uptake of yolk precursors, epithelial cell-oocyte gap junctional communication is required by insects representing six different orders.

For uptake of vitellogenin protein into nascent yolk spheres, communication through open gap junction channels between the follicle epithelium and oocyte is required by six different insects representing six different orders. It was recently shown in the hemipteran, Oncopeltus fasciatus, that endocytic uptake of yolk protein resulting in the formation of nascent yolk spheres depended upon an intact epithelium communicating with the oocyte through patent gap junctions. Following treatment with octanol, which down-regulated gap junctions below the level of dye coupling, vitellogenin uptake was terminated. Yet, for another hemipteran, Dysdercus intermedius, it has been shown that yolk spheres can form even when all epithelial cells have been stripped from the oocyte. To determine if the mechanism seen in Oncopeltus is present in other insects, we utilized the same techniques to study nascent yolk sphere production in a dipteran, Drosophila melanogaster, a lepidopteran, Actias luna, a hymenopteran, Xylocopa virginica, a coleopteran, Tenebrio molitor and an orthopteran, Acheta domesticus. In each of these, when gap junctions were down-regulated yolk uptake quickly stopped. That six different insects from six different orders all required a gap junctionally transmitted chemical signal of epithelial cell origin suggests that this mechanism is widespread throughout the insects.

Charge-Dependent Molecular Movement Through Intercellular Bridges in Drosophila Follicles.

The steady-state membrane potentials of the nurse cells of Drosophila melanogaster incubated in a Ringer's solution containing 2.5 mM K+ averaged 5.2 mV more negative than those of oocytes to which they were attached by intercellular bridges (n = 159; P < 0.001, paired t-test). In sodium azide, the resting potentials of the two cells diminished from a range of -50 to -70 mV to an average of -10 mV, and the transbridge difference disappeared. Therefore, the transbridge difference is dependent upon metabolic energy. Lucifer yellow CH, when microinjected into either nurse cells or oocytes, became distributed in less than 30 min in a manner to be expected for a negatively charged fluorochrome in an electrically polarized system. Polarity was even more striking in this study than reported earlier for Drosophila follicles incubated in a higher K+ medium (Woodruff et al., 1988).

Electrically mediated protein movement inDrosophila follicles.

Distribution of rhodamine-conjugated lysozyme injected into the sixteen-cell syncytium comprising the germ-line portion of theDrosophila follicle is shown to be affected by charge. Positive molecules are able to migrate through intercellular bridges from the oocyte to the nurse cells, but are unable to migrate detectably from nurse cells to the oocyte. Their negatively charged counterparts can move from the nurse cells to the oocyte, but are unable to traverse the intercellular bridges in the counter direction. This charge-dependent movement of molecules is accompanied by an electrical potential difference, focused across the nurse cell-oocyte bridges, which makes the nurse cells negatively charged to the oocyte. The addition of insect hemolymph to the physiological salt solution in which the experiments were performed resulted in only a small increase in the transmembrane resistance, but enhanced the potential difference between oocyte and nurse cells from 0.2±0.3 (SE) mV (nurse cells negative) to 2.3±0.45 (SE) mV (nurse cells negative).

Nutritive cord connection and dye-coupling of the follicular epithelium to the growing oocytes in the telotrophic ovarioles inOncopeltus fasciatus, the milkweed bug.

Microinjected dyes were used to study transport routes between cells in the telotrophic ovarioles of the milkweed bug,Oncopeltus fasciatus. Dye movement revealed that transport through the nutritive cords is not dependent upon microtubules, and that the physiological continuity of the cords between the nurse cells and the oocyte is lost shortly after the onset of vitellogenesis. Dye-coupling of epithelial cells to the oocyte is shown to begin in a band of cells at the follicle equator. Coupling begins coincidentally with cord severance and the onset of vitellogenesis.