Passage through vertebrate gap junctions of 17/18kDa molecules is primarily dependent upon molecular configuration.

In fish, amphibians and mammals, gap junctions of some cells allow passage of elongate molecules as large as 18kDa, while excluding smaller, less elongate molecules. Fluorescently labeled Calmodulin (17kDa) and fluorescently labeled Troponin-C (18kDa), when microinjected into oocytes of Danio rerio, Xenopus laevis or Mus domestica, were able to transit the gap junctions between these oocytes and the granulosa cells which surrounded them. Co-microinjected with these Ca(2+)-binding proteins, Texas-red-labeled dextran (10kDa) remained in the microinjected cell. Osteocalcin (6kDa), also a Ca(2+)-binding protein, but with a wide "V" shape proved unable to transit these gap junctions. Calmodulin, but not Troponin-C, was able to transit gap junctions of gonadotropin treated WB cells in culture. We show evidence that molecules as large as 18kDa can pass through some vertebrate gap junctions, both homologous and heterologous, and that it is primarily molecular configuration which governs gap junctional permeability.

Passage of 17 kDa calmodulin through gap junctions of three vertebrate species.

Gap junctions of some vertebrates are capable of passing the elongate molecule, calmodulin, with a molecular weight 8-17 times greater than the previously recognized size limits. Fluorescently labeled calmodulin (FCaM) (17.34 kDa) microinjected into oocytes of ovarian follicles from an amphibian, Xenopus laevis, and from two species of teleost fish, Danio rerio (Zebrafish) and Oryzias latipes (Medaka), is shown to transit their gap junctions and enter the surrounding epithelial cells. Passage of FCaM was terminated when follicles were first treated with 1 mM octanol, a molecule known to down-regulate gap junctions. There was no FCaM detected in the surrounding medium, nor did epithelial cells become fluorescent when follicles were incubated in medium containing dye. Calmodulin is well known to modulate many cytoplasmic reactions; thus, its passage through gap junctions opens possibilities of additional means by which cells may be supplied with this signaling molecule, and by which their supply may be regulated.

HSV-2 disrupts gap junctional intercellular communication between mammalian cells in vitro.

Infection by herpes simplex virus-2 (HSV-2) disrupts both dye and electrical coupling in Vero (African green monkey kidney) cell cultures. Vero cells in vitro were iontophoretically injected with the fluorescent dye Lucifer yellow CH, the spread of which revealed that cells throughout the confluent sheet shared open gap junctions. However, 24 h after infection with the virus (but before cells became rounded), dye always remained only within the target cell. Intracellular electrophysiological measurements of ionic coupling revealed a 0.4 coupling coefficient for adjacent cells in uninfected control cultures. By 3 h following infection significant down-regulation of gap junctions had begun, preceding by many hours any signs of infection visible with the light microscope. Measurements between adjacent cells 3 h post-infection, a period when HSV-2 gene expression is known to be at a maximum, yielded an average coupling coefficient of 0.35. By 6 h post-infection (a period of known viral DNA replication) average coupling coefficient for adjacent cells was 0.25, while by 24 h post-infection the average fill still further to <0.08. A coupling coefficient of <0.08 suggests that infection by HSV-2 completely disabled the gap junctions.

A gap junctionally transmitted epithelial cell signal regulates endocytic yolk uptake in Oncopeltus fasciatus.

For endocytic uptake of vitellogenins, developing oocytes of Oncopeltus require a soluble, diffusible molecular signal from their surrounding epithelial cells, and this signal must be transmitted through open gap junctions. Hormonal stimulation triggering synthesis and processing of vitellogenins into mature insect yolk spheres has been intensely studied, and follicle epithelial cells are known in several insects to contribute to the blood products which are endocytosed along with vitellogenins synthesized in the fat bodies. However, there has been little evidence that direct gap junctional communication is a requirement for endocytic activity by oocytes. In untreated control follicles, both electrical and dye coupling occur, and follicles incubated in vitro in physiological salt solution containing small amounts of blood and fluorescent dye produce fluorescently labeled nascent yolk spheres. Labeled yolk spheres were visible in both sectioned material, and, with (Laser) Confocal Scanning, in living material. Dye coupling was abolished by treatment with either 1 mM octanol, 0.5 mM ethyl methanesulfonate (EMS), or cytoplasmic acidification, with coupling coefficients also being affected as each of these gap junction antagonists down-regulated the connexons. With each of these treatments, after gap junctions were down-regulated, receptor-mediated endocytic uptake of blood-born vitellogenins came to a halt. Furthermore, Oncopeltus follicles with endocytic activity blocked in this manner could be rescued by microinjection of the soluble fraction of lysed epithelial cell cytoplasm, confirming that the process depended upon a molecular signal from the epithelial cells.

Varied effects of 1-octanol on gap junctional communication between ovarian epithelial cells and oocytes of Oncopeltus fasciatus, Hyalophora cecropia, and Drosophila melanogaster.

In insect gap junctions, species-specific differences occur in response to the purported gap junction uncoupling agent, 1-octanol. Changes in gap junctional communication between oocytes and their epithelial cells following treatment with 1-octanol were assayed in Oncopeltus fasciatus (the milkweed bug), Hyalophora cecropia (the American silk moth), and Drosophila melanogaster. In all three species, microinjection of untreated control follicles with Lucifer yellow CH revealed extensive dye coupling among epithelial cells and between epithelial cells and their oocytes. Also for all three species, treatment with octanol appeared to completely block dye coupling and increase oocyte input resistance. The effect on electrical coupling varied. In Drosophila, octanol diminished the electrical coupling from 64% (0.64 coupling coefficient) in controls to 53% in treated follicles. In Hyalophora, the coupling ratio remained the same following treatment. In Oncopeltus, octanol actually increased the electrical coupling ratio from 84% in controls to 94% in treated follicles. While 0.5 mM octanol left some Oncopeltus epithelial cells dye coupled to the oocyte, the electrical coupling ratio was increased slightly more by this concentration than by 1 or 5 mM octanol solutions, although the differences were not significant. While input resistance (R(o )) increased in all three following treatment with octanol, there was considerable difference in the magnitude of the response. Average oocyte R(o ) for Oncopeltus increased the least of the three species, rising from 196-240 kOhm. Both Hyalophora, with a nearly fourfold increase from 230-900 kOhm or more, and Drosophila, with a twofold increase from 701 kOhm to over 1.2 MegOhm showed much larger changes. Results shown here indicate that insect gap junctions have more varied responses to this common gap junction antagonist than have been reported for their vertebrate counterparts. Arch.

Intercellular bridges between epithelial cells in the Drosophila ovarian follicle: a possible aid to localized signaling.

In the epithelium of Drosophila ovarian follicles, cytoplasm-filled intercellular bridges connect epithelial cells. This study presents further descriptive information about the morphology of these intercellular bridges and the extent of their distribution. We also offer speculations concerning the possible developmental importance of the epithelial bridges. These bridges, whose luminal diameters averaged 0.25 microm, are smaller than those forming the ring canals joining germline cells: nor do they increase their size over time. The membranes limiting the bridges are lined on the cytoplasmic side with an electron-dense material to which is attached a monolayer of filaments which encircle the bridge. By decoration with the S1 fragment of myosin, these filaments are confirmed as actin filaments. Following disruption of gap junctional dye coupling by treatment with 1 mM octanol, microinjection of Lucifer yellow CH revealed the extent and distribution of follicle cell intercellular bridges to be confined to arrays of no more than eight cells/cluster, with many such independent clusters comprising the epithelium. Thus cell-to-cell movement throughout the epithelium of cytosolic regulatory molecules cannot occur via these intercellular bridges. However, weak signals affecting only one or a few cells in each cluster would be amplified throughout the group by spread through the intercellular bridges.

Charge Dependent Distribution of Endogenous Proteins within Vitellogenic Ovarian Follicles of Actias luna.

In vitellogenic ovarian follicles of Actis luna, internal Ca(2+) activity currents create an electrical gradient which influences the distribution of charged macromolecules between nurse cells and oocyte. We show that, between oocyte and nurse cells, there is an ionic gradient of 1-12 mV with the nurse cells being more electronegative than the oocyte by an average 3.5+/-0.2 mV(s.e.)(p<0.001). As previously reported for another saturniid, Hyalophora cecropia, the transbridge ionic gradient of luna: (1) is focused across the intercellular bridges, (2) is abolished by 200 &mgr;M vanadate and (3) includes a [Ca(2+)](i) gradient. Endogenous soluble proteins collected from control and from vanadate treated populations of nurse cells and oocytes were separated by two-dimensional (2-D) gel electrophoresis and visualized with sliver stain. Densitometric analysis showed that 14 out of the 19 acidic proteins and six of the eight basic proteins studied, changed their oocyte-to-nurse cell distribution in consort with change in the transbridge ionic gradient. This suggests that a transbridge ionic gradient may be, at least within the saturniidae, a method for maintaining different molecular concentrations in nurse cells compared to oocytes. Copyright 1997 Elsevier Science Ltd. All rights reserved

The osmolarity of adult Drosophila hemolymph and its effect on oocyte-nurse cell electrical polarity.

In ovarian follicles of Drosophila, changes in external osmolarity affect the steady-state potentials of oocytes more than those of nurse cells. Thus the osmolarity of the incubation medium affects the occurrence and the direction of an electrical gradient across the connecting intercellular bridges. At 255 mOsm nurse cell Em averaged 2.5 mV negative to oocyte Em (P < 0.001). At 275 and at 300 mOsm there was no significant difference between oocyte and nurse cell. At 400 mOsm nurse cell Em averaged 1.1 mV positive to oocyte Em (P = 0.007). The osmolarity of adult Drosophila hemolymph was measured by a variation of freezing point depression and averaged 251 +/- 9 (SE) mOsm. The measured osmolarity and measured ionic concentrations of adult Drosophila hemolymph were used to develop an incubation medium, which was used to incubate developing ovarian follicles. Electrical measurements made in this saline, which mimics in vivo conditions, confirmed reports of a nurse cell-oocyte electrical gradient, with nurse cell Em significantly more negative than oocyte Em. Microinjections of the negatively charged dye Lucifer yellow CH showed that this charged molecule accumulated in the oocyte at the in vivo osmolarity, and in the nurse cells at highly elevated osmotic levels.

Steady-state gradient in calcium ion activity across the intercellular bridges connecting oocytes and nurse cells in Hyalophora cecropia.

Intracellular activities of K+, H+, Mg2+, Ca2+, and Cl-, measured with ion selective microelectrodes in the oocyte and the nurse cells in ovarian follicles of Hyalophora cecropia, indicated that a Ca2+ current is a key component of the electrical potential that is maintained across the intercellular bridges connecting these two cells. In vitellogenic follicles, Ca2+ activity averaged 650 nM in the oocyte and 190 nM in the nurse cells, whereas activities of the other ions studied differed between these cells by no more than 6%. Incubation in 200 microM ammonium vanadate caused a reversal of electrical potential from 8.3 mV, nurse cell negative, to 3.0 mV, oocyte negative, and at the same time the Ca2+ gradient was reversed: activities rose to an average 3.0 microM in the nurse cells and 1.6 microM in the oocyte, whereas transbridge ratios of the other cations remained at 0-3%. In immature follicles that had not yet initiated their transbridge potentials, Ca2+ activities averaged approximately 2 microM in both oocyte and nurse cells. The results suggest that vitellogenic follicles possess a vanadate-sensitive Ca2+ extrusion mechanism that is more powerful in the nurse cells than in the oocyte.

Activation of a new physiological state at the onset of vitellogenesis in Hyalophora follicles.

Ovarian follicles of Hyalophora cecropia are shown here to undergo a comprehensive transformation about 2 days after they are first formed and several hours before the first yolk spheres are visible. (1) Electrical coupling initiates between the follicle cells and the oocyte-nurse cell complex, as well as between adjacent follicles. (2) Oocyte and nurse cell membranes begin to hyperpolarize, adding an azide- and vanadate-sensitive component to a basal potential that is unaffected by these inhibitors. (3) The cytoplasmic pH of the oocyte rises from 6.7 to 7.4. (4) The nurse cells hyperpolarize more strongly than the oocyte, so that the charge-dependent restrictions on protein movement across the cytoplasmic bridges that connect these cells arise at this time. (5) The follicle swells and becomes more turgid. (6) Uridine incorporation shuts down in the germinal vesicle and accelerates in the other nuclei of the follicle. The changes are sufficiently synchronous to suggest that they may be responses to a single, branching cascade of activation. The altered cell potentials, cytoplasmic pH, and turgidity implicate the cell membrane in an early stage of activation.

Electrophoresis of proteins in intercellular bridges.

Cytoplasmic polarity giving rise at mitosis to daughter cells with distinct but complementary morphogenetic functions has been proposed by Jaffe and co-workers to have transcellular ion currents as one of its essential physiological steps. The clearest evidence is from fucoid eggs in which such currents have been shown to parallel the prospective axis of germination. The currents are caused by stabilized accumulations of cation pumps on one side of the cell, and of permeability channels on the opposite side; they are strong enough to suggest that the accompanying gradient in electrical potential could have an electrophoretic effect on the distribution of cytoplasmic constituents, including those that serve as morphogenetic determinants. Although ion currents correlated with axes of morphogenesis have thus been clearly established, their effects on intracellular localizations and morphogenetic activity have remained speculative. We report here evidence that an endogenously generated gradient in electrical potential in the oocyte-nurse cell syncytium of Hyalophora cercropia can, in fact, influence the distribution of soluble proteins in a cytoplasmic continuum.

Large electrical currents traverse developing Ceropia follicles.

An intense (up to 20 muA/cm(2)) steady electrical current enters the anterior or nurse cell end of the growing follicle (or oocyte-nurse cell complex) of the Cecropia moth and is balanced by a more diffuse current leaving elsewhere. In late growth stages, the total transfollicular current is about 100 nA. Moreover, a separate small current, of about 1 nA, seems to leave the furrow between the oocyte and the nurse cells. After the nurse cells collapse, but before shell formation, the transfollicular current is redistributed so that a second relatively localized inward current appears at the posterior pole of the follicle. Thus, at this later stage currents enter both poles of the follicle and leave its sides. Previous measurements, with intracellular microelectrodes, seem to imply a very large (order of 1000 nA) back current across the cytoplasmic bridge between the oocyte and nurse cells. A simple model is presented that attributes the apparent bridge current, and the more directly measured transfollicular and furrow currents, to the action of an ion pump lying within the nurse cell face of the furrow membrane.

Polarized intercellular bridges in ovarian follicles of the cecropia moth.

Fluorescein-labeled rabbit serum globulin was injected into vitellogenic oocytes of the cecropia moth. Though the label spread throughout the ooplasm in less than 30 min, it was unable even after 2 h to cross the complex of intercellular bridges connecting the oocyte to its seven nurse cells. After injection into a single nurse cell, fluorescence was detected in the oocyte adjacent to the bridge complex within 3 min and had spread throughout the ooplasm in 30 min. Here also, the cell bodies of the six uninjected nurse cells remained nonfluorescent. Four of the nurse cells are not bridged directly to the oocyte but only through the apical ends of their siblings. Unidirectional movement must therefore occur in the apical cytoplasm of the nurse cells, as well as in the intercellular bridges. The nurse cells of healthy follicles had an intracellular electrical potential -40 mV relative to blood or dissecting solution, while oocytes measured -30 mV. A mV difference was also detected by direct comparison between a ground electrode in one cell and a recording electrode in the other. Three conditions were found in which the 10 mV difference was reduced or reversed in polarity. In all three cases fluorescent globulin was able in some degree to cross the bridges from the oocyte to the nurse cells.