Gap-junction quantification in biological tissues: freeze-fracture replicas versus thin sections

The shaking B gene in Drosophila regulates the number of gap junctions between photoreceptor terminals in the lamina

The molecular structure of insect gap junctions differs from that in vertebrates, and in Drosophila is possibly encoded by the shaking B (= Passover) locus. shaking B2 is a null allele that acts in the nervous system. In the shakB2 mutant, one site of action are gap junctions between photoreceptor terminals in the cartridges of the lamina, beneath the compound eye, which we assayed from the number of close-apposition profiles in thin-section EM. The number of profiles in the Canton-S (C-S) wild type is about 0.5 per cartridge per section in distal and mid-lamina depths, and significantly less, about one quarter this value, closer to the brain, in the proximal lamina. In shakB2, there are fewer profiles, approximately one quarter the number of appositions in distal and mid-lamina depths as in C-S, and their number does not differ significantly from those at the proximal depth in either the mutant or wild type. Thus mutant action is associated with a reduced number of appositions at distal and mid-lamina depths. We propose that R1-R6 gap junctions are partitioned into at least two strata, proximal and distal, and that two populations of gap junctions exist, one extending throughout the lamina that does not require shakB, and a second at distal and mid-depth levels, which does. The number of gap junctions is reduced in mutant shakB2, and surviving appositions at distal and middle lamina depths possibly have wider clefts than in C-S. Gap junctions are reduced equally between all R1-R6 terminals, so the two different types of junction proposed, shakB2- and non-shakB2-dependent, can apparently express in a single receptor terminal.

Gap junctions and growth control in liver regeneration and in isolated rat hepatocytes

The hepatocytes in the mature normal liver are tightly coupled through gap junctions, except during compensatory hyperplasia (regeneration) after partial hepatectomy when the gap junctions become down-regulated. The significance of this down-regulation has been a long-standing enigma. The present study of hepatocytes in primary culture and in the regenerating liver aimed at defining the relationship, if any, between hepatocyte gap junctional communication and proliferation. Gap junctional down-regulation in the regenerating liver appeared to be a specific phenomenon because desmosomes and the surface contact area between neighboring hepatocytes remained constant. All agents and conditions (dexamethasone in vivo; dexamethasone, cyclic adenosine monophosphate, serum, and high cell density in vitro) delaying gap junctional down-regulation also increased the lag before the cells reached competence to enter S phase. This raised the possibility that hepatocyte DNA replication was inhibited through preservation of gap junctions. However, we disproved this assumption by showing that the DNA replication (more specifically the G1/S transition rate constant) was inhibited even in hepatocytes completely devoid of gap junctional communication. The teleological advantage of linking gap junctional down-regulation to hepatocyte G1 progression therefore may not be to trigger DNA replication but to ensure that proliferating hepatocytes and hepatocytes responsible for liver-specific metabolic functions maintain separate pools of metabolites and signaling molecules.

An ultrastructural analysis of the epithelial-fiber interface (EFI) in primate lenses

The purpose of this study was to conduct a comprehensive ultrastructural analysis of the epithelial-fiber interface (EFI) in normal adult primate (Macaque nemestrina and fascicularis; 6-9 years old, n = 10) lenses. Scanning electron microscopy (SEM) was used to initially characterize the gross size, shape and three-dimensional organization of central zone (cz) epithelial cells and the anterior ends of elongating fibers beneath these cells. This fiducial information was essential to properly orient lens pieces in freeze fracture specimen carriers for the production of replicas with unambiguously identifiable EFI. Transmission electron microscopy (TEM) of replicas and thin-sectioned material were used to ultrastructurally analyse the cz EFI. TEM thin-sectioned material was also used to ultrastructurally analyse the pregerminative (pgz), germinative (gz) and transitional zone (tz) EFI. Correlative SEM and TEM of cz EFI components revealed that the apical membrane of both epithelial and elongating fiber cells were irregularly polygonal in shape, and aligned in parallel as smooth, concave-convex surfaces. However, whereas epithelial cell apical surfaces had minimal size variation, elongating fibers were larger and considerably variable in size. Quantitative analysis of > 10000 micron2 cz elongating fiber apical surfaces failed to detect any gap junctions defined in freeze fracture replicas as complementary aggregates of transmembrane proteins (connexons) conjoined across a narrowed extracellular space. However, a comparable frequency of vesicular events was noted in this region as quantified previously in adult and embryonic chick lens. Correlative TEM analysis > 1500 linear micrometers of thin-sectioned EFI from this region confirmed the presence of epithelial-epithelial gap junctions, elongating fiber-elongating fiber gap junctions, and an extreme paucity of epithelial-elongating fiber gap junctions. In contrast, TEM analysis of > 1000 linear micrometers of thin-sectioned pgz, gz and tz EFI, confirmed the presence of epithelial-epithelial gap junctions, elongating fiber-elongating fiber gap junctions, numerous epithelial-elongating fiber adherens junctions and a few epithelial-elongating fiber gap junctions. Thus, the results of this and previous quantitative morphological and physiological studies (electronic and dye coupling) demonstrate that there is limited coupling between cz epithelial cells and underlying elongating fibers. Furthermore, the absence of gap junctional plaques in cz EFI freeze-fracture replicas and either pentalaminar or septalaminar profiles in correlative thin-sections, suggests that this limited coupling could be mediated via isolated gap junction channels. However, the results of this and previous quantitative studies further show that a greater degree of coupling exists across the pgz, gz and tz regions of the EFI and that this coupling is likely to be mediated by gap junction plaques. Finally, this and other studies continue to demonstrate that transcytotic processes play a role in lens physiology at the EFI.

Immunocytochemical, electrophoresis, and immunoblot analysis of Heliothis virescens gap junctions isolated in the presence and absence of protease inhibitors

Gap junction-enriched fractions were prepared from larvae of the tobacco budworm Heliothis virescens using the NaOH procedure in the presence or absence of protease inhibitors and were analyzed by SDS-PAGE immunoblotting and EM immunocytochemistry. Protease inhibitor fractions contained a 48-kDa protein in addition to the approximately 10 proteins in fractions with and without inhibitors. Three polyclonal antibodies were used as probes for gap junction plaques and proteins: R16, against an approximately 40-kDa candidate gap junction protein from Drosophila melanogaster; R17, against the 40-kDa candidate gap junction protein from H. virescens; and R18AP, an affinity purified antibody against a consensus sequence of N-terminal amino acids 2-21 of the H. virescens 40-kDa protein. R16, R17, and R18AP stain the 40- and 48-kDa proteins, R16 and R18AP stain a 64-kDa protein, and R16 stains an approximately 30-kDa protein in the absence of inhibitors. Inclusion of protease inhibitors had no effect on gap junction ultrastructure. R16 and R17 label gap junction plaques in crude membrane and NaOH fractions, whereas R18AP exhibits only a low level of reactivity with gap junctions in crude membrane fractions and none with gap junctions in NaOH fractions. The results show that the 30-, 40-, 48- and 64-kDa proteins are immunologically related and are associated with gap junctions in H. virescens, the N-terminus of the 40-kDa protein is relatively inaccessible or easily lost, and the 48-kDa protein is protease-sensitive.

Gap junction protein tissue distribution and abundance in the adult brain in Drosophila

The distribution of gap junction (GJ) protein in Drosophila tissues and developmental stages was determined by probing immuno-blots with an anti-Drosophila GJ protein antibody (R2AP18). All tissues and developmental stages examined contained 18, 24 or 72 kD GJ protein. GJ protein was notably abundant in immuno-blots of homogenates of adult brain tissue. This was confirmed by the direct visualization of GJs in thin sections of adult brain by electron microscopy. GJs were particularly large and numerous between glial cells in the optic lobes and peripheral glial sheath. R2AP18 reactivity was used to identify GJ protein in immunoblots of cell fractions from isolated adult heads. The final GJ-enriched pellets, derived by extracting crude membrane fractions with urea and N-lauroyl sarcosine, contained GJs with reduced profile widths (13-15 nm vs 16-18 nm for native GJs) and which, unlike native GJs in the crude membrane fractions, were immuno-labelled by R2AP18. Immuno-blots of the urea-sarcosine extracted GJ pellets and supernatant contained higher molecular weight R2AP18 immuno-reactive proteins in addition to the 18 kD form which was present in the tissue homogenate and crude membrane fractions. The results confirm previous observations that urea-sarcosine causes alterations in GJ structure and suggest that urea-sarcosine treatment exposes antigenic determinant(s) which are unavailable for R2AP18 binding in non-extracted native GJs. The abundance of GJs in the adult brain and the relatively simple R2AP18 staining patterns in immuno-blots of GJ-enriched fractions from isolated adult heads suggest that this tissue will be useful for further biochemical and molecular studies of GJs in Drosophila.