Direct communication of homologous and heterologous endocrine islet cells in culture.

Islet Cell Cultures

Monolayer cultures of islet cells were prepared by a modification of a collagenase-trypsin method (6) as described (18) . Briefly, the pancreases of newborn Wistar rats were enzymatically dissociated by ten successive 10-min incubations at 37C in a Ca"- and Mg"-free phosphate-buffered saline (PBS) containing 5.6 mM glucose, 5 mg/ml trypsin (1 :250, Difco Laboratories, Detroit, Mich.) and 0.2 mg/ml ofcollagenase (CLS IV, Worthington Biochemical Corp., Freehold, N. J.) . The dissociated cells, collected by decantation at the end of THE JOURNAL OF CELL BIOLOGY " VOLUME 92 JANUARY 1982 221-226 (c)The Rockefeller University Press - 0021-9525/82/01/0221/06 $1 .00 RAPID COMMUNICATIONS each ofthe last eight incubations, were pooled, rinsed twice by centrifugation in medium 199 (Gibco Laboratories, Grand Island Biological Co., Grand Island, N. Y.) supplemented with 10% heat-inactivated fetal calf serum and 16.7 mM glucose, resuspended in this medium, and plated at a concentration of 7.5 x 105 cells/ml on 100-mm plastic dishes (Falcon Plastics, Oxnard, Calif) . This primary cell suspension was cultured at 37C in a water-saturated air/C02 (95/5%) atmosphere and decanted twice in newdishes, after 16 and 22 h ofculture. At the time ofthelast decantation, the enriched suspension ofislet cells was transferred to round cover glasses fitted in 35-mm Petri dishes and grown on medium 199. One-week-old cultures were used in the experiments. At that time, islet cells are differentiated as judged by their ultrastructure (15), their hormone content (6, 14), and their normal response to secretagogues (6, 18). Microinjection Experiments Individual islet cells were microelectrophoretically injected with 6-carboxyflu- orescein (Eastman Kodak, Rochester, N. Y.) and observed by fluorescence microscopy, as previously described (4). After removal ofthe microinjectors, the selected cluster was photographed under both phase-contrast and fluorescence microscopy and identified on the cover glass with a diamond marker objective. In some experiments, two or three injections were carried out successively in different islet cells of the same cluster. Forlight microscopy, the microinjected cultures were fixed in Zamboni's fluid (21), rinsed several times in PBS, dehydrated, and subsequently rehydrated through a graded ethanol series. They were then incubated for 2 h at room temperature with an anti-insulin serum (a gift from Dr. P. H. Wright) at a 1:200 dilution in PBS. Aftertwo 5-min washes in PBS, ananti-guineapig IgG antiserum conjugated with horseradish peroxidase (RAGp/IgG (H + Ll PO) (Nordic Immunological Laboratories, Tilburg, The Netherlands) was applied at a 1:50 dilution for l h at room temperature. After washing in PBS, the peroxidase was revealed, according to (3), by incubating the cultures in 0.05 MTris buffer, pH 7.6, containing 100mg% of 3,3'-diaminobenzidine and 30fal of 30% H202. When 221 adequate staining was observed, the cover glass was immersed in distilled water, mounted on a microscope slide, and examined on a Zeiss lCM 405 inverted microscope. For ultrastructural studies, the microinjected cultures were fixed at room temperature in 2.5% glutaraldehyde solution in 0.1 Mcacodylate buffer, pH 7.4, postfixed for 1 h in 1% osmium tetroxide, dehydrated in graded ethanols, and embedded in Epon. After removal of the bottom of the dish by immersion in liquid nitrogen, the identification mark surrounding the injected cluster was localized on the embedded cover glass under a phase-contrast microscope . A small block of Epon supporting the microinjected cluster was then cut from the sample, immersed for 5 min in warm (85C) water and, subsequently, rapidly 222 Number of Number of microin- intracellular Absence of inter- jected clus- microinjec- cellular communi- ters tions cation RAPID COMMUNICATIONS TABLE Presence of intercel- lular communication transferred to liquid nitrogen. This procedure usually resulted in a clean separa- tion of the cover glass from the polymerized plastic without damage to the cells. The injected cluster, sometimes stained in bloc (13), was recognized by compar- ison with the initial photograph taken in the Petri dish and serially sectioned, parallel to the support plane, with a LKB Ultratome (LKB-Products AB, Bromma, Sweden). Sections were collected on single-hole carbon-coated grids, stained with uranyl acetate and lead citrate, and observed in a Philips EM 301 electron microscope.

Identification of Communicating Islet Cells

At the light microscopic level, communicating B-cells were readily identified by comparing the phase-contrast and fluorescence photographs ofeach injected cluster with photographs ofthe same field after immunoperoxidase staining.

I 6-Carboxyfluorescein Microinjections in Monolayer Cultures of Islet Cells

Number of commu- nicating islet cells

Coexistence of communicating and noncommu- nicating islet cells in the same clus- ter* Presence of sepa- rate territories of communicating is- let cells* 96 172 57 (33.1) 115 (66.9) 3 .5 t0.2$ 18 (45) 37 (92.5) * In these columns, values in parentheses are expressed as percent of the number (n = 40) of clusters in which two or more islet cells were successively microinjected. $ Expressed as mean t SEM. Values in parentheses are expressed as percent of the total number of microinjections. FIGURE 1 (a) In this cluster, two different islet cells (asterisks) were successively microinjected with 6-carboxyfluorescein . (b) After both injections, the dye was rapidly transferred from the injected into 2-3 neighbouring islet cells, delineating two separate territories of communicating cells. In this photograph, taken at the end of the experiment, the fluorescence of the first injected territory (right) has slightly faded whereas that of the second territory (left) is bright . (c) After immunoperoxidase staining using an anti-insulin serum, dark peroxidase reaction product stains most of the cells forming the injected cluster, including those of the two communicating territories. Thus, in this case, carboxyfluorescein was directly exchanged between insulin-containing B-cells. (a, b, and c: Bar, 25 gm . x 540) . (d) Three islet cells (asterisks) of this large cluster were successively microinjected with 6- carboxyfluorescein . (e) In the three cases, the dye spread from the injected into some of the adjacent islet cells and delineated, within the cluster, three separate territories of coupled cells. (d and e : Bar, 25 pm. x 330) . At the ultrastructural level, communicating islet cells were identified by comparing the electron microscopic reconstruction of the whole injected cluster with its phase-contrast and fluorescence photographs. In allcommunicating islet cells that could be positively identified, a swelling of the rough endoplasmic reticulum and mitochondria and a decreased electron density of all cellular compartments was observed . As these alterations showed a decreasing gradient from the injected to the adjacent communicating cells and contrasted with the good preservation of the neighboring noncommunicating cells, we used them as an additional marker to localize the coupled cells within each cluster. A-, B- and D-cell types were identified on the basis of the characteristic ultrastructural appearance of their respective secretory granules (14, 15), a procedure that is consistent with the immunocytochemical localization of the corresponding islet hormones (l9; unpublished data for the monolayer cultures).