ABSTRACT
Secretory cells store a notable array of preformed molecules of various sizes in their secretory granules. When these cells are activated to degranulate, granule content is secreted through a fusion pore to the extracellular space. Analysis of secretory granule size (or content) distributions in various cells documented a periodic multimodal behavior, as would be the case under homotypic fusion of granules (Gn) with unit granules of quantal size (G1), a mechanism we proposed1 for granule growth. We irst documented quantal size characteristics of mast cell secretory granules,1 whose cross-
sectional areas, measured by digitized planimetry, were recorded in transmission electron micrographs (TEM) of sections of rat peritoneal mast cells. A histogram of equivalent volumes calculated from the measured areas (assuming spherical granules) showed a periodic multimodal distribution in which the modes fell at volumes that were successive integral multiples of the volume at the irst mode. Application of a moving-bin technique to the data conirmed the presence of these periodic modes of discrete steps.2 Biochemical analysis3 and pulse and chase autoradiography,4 as well as immunohistochemical approaches correlated with quantitative electron microscopy,5 assisted to establish the notion of granule quantal size. The quantal model of granule size and content gained acceptance after it was conirmed in different cells by other methodologies, e.g., the patch clamp technique,6 intra-granule content estimation3 and in vitro biochemical studies correlated with granule morphology.2 For instance, follow-up of single granule degranulation of eosinophils by time-resolved patch-clamp capacitance measurements discloses that the plasma membrane increases in discrete steps. The capacitance step size distributions in promyelocytes and myelocytes7-8 conirm that mature large speciic granules are formed by homotypic fusion of unit granules of similar size. Homotypic fusion is facilitated during early stages of differentiation associated with granulogenesis, thus conirming our earlier work on maturation of rat bone marrow and peritoneal eosinophils.9 The morphometric study demonstrated a periodic, multimodal granule volume distribution, in mature and immature eosinophils. Since the basic volume quantum was observed to be equal in both cases, we suggested that turning the vesiculated young eosinophil granule into a mature dense one depends on intragranule processes rather than signiicant volume change. Molecular players coupled with secretion processes are progressively becoming identiied. Although these players appear to be highly conserved in all non-virus-mediated membrane fusion processes, the dynamic mechanisms of granule homotypic fusion and exocytosis remain to be resolved.9-12
One of the major cornerstones for the establishment of granule growth emerged from quantitative follow-up of granule reconstitution (in various cells) following massive secretion. Subsequent to secretion, the earliest identiiable activated cell showed an extensive diminution in cell volume associated with
granule loss (90-95%). Cell volume then increased almost to the original level over a period of a month (mast cells) or a day (pancreas). The size of the Golgi apparatus increased markedly during the granule packaging period and then returned to its original size.13 The reconstitution of mast cells and pancreatic acinar cells after secretion is a prolonged process with several phases, resulting in cells of varying appearance and content. Granule periodicity remained in evidence throughout the re-granulation time, with constant G1 volume but increasing mean granule size, stabilizing at a constant mean value.13-16 Pulse and chase autoradiography supported the conclusion that granule growth is the result of homotypic fusion.4 In addition, we were able to document TEM micrograph of dumb-bell shaped granules, suggesting homotypic fusion.1 On the other hand, VAMP8-knockout mice result in lack of homotypic fusion.17
