Studies on microtubules in Heliozoa I. : The fine structure ofActinosphaerium nucleofilum (Barrett), with particular reference to the axial rod structure

Yolk proteins are thought to enter certain eggs by a process akin to micropinocytosis but the detailed mechanism has not been previously depicted. In this study the formation of protein yolk was investigated in the mosquito Aedes aegypti L. Ovaries were fixed in phosphate-buffered osmium tetroxide, for electron microscopy, before and at intervals after a meal of blood. The deposition of protein yolk in the oocyte was correlated with a 15-fold increase in 140 mµ pit-like depressions on the oocyte surface. These pits form by invagination of the oocyte cell membrane. They have a 20 mµ bristle coat on their convex cytoplasmic side. They also show a layer of protein on their concave extracellular side which we propose accumulates by selective adsorption from the extraoocyte space. The pits, by pinching off from the cell membrane become bristle-coated vesicles which carry the adsorbed protein into the oocyte. These vesicles lose the coat and then fuse to form small crystalline yolk droplets, which subsequently coalesce to form the large proteid yolk bodies of the mature oocyte. Preliminary radioautographs, and certain morphological features of the fat body, ovary, and midgut, suggest that the midgut is the principal site of yolk protein synthesis in the mosquito.

With a technique of preselecting the mitotic cell in the living state for subsequent electron microscopy, it has been possible to examine the ultrastructure of the various stages of mitosis with greater precision than has been reported previously. The early dissolution of the nuclear envelope has been found to be preceded by a marked undulation of this structure within the nuclear "hof." This undulation appears to be intimately related to the spindle-forming activity of the centriole at this time. Marked pericentriolar osmiophilia and extensive arrays of vesicles are also prominent at this stage, the former continuing into anaphase. Progression of the cell through prophase is accompanied by a disappearance of these vesicles. A complex that first makes its appearance in prophase but becomes most prominent in metaphase is a partially membrane-bounded cluster of dense osmiophilic bodies. These clusters which have a circumferential distribution in the mitotic cell are shown to be derived from multivesicular bodies and are acid phosphatase-positive. The precise selection of cells during the various stages of anaphase has made it possible to follow chronologically the morphological features of the initiation of nuclear membrane reformation. The nuclear membrane appears to be derived from polar aggregates of endoplasmic reticulum, and the process begins less than 2 minutes after the onset of karyokinesis. While formation of the nuclear envelope is initiated on the polar aspects of the chromatin mass, envelope elements appear on the equatorial aspect long before the polar elements fuse. Apparently interfering with this fusion are continuous spindle tubules which traverse the chromatin mass in striking density at characteristic points. Several cortical changes, also most pronounced in anaphase, have been described, as has the kinetochore which is seen to good advantage only in this stage. The Golgi complex has been found to disappear both morphologically and histochemically during mitosis and to reappear rapidly in telophase. Evidence is presented which implicates the continuous spindle tubules in certain phases of chromosome movement.

Dividing nuclei from the giant ameba Pelomyxa carolinensis were fixed in osmium tetroxide solutions buffered with veronal acetate to pH 8.0. If divalent cations (0.002 M calcium, magnesium, or strontium as chlorides) were added to the fixation solution, fibrils that are 14 mµ in diameter and have a dense cortex are observed in the spindle. If the divalent ions were omitted, oriented particles of smaller size are present and fibrils are not obvious. The stages of mitosis were observed and spindle components compared. Fibrils fixed in the presence of calcium ions are not so well defined in early metaphase as later, but otherwise have the same diameter in the late metaphase, anaphase, and early telophase. Fibrils are surrounded by clouds of fine material except in early telophase, when they are formed into tight bundles lying in the cytoplasm unattached to nuclei. Metaphase and anaphase fibrils fixed without calcium ions are less well defined and are not observably different from each other. The observations are consistent with the concept that spindle fibrils are composed of polymerized, oriented protein molecules that are in equilibrium with and bathed in non-oriented molecules of the same protein. Partially formed spindle fibrils and ribosome-like particles were observed in the mixoplasm when the nuclear envelope had only small discontinuities. Remnants of the envelope are visible throughout division and are probably incorporated into the new envelope in the telophase. Ribosome-like particles are numerous in the metaphase and anaphase spindle but are not seen in the telophase nucleus, once the envelope is reestablished, or in the interphase nucleus.