The development of epidermal feet in preparation for metamorphosis in an insect

Insect epidermal cell surfaces can be seen by scanning electron microscopy after removal of the basal lamina. This let us study surface changes in the 5th larval stage of Calpodes ethlius (Lepidoptera, Hesperiidae) in preparation for metamorphosis at the end of the stadium, in particular changes in the basal cell processes or feet, intercellular lymph spaces, filopodia and hemidesmosomes. The feet develop in three phases, initiation, elongation and contraction. Initial growth begins immediately after ecdysis and continues until commitment to pupation 66 hr later. During this phase the feet are randomly oriented. Elongation and orientation begin after commitment to pupation. Orientation is probably achieved by selective survival and growth of those feet that are axially oriented rather than by reorientation. As the larva shortens to the pupal form late in the stadium, contraction of the feet occurs and the cells become columnar. The feet finally disappear as the cells rearrange themselves into new positions in the pupal epidermis. The lateral margins of the feet are united by adhesions even when their interdigitations are most complex. The adhesions separate an intercellular lymph space from the haemolymph. The lymph space remains small through most of the stadium, but enlarges with the loss of lateral junctions as the feet contract and eventually extends along most of the length of the columnar cells. Filopodia then form and span the gaps between the cells as though they have been induced by the separation and loss of lateral cell to cell contact. Scanning electron microscopy also shows that hemidesmosomes reflect the axial alignment of the cells even before the orientation of the feet. The hemidesmosome plaques are linear structures having a constant width of 0.15 - 0.2 mum and variable length. They arise in short sections and lengthen by the linear addition of more sections with the same width. Late in the stadium they lose their axial alignment and may become branched.     

Apoferritin in the vacuolar system of insect hemocytes

Electron microscopy showed no holoferritin in either the cytosol or the vacuolar system of hemocytes (granulocytes) from normal Calpodes ethlius larvae. This does not mean that ferritin is normally absent from hemocytes, since apoferritin lacks contrast and would not be observed. In vitro iron in glycerol treatment of hemocytes from normal larvae caused holoferritin cores to be visible in the rough endoplasmic reticulum, suggesting that hemocytes from normal larvae contain apoferritin. Hemocytes are therefore like the fat body, and could also be a source of hemolymph ferritin. After loading the hemolymph with iron in vivo, many holoferritin cores were resolvable in the vacuolar system of some hemocytes. Ferritin synthesis can therefore be induced by elevated hemolymph iron levels. Iron loading of epidermis and heart showed similar ferritin cores but more rarely. In all tissues they occurred in the secretory pathway and not in the cytosol.     

Paired cytoskeletal patterns in an epithelium of siamese twin cells.

The epidermis of some insects is a sheet of siamese twin cells which are formed by conserving the midbody between siblings after cell division. We have found that for about 36 h after ecdysis to the 5th stage, the cells of Calpodes caterpillars contain one to five or more actin bundles. The variation in number of bundles occurs in an epithelium that is presumed to be otherwise genetically and developmentally homogeneous. The number of bundles is paired in adjacent cells (P less than 0.005, n = 617). Confocal microscopy shows midbodies between paired but not between unpaired cells. The pairing is reminiscent of the paired nucleolar patterns in these siamese twin cells (Locke, M., H. Leung, Tissue and Cell 17, 573-588 (1985)) or the mirrored patterns of stress fibers in newly divided 3T3 cells (Albrecht-Buehler, G., J. Cell Biol. 72, 595-603 (1977)). The pairing provides further evidence for the operation of transiently heritable factors as determinants for cell pattern.     

Nucleolar necklace formation in response to hemolymph ecdysteroid peaks.

Previous work on the last (fifth) larval stadium of Calpodes showed two phases of elaboration of epidermal nucleoli correlated with RNA synthesis, the first after ecdysis at the beginning of the intermolt and the second near the end of the stadium prior to molting. Both phases followed periods of elevated hemolymph ecdysteroid. The demonstration of four hemolymph ecdysteroid peaks and an improvement in the bismuth-staining procedure for nucleoli has prompted further study of nucleolar changes in relation to hemolymph edcysteroids. We have found that three of the four ecdysteroid peaks (I, II and IV) are followed by nucleolar changes. The exception is the commitment peak (III) for which there is no corresponding nucleolar change. The three nucleolar cycles are similar in their essential features. An intercycle nucleolus consists of one or a few irregularly shaped particles that become more densely stained and condense into a knot at the beginning of each cycle. The knot unfolds into a necklace which beomes beaded as it elongates to a length of about 23 mum. Cells have one or two, rarely more, necklaces presumably depending on their ploidy. At the end of the cycle the necklaces contract, becoming coarser and fragmented before they condense to the intercycle condition of central irregular cores. Whereas nucleolar necklaces are a general response to hemolymph ecdysteroids, mitoses are locally determined and are imposed over other nuclear activities at any time in the third nucleolar cycle.     

The structure of epidermal feet during their development

Epidermal cells in Calpodes and other insects form basal processes or feet that at first extend axially and later shorten at the same time as the larval segment shortens to the pupal shape. The feet grow into spaces at the surfaces of other cells to make a basal interlacing meshwork of cellular extensions that are combined mechanically by their desmosomal attachments to cell bodies above and to the basal lamina below. Microtubules and microfilaments are linked to these junctions by a reticular fibrous matrix. Gap junctions on the feet may couple cells that are several cell bodies removed from one another. The meshwork is also a sieve separating the hemolymph from the spaces between cells to form an intercellular compartment. Entry to the intercellular compartment is through the sieve made by the negatively charged basolateral cell surfaces that can prevent the entry of positively charged molecules such as cationic ferritin. As the cells become columnar, coincident with the metamorphic change in segment shape, the feet shorten and pack more densely together. At this time the basal lamina buckles axially as if responding to contraction of the feet. Segment shape change involves cell rearrangement and relative cell movement, necessitating the transient loss of plasma membrane plaque attachments to the cuticle apically and the loss of junctions laterally. Gap junctions involute in characteristic vacuoles. The metamorphic reduction in cell surface area coincides with the loss of basolateral membrane in smooth tubes and vesicles and the turnover of the apical surface in multivesicular bodies. New apical plasma membrane plaques and new lateral and basal junctions stabilize the cells in their pupal positions.     

Rapid detection of chromosome aneuploidies in uncultured amniocytes by using fluorescence in situ hybridization (FISH).

Herein we report the results of the first major prospective study directly comparing aneuploidy detection by fluorescence in situ hybridization of interphase nuclei with the results obtained by cytogenetic analysis. We constructed probes derived from specific subregions of human chromosomes 21, 18, 13, X, and Y that give a single copy-like signal when used in conjunction with suppression hybridization. A total of 526 independent amniotic fluid samples were analyzed in a blind fashion. All five probes were analyzed on 117 samples, while subsets of these five probes were used on the remaining samples (because of insufficient sample size), for a total of over 900 autosomal hybridization reactions and over 400 sex chromosome hybridization reactions. In this blind series, 21 of 21 abnormal samples were correctly identified. The remaining samples were correctly classified as disomic for these five chromosomes. The combination of chromosome-specific probe sets composed primarily of cosmid contigs and optimized hybridization/detection allowed accurate chromosome enumeration in uncultured human amniotic fluid cells, consistent with the results obtained by traditional cytogenetic analysis.