The duplication of entire genomes has long been recognized as having great potential for evolutionary novelties, but the mechanisms underlying their resolution through gene loss are poorly understood. Here we show that in the unicellular eukaryote Paramecium tetraurelia, a ciliate, most of the nearly 40,000 genes arose through at least three successive whole-genome duplications. Phylogenetic analysis indicates that the most recent duplication coincides with an explosion of speciation events that gave rise to the P. aurelia complex of 15 sibling species. We observed that gene loss occurs over a long timescale, not as an initial massive event. Genes from the same metabolic pathway or protein complex have common patterns of gene loss, and highly expressed genes are over-retained after all duplications. The conclusion of this analysis is that many genes are maintained after whole-genome duplication not because of functional innovation but because of gene dosage constraints.Ciliates are unique among unicellular organisms in that they separate germline and somatic functions 1 . Each cell harbours two kinds of nucleus, namely silent diploid micronuclei and highly polyploid macronuclei. The latter are unusual in that they contain an extensively rearranged genome streamlined for expression and divide by a non-mitotic process. Only micronuclei undergo meiosis to perpetuate genetic information; the macronuclei are lost at each sexual generation and develop anew from the micronuclear lineage.In Paramecium the exact number of micronuclear chromosomes (more than 50) and the structures of their centromeres and telomeres remain unknown. During macronuclear development, these chromosomes are amplified to about 800 copies and undergo two types of DNA elimination event. Tens of thousand of short, unique copy elements (internal eliminated sequences) are removed by a precise mechanism that leads to the reconstitution of functional genes 2 .Transposable elements and other repeated sequences are removed by an imprecise mechanism leading either to chromosome fragmentation and de novo telomere addition or to variable internal deletions 3 . These rearrangements occur after a few rounds of endoreplication, leading to some heterogeneity in the sequences abutting the imprecisely eliminated regions 3 . The sizes of the resulting, acentric macronuclear chromosomes range from 50-1,000 kilobases (kb) as measured by pulsed-field gel electrophoresis. Because the sexual process of autogamy results in an entirely homozygous genotype 4 , the macronuclear DNA that was sequenced was genetically homogeneous.The Paramecium genome sequence The Paramecium macronuclear genome sequence was established with the use of a whole-genome shotgun and assembly strategy. Paired-end sequencing of plasmid and bacterial artificial chromosome (BAC) clones provided a coverage of 13 genome equivalents (Supplementary Table S1). We assembled the sequence reads with Arachne 5 in 1,907 contigs connected in 697 scaffolds of size greater than 2 kb, giving a total coverage of 72...

Using confocal laser scanning and double immunogold electron microscopy, we demonstrate that reggie-1 and -2 are colocalized in Յ0.1-m plasma membrane microdomains of neurons and astrocytes. In astrocytes, reggie-1 and -2 do not occur in caveolae but clearly outside these structures. Microscopy and coimmunoprecipitation show that reggie-1 and -2 are associated with fyn kinase and with the glycosylphosphatidyl inositol-anchored proteins Thy-1 and F3 that, when activated by antibody cross-linking, selectively copatch with reggie. Jurkat cells, after crosslinking of Thy-1 or GM1 (with the use of cholera toxin), exhibit substantial colocalization of reggie-1 and -2 with Thy-1, GM1, the T-cell receptor complex and fyn. This, and the accumulation of reggie proteins in detergent-resistant membrane fractions containing F3, Thy-1, and fyn imparts to reggie-1 and -2 properties of raft-associated proteins. It also suggests that reggie-1 and -2 participate in the formation of signal transduction centers. In addition, we find reggie-1 and -2 in endolysosomes. In Jurkat cells, reggie-1 and -2 together with fyn and Thy-1 increase in endolysosomes concurrent with a decrease at the plasma membrane. Thus, reggie-1 and -2 define raft-related microdomain signaling centers in neurons and T cells, and the protein complex involved in signaling becomes subject to degradation.

Neurons are believed to possess plasmalemmal microdomains and proteins analogous to the caveolae and caveolin of nonneuronal cells. Caveolae are plasmalemmal invaginations where activated glycosyl‐phosphatidylinositol (GPI)‐anchored proteins preferentially assemble and where transmembrane signaling may occur. Molecular cloning of rat reggie‐1 and ‐2 (80% identical to goldfish reggie proteins) shows that reggie‐2 is practically identical to mouse flotillin‐1. Flotillin‐1 and epidermal surface antigen (ESA) (flotillin‐2) are suggested to represent possible membrane proteins in caveolae. Rat reggie‐1 is 99% homologous to ESA in overlapping sequences but has a 49‐amino‐acid N‐terminus not present in ESA. Antibodies (ABs) which recognize reggie‐1 or ‐2 reveal that both proteins cluster at the plasmamembrane and occur in micropatches in neurons [dorsal root ganglia (DRGs), retinal ganglion, and PC‐12 cells] and in nonneuronal cells. In neurons, reggie micropatches occur along the axon and in lamellipodia and filopodia of growth cones, but they do not occur in caveolae. By quantitative electronmicroscopic analysis we demonstrate the absence of caveolae in (anti‐caveolin negative) neurons and show anti‐reggie‐1 immunogold‐labeled clusters at the plasmamembrane of DRGs. When ABs against the GPI‐anchored cell adhesion molecules (CAMs) F3 and Thy‐1 are applied to live DRGs, the GPI‐linked CAMs sequester into micropatches. Double immunofluorescence shows a colocalization of the CAMs with micropatches of anti‐reggie antibodies. Thus, reggie‐1 and reggie‐2 identify sites where activated GPI‐linked CAMs preferentially accumulate and which may represent noncaveolar micropatches (domains). © 1998 John Wiley & Sons, Inc. J Neurobiol 37: 502–523, 1998

The cellular prion protein (PrP c ) resides in lipid rafts, yet the type of raft and the physiological function of PrP c are unclear. We show here that cross-linking of PrP c with specific antibodies leads to 1) PrP c capping in Jurkat and human peripheral blood T cells; 2) to cocapping with the intracellular lipid raft proteins reggie-1 and reggie-2; 3) to signal transduction as seen by MAP kinase phosphorylation and an elevation of the intracellular Ca 2+ concentration; 4) to the recruitment of Thy-1, TCR/CD3, fyn, lck and LAT into the cap along with local tyrosine phosphorylation and F-actin polymerization, and later, internalization of PrP c together with the reggies into limp-2 positive lysosomes. Thus, PrP c association with reggie rafts triggers distinct transmembrane signal transduction events in T cells that promote the focal concentration of PrP c itself by guiding activated PrP c into preformed reggie caps and then to the recruitment of important interacting signaling molecules. ) is a glycosylated glycosylphosphatidyl inositol (GPI-) anchored protein that is mostly expressed on the surface of neurons and immune cells (1−4). PrP c has gained considerable attention due to the conversion of α helix to β sheet structures leading to the protease-resistant conformer designated PrP scrapie (PrP sc ) and the spreading of prion disease (1,4,5). The physiological function of PrP c is still under debate: PrP c has been implicated in cell adhesion, differentiation, copper binding (6), neuroprotection against oxidative stress (7,8), apoptosis (9), and transmembrane signaling via a lipid raft-based mechanism (10).Clearly, PrP c resides in plasma membrane lipid rafts/microdomains (2,11,12). Lipid rafts are discussed as platforms for proteins involved in signal transduction, allowing for example GPIanchored proteins to signal across the plasma membrane (13,14 by natural ligands or antibodies (Abs) leads to so-called clustered rafts, ~100−200 nm in size (15, 16) that can be visualized at the light microscopic (LM) level. In fact, an activation of the nonreceptor Src kinase fyn was reported to occur in neurites of a neuroectodermal cell line in a caveolin-1-dependent manner using AB-induced PrP c cross-linking (10).Conflicting views exist, however, concerning the association of PrP c with caveolin-1 and caveolae as opposed to its association with noncaveolar lipid rafts (16), particularly as neurons and lymphocytes lack caveolin-1 and caveolae (14,17,18).The existence of noncaveolar lipid raft microdomains is clearly revealed by the pattern of the two proteins reggie-1 and reggie-2 (18,19,20), also known as flotillin-2 and flotillin-1 (22). In lymphocytes, the reggie proteins exhibit a strikingly polarized expression known as "capping" (20, 23, 24). AB-mediated sequestration of GPI-anchored proteins such as Thy-1 results in Thy-1 capping and cocapping with the reggies (20) and seems to involve transmission of signals into the cell (13, 25, 26, 27, reviewed in 14). Signaling leading to full T cell activati...

We have analyzed ultrathin sections from isolated bovine chromaffin cells grown on plastic support, after fast freezing, by quantitative electron microscopy. We determined the size and intracellular distribution of dense core vesicles (DVs or chromaffin granules) and of clear vesicles (CVs). The average diameter of DVs is 356 nm, and that of CVs varies between 35–195 nm (average 90 nm). DVs appear randomly packed inside cells. When the distance of the center of DVs to the cell membrane (CM) is analyzed, DV density is found to decrease as the CM is approached. According to Monte Carlo simulations performed on the basis of the measured size distribution of DVs, this decay can be assigned to a “wall effect.” Any cortical barrier, regardless of its function, seems to not impose a restriction to a random cortical DV packing pattern. The number of DVs closely approaching the CM (docked DVs) is estimated to be between 364 and 629 (average 496), i.e., 0.45 to 0.78 DVs/μm2 CM. Deprivation of Ca2+, priming by increasing [Ca2+]i, or depolarization by high [K+]e for 10 s (the effect of which was controlled electrophysiologically and predicted to change the number of readily releasable granules [RRGs]) does not significantly change the number of peripheral DVs. The reason may be that (a) structural docking implies only in part functional docking (capability of immediate release), and (b) exocytosis is rapidly followed by endocytosis and replenishment of the pool of docked DVs. Whereas the potential contribution of DVs to CM area increase by immediate release can be estimated at 19–33%, that of CVs is expected to be in the range of 5.6–8.0%.