Microtubule-based peroxisome movement

Stephan Rapp 2 Rainer Saffrich 1 Markus Anton 2 Ursula Jkle 2 Wilhelm Ansorge 1 Karin Gorgas 0 Wilhelm W. Just 2 0 Institut fur Anatomie und Zellbiologie II der Universitat Heidelberg , Heidelberg , Germany 1 EMBL , Heidelberg , Germany 2 Institut fur Biochemie I der Universitat Heidelberg , Im Neuenheimer Feld 328, D-69120 Heidelberg , Germany SUMMARY The association of peroxisomes with cytoskeletal structures was investigated both by electron microscopy and by kinetic analysis of peroxisome movement. The morphological studies indicated distinct interactions of peroxisomes with microtubules and frequently revealed multiple contact sites. The kinetic approach utilised microinjection and import of fluorescein-labeled luciferase in order to mark and track peroxisomes in vivo. Peroxisomal motility was analysed by time-lapse imaging and fluorescence microscopy. According to their movement peroxisomes were classified into two groups. Group 1 peroxisomes comprising the majority of organelles at 37C moved slowly with an average velocity of 0.0240.012 m m/second whereas the movement of group 2 peroxisomes, 10-15% of the total population, was saltatory exhibiting an average velocity of 0.260.17 m m/second with maximal values of more than 2 m m/second. Saltations were completely abolished by the microtubule-depolymerising drug nocodazole and were Since the observation of Goldfischer et al. (1973) who identified the cause of the cerebro-hepato-renal syndrome of Zellweger as a defect in peroxisome biogenesis (for reviews see Schutgens et al., 1986; Roscher and Rolinski, 1992) peroxisome research mainly focused on targeting and import of peroxisomal matrix proteins (for a review see Subramani, 1993), the insertion of proteins into the peroxisomal membrane (Diestelktter and Just, 1993), and the identification of membrane components participating in the biogenesis of peroxisomes (McCammon et al., 1990; Tsukamoto et al., 1991; Hhfeld et al., 1991; Shimozawa et al., 1992; Allen et al., 1994; Tan et al., 1995; Erdmann and Blobel, 1995). Besides the biogenetic aspects on targeting and import of peroxisomal proteins there are other intriguing questions concerning the biogenesis of peroxisomes such as for example the generation and maintenance of size and shape of peroxisomes, the formation of new peroxisomes, the maintenance of their cytoplasmic distribution or the control of their degradation. In higher eukaryotes only liver and kidney contain large spherical peroxisomes of about 0.5 m m in average diameter. In most other organs slightly reduced by about 25% by cytochalasin D which disrupts the actin microfilament system. Double fluorescence labeling of both peroxisomes and microtubules revealed peroxisome saltations linked to distinct microtubule tracks. Cellular depletion of endogenous levels of NTPs as well as the use of 5 -adenylylimidodiphosphate, a nonhydrolysable ATP analog, applied to a permeabilised cell preparation both completely blocked peroxisomal movement. These data suggest an ATPase dependent, microtubule-based mechanism of peroxisome movement. Both the intact and the permeabilised cell system presented in this paper for the first time allow kinetic measurements on peroxisomal motility and thus will be extremely helpful in the biochemical characterisation of the motor proteins involved. peroxisomes are of a more tubular structure resembling the endoplasmic reticulum (Gorgas, 1984, 1985). There is strong biochemical evidence that new peroxisomes are formed by budding and fission from parent organelles (for a review see Lazarow and Fujiki, 1985) although the underlying mechanisms are not clear and thus far could not be proved experimentally. Peroxisomes in most higher eukaryotic cells seem to be distributed throughout the cytoplasm with no preferential location. On the other hand lysosomes, which are the compartment of final peroxisomal degradation, are preferentially found in juxtaposition to the cell nucleus (Matteoni and Kreis, 1987). Analogous to other organelles some of these biogenetic aspects may be intimately related to the functions of the cytoskeleton. For example microtubules have been shown to be involved in the organization of the endoplasmic reticulum (Dabora and Sheetz, 1988; Allan and Vale, 1994), the Golgi apparatus (Rogalski and Singer, 1984; Marks et al., 1994) and the endosomal compartment (Bomsel et al., 1990; Aniento et al., 1993). Cell organelles move along cytoskeletal tracks and are driven by microtubule- and/or actin-based motor enzymes (for reviews see Schliwa, 1984; Walker and Sheetz, 1993; Fig. 1. Light and electron micrographs of tubular epithelial cells of the P3-segment of canine nephron incubated for catalase activity. (Inset in A) Peroxisomes are abundant and clustered in the basolateral cytoplasm. (A) Microtubules are frequently seen in close association with lysosomes (L) and ER-profiles as well as peroxisomes (large arrowheads). (B) Over a rather long distance a microtubule (small black asterisks) is decorated with numerous filamentous side-arms forming cross-bridges to a peroxisome (small arrowheads) and a mitochondrium (large arrowheads). Note the filamentous matrical inclusions (white arrow in A) and the marginal plates (small arrowhead in A and white asterisk in B). Tubular profile of the endoplasmic reticulum (ER), Mit, mitochondrium. Bars: 3 m m (inset in A); 100 nm (A); 50 nm (B). Langford, 1995). These data strongly suggest that such cytoskeletal interactions may also exist for peroxisomes and may play an important role in their biogenesis. Therefore, in the present study, we analysed in detail the peroxisomalcytoskeletal associations at the electron microscope level in appropriate tissues and established an in vivo system allowing the continuous observation of peroxisomes within the living cell. Based on our recent findings that firefly luciferase (FL) upon microinjection is imported into peroxisomes of various mammalian cell lines (Soto et al., 1993) we were able to visualise and track peroxisome movement, and for the first time we report on the spatial and temporal characteristics of intracellular peroxisome movement in vivo. Our data provide evidence for a specific interaction of peroxisomes with cytoskeletal elements, particularly the microtubule system. MATERIALS AND METHODS Light and electron microscopy Perfusion fixation, cytochemical staining, postfixation as well as embedding of the tissue specimens was carried out as described previously (Gorgas, 1984, 1985; Zaar et al., 1984). CHO wild-type cells (ATCC, Rockville, USA), free of mycoplasma contamination, were grown in 10 cm Falcon plastic dishes in a -MEM (Sigma) supplemented with 7.5% fetal bovine serum and penicillin and streptomycin at a concentration of 10,000 U and 10 mg per 100 ml medium, respectively. Cells were detached from the substratum with trypsin/EDTA 24 hours before microinjection and 1 105 cells per 2 ml of medium were transferred to a 3.5 cm Nun (...truncated)