| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biological Sciences, Dartmouth College, Hanover, NH 03755
1 Rostock University, Germany
Ca2+ is a key regulator of cytokinesis, the process by which the contractile ring constricts the cell to form the cleavage furrow. BAPTA (1,2-bis (aminophenoxy)-ethane-N,N,N',N'-tetraacetic acid) has been used to investigate the role of Ca2+ in this process. Several studies have shown that buffering cytosolic Ca2+ with BAPTA blocks cytokinesis in a variety of cultured cells and fertilized eggs (1,2). These results support a model of cytokinesis in which a local rise in free Ca2+ is part of the signaling pathway that regulates assembly of the cytokinetic ring of actin filaments. In addition, Ca2+ is required for Ca2+-calmodulin (CaM)-stimulated phosphorylation of the regulatory light chain of myosin-II by myosin light chain kinase (MLCK). Therefore, Ca2+ serves both as a signal for cytokinesis and as a regulator of myosin-II activity.
We studied cytokinesis in vitro by reconstituting pseudo-contractile rings in M-phase extracts made from clam oocytes. In these extracts, actin filaments spontaneously organized into a cooperative, 3-D network of interconnected filaments. We refer to this self-organized network of actin filaments as a pseudo-contractile ring because it shares the following two fundamental properties with the contractile ring: (i) it exhibits myosin-II-mediated, anti-parallel sliding of actin filaments; and (ii) it assembles during the M-phase of the cell cycle. Fortuitously, actin filaments within the pseudo-contractile ring co-aligned to form bundles (10–20 parallel filaments) that were thick enough to be visualized by Allen Video Enhanced Contrast (AVEC) - Differential Interference Contrast (DIC) microscopy. Therefore, sliding of actin filaments could be monitored and studied in networks reconstituted in vitro. The term "contraction" refers to the myosin-II-mediated sliding of actin filaments in the network; the network itself does not shorten.
In a previous study, we used a function-blocking myosin-II-specific antibody to establish that the sliding of actin bundles observed in these pseudo-contractile rings is mediated by myosin-II (3). Using the same antibody, we showed that vesicle transport on actin filaments observed in these extracts is also mediated by myosin-II. In this report, we determined the effects of Ca2+ buffering by BAPTA on actin filament sliding and vesicle transport in M-phase extracts of clam oocytes.
Extracts were prepared from mature oocytes obtained from gravid female clams. The cytoplasmic extracts were diluted 2-fold to a final protein concentration of about 15 mg/ml and clarified to remove large membrane aggregates. Nocodazole (30 µM) was added to the extracts to block microtubule assembly, an ATP regenerating system was added to maintain ATP levels, and the preparation was incubated for 45 min at 18 °C. Rhodamine-phalloidin (0.5 µM) was added to stain the actin filaments that assembled during the 45-min incubation. The myosin-II motor activity was monitored by AVEC-DIC and fluorescence microscopy.
In control extracts, actin filaments organized spontaneously into a 3-D network, or pseudo-contractile ring (Fig. 1C; control). The actin filaments in the network moved by sliding relative to each other. An average of 2 sliding actin bundles/field/min was observed (Fig 1B; control). The advancing tips of the sliding filaments were tracked at an average speed of 0.2 µm/s and for distances greater than 25 µm. In addition, the movement of vesicles on actin filaments was observed. In a given video field that measured 25 µm2, the motile activity on actin filaments was 46 ± 2 vesicles/field/min (v/f/m). The average speed of vesicle movement was 1.0 µm/s (Fig. 1A; control). The difference in the speed of vesicle movement and filament sliding provided evidence that vesicle movement was not due to the passive attachment of vesicles to sliding filaments. In addition, vesicle movement was observed on stationary actin filaments that were tightly bound to the glass coverslip.
|
In summary, when the calcium concentration in the extract was reduced by the addition of 10 mM BAPTA, myosin-II-mediated vesicle transport increased 1.7-fold. Therefore the activation of MLCK by Ca2+-CaM was not attenuated by the addition of this calcium chelator. However, actin filament bundling was inhibited, suggesting that actin filament bundling activity involves a calcium-sensitive step.
The stimulation of motile activity in these extracts was due, most likely, to the demonstrated effect of BAPTA on membrane network formation. BAPTA (10 mM) has been shown by Drier and Rapoport (6) to stimulate the formation of ER-like networks in extracts of Xenopus oocytes. A similar effect was observed in the extracts that we made from clam oocytes. The vesicles in these extracts underwent fusion events, forming large clusters of vesicles connected to each other by thin membrane tubules (Fig. 1C, arrows in DIC micrograph). The clustering of vesicles into large complexes increased the number of vesicles observed moving per unit time. Our results are consistent with the observation that high calcium levels inhibit membrane fusion and the formation of ER-like networks.
These data suggest that calcium fluctuations of about 1 µM may be important for ER-like network formation. BAPTA, a calcium chelator with a binding constant in this range, affects membrane fusion while chelators with either a lower or higher binding constant do not (6). Therefore, the lack of a strong inhibitory effect of BAPTA on MLCK activity suggests that the binding constant for calmodulin lies outside the range of calcium fluctuations reported for BAPTA (1).
The generation of sliding forces between anti-parallel actin filaments is a well-established activity of bipolar filaments of myosin-II. Therefore, the observation that actin filaments self-organized and moved in an anti-parallel fashion in these extracts fits current models of the contractile ring. These data support models in which the bundling of actin filaments involves a calcium-sensitive step. Calcium signaling through Rho G-protein pathways, which modulate the actin cytoskeleton, is the most likely mechanism by which cytosolic Ca2+ controls actin bundle formation (7). The involvement of actin bundling activity in the formation of contractile rings has yet to be established, although these data raise such a possibility.
Supported by NSF grant IBN-0131470 and MBL Shifman award to RFT.
Literature Cited
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
