Probing the structure and function of the nuclear pore complex.

Nucleocytoplasmic transport is a bi-directional process mediated by the nuclear pore complex (NPC), which results in a segregation of cytoplasmic and nuclear macromolecules within cells. Some progress has been made in understanding the mechanistic basis of this selective transport phenomenon. In particular, cryo-electron microscopy of frozen-hydrated nuclear envelopes coupled with image processing and labeling studies, has provided a glimpse of the transporter at the center of the NPC.     

Interactions and structure of the nuclear pore complex revealed by cryo- electron microscopy

Nuclear pore complexes (NPCs) play a central role in mediating nucleocytoplasmic transport and exchange processes in eukaryotic cells. The arrangement and interactions of NPCs within amphibian nuclear envelopes have been studied using cryo-electron microscopy of unfixed and frozen hydrated specimens. The nuclear lamina in Necturus forms an orthogonal network with crossover distances which vary between 1,600 and 4,000 A and which may be related to the basic filament repeat of lamins. Furthermore, the NPCs are attached randomly within the confines of the lamin network, presumably by their nucleoplasmic rings. Image analysis of edge-on and en face projections of detergent-extracted NPCs has been combined with data on the coaxial thin rings to provide a quantitative evaluation of the triple ring model of NPC architecture proposed previously (Unwin, P. N. T., and R. Milligan. 1982. J. Cell Biol. 93:63-75). Additional details of the complex have been visualized including an intimate association of the inner spoke domains as an inner spoke ring, extensive domains within the spokes and coaxial thin rings, and interestingly, a central channel-like feature. Membrane-associated NPCs and detergent-extracted NPCs both possess peripherally located radial arms resulting in an effective diameter of approximately 1,450-1,500 A. In projection, the radial arms possess approximate mirror symmetry suggesting that they originate from both sides of the assembly. Moreover, membrane-associated NPCs are asymmetric at most radii and right-handed as viewed from the cytoplasm; detergent-extracted NPCs appear to be symmetric and have approximately 822 symmetry. Taken together, the data suggests that the framework of membrane-associated NPCs is perturbed from a symmetrical configuration, either during isolation of nuclei or by interactions with the lamina and the nuclear envelope in vivo. However, detergent extraction of nuclei appears to result in a more symmetrical alignment of components in apposing halves of the assembly.     

Interaction of ouabain with the Na(+) pump in intact epithelial cells

To determine the specificity and efficacy of [(3)H]ouabain binding as a quantitative measure of the Na(+) pump (Na(+), K(+)-ATPase) and as a marker for the localization of pumps involved in transepithelial Na(+)-transport, we analyzed the interaction of [(3)H]ouabain with its receptor in pig kidney epithelial (LLC-PK(1)) cells. When these epithelial cells are depleted of Na(+) and exposed to 2 muM [(3)H]ouabain in a Na(+)-free medium, binding is reduced by 90 percent. When depleted of K(+) and incubated in a K(+)- free medium, the ouabain binding rate is increase compared with that measured at 5 mM. This increase is only demonstable when Na(+) is present. The increased rate could be attributed to the predominance of the Na(+)-stimulated phosphorylated form of the pump, as K(+) is not readily available to stimulate dephosphorylation. However, some binding in the K(+)-free medium is attributable to pump turnover (and therefore, recycling of K(+)), because analysis of K(+)-washout kinetics demonstrated that addition of 2 muM ouabain to K(+)-depleted cells increased the rate of K(+) loss. These results indicate that in intact epithelial cells, unlike isolated membrane preparations, the most favorable condition for supporting ouabain binding occurs when the Na(+), K(+)-ATPase is operating in the Na(+)-pump mode or is phosphorylated in the presence of Na(+). When LLC-PK(1) cells were exposed to ouabain at 4 degrees C, binding was reduced by 97 percent. Upon rewarming, the rate of binding was greater than that obtained on cells kept at a constant 37 degrees C. However, even at this accelerated rate, the time to reach equilibrium was beyond what is required for cells, swollen by exposure to cold, to recover normal volume. Thus, results from studies that have attempted to use ouabain to eliminate the contribution of the conventional Na(+) pump to volume recovery must be reevaluated if the exposure to ouabain was done in the cold or under conditions in which the Na(+) pump is not operating.     

A synthetic strand of cardiac muscle: its passive electrical properties

The passive electrical properties of synthetic strands of cardiac muscle, grown in tissue culture, were studied using two intracellular microelectrodes: one to inject a rectangular pulse of current and the other to record the resultant displacement of membrane potential at various distances from the current source. In all preparations, the potential displacement, instead of approaching a steady value as would be expected for a cell with constant electrical properties, increased slowly with time throughout the current step. In such circumstances, the specific electrical constants for the membrane and cytoplasm must not be obtained by applying the usual methods, which are based on the analytical solution of the partial differential equation describing a one-dimensional cell with constant electrical properties. A satisfactory fit of the potential waveforms was, however, obtained with numerical solutions of a modified form of this equation in which the membrane resistance increased linearly with time. Best fits of the waveforms from 12 preparations gave the following values for the membrane resistance times unit length, membrane capacitance per unit length, and for the myoplasmic resistance: 1.22 plus or minus 0.13 x 10-5 omegacm, 0.224 plus or minus 0.023 uF with cm-minus 1, and 1.37 plus or minus 0.13 x 10-7 omegacm-minus 1, respectively. The value of membrane capacitance per unit length was close to that obtained from the time constant of the foot of the action potential and was in keeping with the generally satisfactory fit of the recorded waveforms with solutions of the cable equation in which the membrane impedance is that of a single capacitor and resistor in parallel. The area of membrane per unit length and the cross-sectional area of myoplasm at any given length of the preparation were determined from light and composite electron micrographs, and these were used to calculate the following values for the specific electrical membrane resistance, membrane capacitance, and the resistivity of the cytoplasm: 20.5 plus or minus 3.0 x 10-3 omegacm-2, l.54 plus or minus 0.24 uFWITHcm-minus 2, and 180 plus or minus 34 omegacm, respectively.     

Dissociation and reconstitution of human ferroxidase II.

The ferroxidase II protein from human serum is large and structurally complex. It possesses protein-bound lipid and copper components which are essential for the maintenance of its catalytic activity. Treatment of ferroxidase II with 8 M urea, 6 M guanidine hydrochloride, or 6 M guanidine hydrochloride and alkylation does not result in the dissociation of the enzyme into subunits. However, treatment with sodium dodecyl sulfate results in the dissociation of ferroxidase II into two nonidentical subunits, designated S-I and S-II. S-I contains little phospholipid, cholesterol, or copper and has a molecular weight of 3.8-3.9 X 10(5). In contrast, S-II contains bound phospholipid, cholesterol, and copper and has a molecular weight of 2.2-2.4 X 10(5). The lipid compositon of S-II is identical with the native enzyme. Sodium dodecyl sulfate-free S-I exhibits no ferroxidase activity. Immediately following removal of sodium dodecyl sulfate, S-II exhibits ferroxidase activity but S-II rapidly loses its activity in the absence of S-I. The separated subunits spontaneously reassociate upon removal of the sodium dodecyl sulfate to yield a fully active enzyme which chemically appears identical with native ferroxidase II. Furthermore, the reconstituted enzyme is stable. Both native and reconstituted ferroxidase II may be stored at 4 degrees C for 6 weeks without any loss in activity. This suggests that S-II, the copper and lipid-containing subunit, is the catalytic subunit and that S-I is essential for the stabilization of the enzymic activity of S-II. These results provide insight into the molecular structure and chemical composition of ferroxidase II and suggest that the complete native structure of ferroxidase II is required for the maintenance of i-s functional integrity.