According to the valve cusp hypoxia hypothesis (VCHH), deep venous thrombosis is caused by sustained non-pulsatile (streamline) venous blood flow. This leads to hypoxemia in the valve pockets; hypoxic injury to the inner (parietalis) endothelium of the cusp leaflets activates the elk-1/egr-1 pathway, leading to leukocyte and platelet swarming at the site of injury and, potentially, blood coagulation. Here, we propose an extension of the VCHH to account for chronic venous insufficiency. First, should the foregoing events not proceed to frank thrombogenesis, the valves may nevertheless be chronically injured and become incompetent. Serial incompetence in lower limb valves may then generate ''passive'' venous hypertension. Second, should ostial valve thrombosis obstruct venous return from muscles via tributaries draining into the femoral vein, as Virchow illustrated, ''active'' venous hypertension may supervene: muscle contraction would force the blood in the vessels behind the blocked ostial valves to re-route. Passive or active venous hypertension opposes return flow, leading to luminal hypoxemia and vein wall distension, which in turn may impair vasa venarum perfusion; the resulting mural endothelial hypoxia would lead to leukocyte invasion of the wall and remodelling of the media. We propose that varicose veins result if gross active hypertension stretches the valve ''rings'', rendering attached valves incompetent caudad to obstructed sites, replacing normal centripetal flow in perforating veins with centrifugal flow and over-distending those vessels. We also discuss how hypoxemia-related venous/capillary wall lesions may lead to accumulation of leukocytes, progressive blockage of capillary blood flow, lipodermosclerosis and skin ulceration.

Most ideas about the pathogenesis of deep venous thrombosis (DVT) are dominated by a 'consensus model' first articulated around 1962. This model invokes Virchow's triad' and attributes thrombo-genesis in veins to some combination of 'hypercoagulability', 'stasis' and 'intimal injury'. This arose as a by-product of studies on the mechanisms of haemostasis and bleeding diatheses that were at best only indirectly relevant to thrombosis, and there are reasons for doubting the causal significance of 'hypercoagulability' and 'stasis' in the aetiology of DVT. Proponents of the consensus model make little reference to a substantial literature, mostly historical, that: (a) emphasizes the significance of the venous valve pockets (VVP) and blood rheology in DVT pathogenesis;,and (b) describes morphological features specific to venous thrombi that a valid aetiological model must explain. This literature provides the basis for an alternative hypothesis of DVT aetiology, published some 30 years ago, which has been experimentally corroborated and is compatible with recent cell and molecular biological studies of the venous endothelium. We review this alternative hypothesis, considering its potential value for future research on DVT and embolism, and its significance for clinical practice.

For many years, it has been believed that diffusion is the principle motive force for distributing molecules within the cell. Yet, our current information about the cell makes this improbable. Furthermore, the argument that limitations responsible for the relative constancy of cell size--which seldom varies by more than a factor of 2, whereas organisms can vary in mass by up to 10(24)--are based on the limits of diffusion is questionable. This essay seeks to develop an alternative explanation based on transport of molecules along structural elements in the cytoplasm and nucleus. This mechanism can better account for cell size constancy, in light of modern biological knowledge of the complex microstructure of the cell, than simple diffusion.

The possibility that the disposition of subsets of proteins within the cell can retain memory traces and may act therefore in a computational role has been advanced and more recently refined by Bray (Nature (1995) 376, 307–312). The proposition is not without its merits but inevitably has a number of associated difficulties, some of which are discussed in this article. These relate to the nature of the computational units envisaged (analog vs digital), their limitations in the number of stable patterns they can accommodate, the reliance on diffusion at the molecular level as the ‘governing principle’, and the complication of the turnover of proteins through degradative mechanisms. These issues suggest that certain modifications of the original model are required.

Much is known about the mechanism by which mRNAs cross the nuclear envelope (the translocation stage of nucleocytoplasmic transport), but far less is known about the preceding (intranuclear migration/release) and succeeding (cytoplasmic binding) stages. Therefore, existing information suffices for articulating detailed kinetic models of translocation, but not models for the overall mRNA transport process. In this paper, we show that simple kinetic models of translocation can (i) accommodate date about nucleocytoplasmic distributions of endogenous transcripts; (ii) predict the overall effects on these distributions of effectors such as insulin and epidermal growth factor; (iii) throw some light on the machanism(s) of action of the HIV-1 protein Rev and produce experimentally testable predictions about this mechanism; and (iv) account for the action of influenza virus NSI protein. However, the simplest forms of translocation models apparently fail to account for some properties of viral regulators such as HIV Rev and adenovirus E1B-E4 complex. To elucidate these topics, less narrowly focused models of mRNA transport are required, describing intranuclear binding/release as well as translocation. On the basis of our examination of translocation models, we suggest some criteria that the requisite broadly based models must satisfy.

Intracellular movement of any solute or particle accords with one of two general schemes: either it takes place predominantly in the solution phase or it occurs by dynamic interactions with solid-state structures. If nucleocytoplasmic exchanges of macromolecules and complexes are predominantly solution-phase processes, i.e., if the former ("diffusionist") perspective applies, then the only significant structures in nucleocytoplasmic transport are the pore complexes. However, if such exchanges accord with the latter ("solid-state") perspective, then the roles of the nucleoskeleton and cytoskeleton in nucleocytoplasmic transport are potentially, at least, as important as that of the pore complexes. The role of the nucleoskeleton in mRNA transport is more difficult to evaluate than that of the cytoskeleton because it is less well characterized, and current evidence does not exclude either perspective. However, the balance of evidence favors a solid-state scheme. It is argued that ribosomal subunits are also more likely to migrate by a solid-state rather than a diffusionist mechanism, though the opposite is true of proteins and tRNAs. Moreover, recent data on the effects of viral proteins on intranuclear RNA processing and migration accord with the solid-state perspective. In view of this balance of evidence, three possible solid-state mechanisms for nucleocytoplasmic mRNA transport are described and evaluated. The explanatory advantage of solid-state models is contrasted with the heuristic advantage of diffusion theory, but it is argued that diffusion theory itself, even aided by modern computational techniques and numerical and graphical approaches, cannot account for data describing the movements of materials within the cell. Therefore, the mechanisms envisaged in a diffusionist perspective cannot be confined to diffusion alone, but must include other processes such as bulk fluid flow.

It is argued that Brownian motion makes a less significant contribution to the movements of molecules and particles inside cells than is commonly believed, and that the numbers of similar molecules and particles within any near-homogeneous subcompartment of the cell internum are insufficient to justify the statistical assumptions implicit in the derivation of the diffusion equation. For these reasons, it is contended that, contrary to accepted opinion, diffusion theory cannot provide an explanation for intracellular transport at the molecular level. Although attempts have been made to adapt diffusion theory to complex media, the conclusion is that none satisfactorily overcomes the problem of applying the theory to cell biology. However, the heuristic influence of the theory on cellular biophysics and physiology is noted, and possible alternative frameworks for interpreting the valuable experimental data obtained from such studies are outlined.

This paper explores the idea that mRNAs are transported between their transcription and processing sites in the nucleus, and their translation and degradation sites in the cytoplasm, by a 'solid-state' process. The underlying assumption is that negligible quantities of mRNA and of mRNA precursors are in solution in vivo. Therefore, mRNA transport cannot be considered as movement in the aqueous phase of the cell. The main lines of experimental evidence supporting this 'solid-state' concept are summarized and related controversies are outlined. Three possible models for a solid-state transport mechanism are discussed: a direct transfer model, with receptors organized analogously to the components of a multienzyme complex; a motor-driven model, analogous to synaptic vesicle transport in axons; and an assembly-driven model which assumes net movement along a fibril resulting from differential activities at the poles. Qualitative evaluation indicates that each of these models has characteristic advantages and disadvantages. The possibility that other nucleocytoplasmic transport processes might operate by solid-state mechanisms is briefly discussed.

mRNA translocation across the nuclear envelope and the appropriate signal-receptor interactions have been studied using resealed rat liver nuclear envelope ghosts (RNEG). We compared export kinetics of nonadenylated (tRNAs, histone-2 poly(A)− mRNA), and adenylated RNAs (poly(A)+ tRNAs, synthetic histone-2 poly(A)+mRNA, albumin mRNA, β-globin poly(A) +mRNA and a total poly(A) + mRNA extract from rat liver cells). ATP-dependent export of mRNAs and of total poly(A)(+) RNA was prevented by inhibitors of a nuclear envelope NTPase. All adenylated RNA species competed with each other for export, but nonadenylated RNAs did not. This indicates the existence of different translocation mechanisms for different RNA species with their appropriate nuclear envelope associated RNA receptors involved in export. The attachment of a poly(A)250 sequence at the 3′-end of tRNA or histone messenger masks the intrinsic RNA export signal of nonadenylated RNAs and results in efflux comparable to that of β-globin poly(A)+ mRNA. The attachment on oligo(A)5 does not have any comparable effect of nonadenylated RNA translocation. Export of all polyadenylated RNAs from RNEGs is blocked by a monoclonal antibody, which is directed against an intranuclear envelope poly(A) binding protein. The results suggest that the pore complexes do not select RNAs for export to the cytoplasm and are therefore not responsible for nuclear restriction of mRNA precursors.

Transport of mRNA from nucleus to cytoplasm is an ATP-dependent process which occurs strictly vectorially. Because the mRNA is structurally bound during transport, mRNA transport is a "solid-state" process consisting of i) mRNA release from the nuclear matrix, ii) mRNA translocation through the nuclear pore, and iii) cytoskeletal binding. We identified and purified the following components involved in the translocation step: i) the nuclear envelope (NE) nucleoside triphosphatase (NTPase) which is stimulated by the 3'poly(A) tail of mRNA, ii) the poly(A)-recognizing mRNA carrier, iii) the NE protein kinase, and iv) the NE phosphatase. In addition, we found that an RNA helicase activity is present in NE, which also may be involved in RNA transport. Our results show that, besides poly(A), also double-stranded RNA structures may modulate RNA export. The amount of mRNA released from nuclei markedly decreases with age. Evidence is presented that this age-dependent change is caused by an impairment of polyadenylation of mRNA, hnRNA processing, release of mRNA from nuclear matrix, and translocations of mRNA from nuclear to cytoplasmic compartment (decrease in activities of NE NTPase, protein kinase, and phosphatase; decrease in poly(A)-binding affinity of mRNA carrier).

Transport of mRNA from nucleus to cytoplasm is an ATP-dependent process which occurs strictly vectorially. Because the mRNA is structurally bound during transport, mRNA transport is a "solid-state" process consisting of i) mRNA release from the nuclear matrix, ii) mRNA translocation through the nuclear pore, and iii) cytoskeletal binding. We identified and purified the following components involved in the translocation step: i) the nuclear envelope (NE) nucleoside triphosphatase (NTPase) which is stimulated by the 3'poly(A) tail of mRNA, ii) the poly(A)-recognizing mRNA carrier, iii) the NE protein kinase, and iv) the NE phosphatase. In addition, we found that an RNA helicase activity is present in NE, which also may be involved in RNA transport. Our results show that, besides poly(A), also double-stranded RNA structures may modulate RNA export. The amount of mRNA released from nuclei markedly decreases with age. Evidence is presented that this age-dependent change is caused by an impairment of polyadenylation of mRNA, hnRNA processing, release of mRNA from nuclear matrix, and translocation of mRNA from nuclear to cytoplasmic compartment (decrease in activities of NE NTPase, protein kinase, and phosphatase; decrease in poly(A)-binding affinity of mRNA carrier).

We have purified a 110 kDa poly(A) binding protein (P110) from rat liver which is thought to be involved in mRNA translocation through the nuclear pores and have demonstrated its localisation in the nuclear envelope using polyclonal antibodies and confocal laser scanning microscopy. Although P110 was prepared from highly purified nuclear envelopes, the polyclonal antibodies raised against them bind to nucleo- and cytoplasmic structures to a minor extent, but not to nucleolar structures. P110 decays spontaneously into several fragments which are also recognized by the polyclonal antibodies. The 110 kDa polypeptide and its fragments were phosphorylated by a nuclear envelope kinase and this phosphorylation was inhibited by a monoclonal antibody against protein kinase C and by a specific protein kinase C inhibitor obtained from bovine brain. Scatchard analysis was used to determine the influence of protein kinase C activators and inhibitors on nuclear envelope protein phosphorylation and RNA binding. The data indicate a close association between the RNA translocation machinery (the 110 kDa protein) and protein kinase C within the nuclear envelope. We suggest that the fragmentation of P110 is triggered before or during mRNA export and is not due to nonspecific proteolysis.

Establishment and maintenance of correct partitioning of proteins and RNA molecules between nucleus and cytoplasm in a sine qua non of the viability of eukaryotic cells. Cytoskeletal elements play several roles in such partitioning: controlling the diffusion of proteins within the main cell compartments; presenting transportable macromolecular ligands to receptor sites within the pore complexes; maintaining the structure and dynamics of the pore complexes themselves. The solid-state transport machinery which moves mRNA molecules between particular sites in nucleus and cytoplasm is dependent on actin and other fibrils, and the migration of other major RNA types might show similar dependence. These various aspects of macromolecule partitioning illustrate one way in which the cytoskeleton is fundamental to the eukaryotic state.

We have used a photoreactive cross-linking reagent, poly(A/8-N3-A) (a poly(A) of average molecular mass of 100 kDa in which 5-10% of the A residues are replaced by 8-N3-A), to label poly(A) binding proteins of rat liver nuclear envelopes. This reagent was prepared by polymerizing a mixture of ADP and 8-N3-ADP with polynucleotide phosphorylase. The purified poly(A) was labeled in the 5'-position with a 32P group. In nuclear envelopes prepared by a low salt DNase I procedure, the poly(A/8-N3-A) labeled a protein-nucleic acid complex of approximately 270 kDa, which on degradation with RNase U2 or NaOH at pH 10 yielded two polypeptides of approximately 50 and 30 kDa. These photoreaction products were markedly decreased when resealed nuclear envelopes or non-nuclear envelope proteins were irradiated in the presence of poly(A/8-N3-A). The affinity labeling was intensified when resealed vesicles were made leaky by freezing or ultrasonication, suggesting that the poly(A) binding proteins are accessible from the nucleoplasmic but not the cytoplasmic face of the envelope. Moreover binding was specific for poly(A). Alternative reagents, random poly(A/8-N3-A,C,G,U) of about 100 kDa and poly(dA) (molecular mass between 350 and 515 kDa), showed a very low affinity for poly(A) recognition proteins in the low salt DNase I-treated nuclear envelopes; the 270-kDa band was labeled only weakly. The binding site was not protected by poly(A,C,G,U), weakly by poly(dA), and distinctly by poly(A).

Two polysome-associated proteins with particular affinities for poly(A) have been purified from rat liver. These proteins stimulate the efflux of mRNA from isolated nuclei in conditions under which such efflux closely stimulates mRNA transport in vivo, and they are therefore considered as mRNA-transport-stimulatory proteins. Their interaction with the mRNA-translocation system in isolated nuclear envelopes has been studied. The results are generally consistent with the most recently proposed kinetic model of mRNA translocation. One protein, P58, has not been described previously. It inhibits the protein kinase that down-regulates the NTPase, it enhances the NTPase activity in both the presence and the absence of poly(A) and it seems to increase poly(A) binding in unphosphorylated, but not in phosphorylated, envelopes. The other protein, P31, which probably corresponds to the 35,000-Mr factor described by Webb and his colleagues, enhances the binding of poly(A) to the mRNA-binding site in the envelope, thus stimulating the phosphoprotein phosphatase and, in consequence, the NTPase. The possible physiological significance of these two proteins is discussed.

The effects of (a) oxidative cross-linking of protein thiol groups and (b) the presence or absence of Ca2+ ions on rat liver nuclear-envelope isolation were studied. Two envelope-isolation procedures were compared: a well characterized low-ionic-strength method and a recently developed high-ionic-strength method. The latter method seems preferable to the former in respect of lower intranuclear contamination of the envelopes, suppression of endogenous serine proteinase, and maintenance of high specific activities of envelope-associated enzymes. In both procedures, however, the presence of Ca2+ gave rise to a rapid, apparently irreversible, contamination of the envelopes by intranuclear material. This effect was half-maximal at 20 microM-Ca2+. In addition, the envelopes became contaminated with intranuclear material by a Ca2+-independent mechanism, apparently resulting from N-ethylmaleimide-sensitive intermolecular disulphide-bond formation. This oxidative process seemed to have two major kinetic components (half-life, t1/2, approx. 2 min and 10 min). In view of these findings, it is recommended that (i) for most purposes, nuclear envelopes be isolated by the newly developed high-ionic-strength procedure, (ii) irrespective of the method used, Ca2+-chelators be included in all the buffers, (iii) thiol-group oxidation be prevented or reversed during the procedure.