Hypoxia-induced des-gamma-carboxy prothrombin production in hepatocellular carcinoma.

Des-gamma-carboxy prothrombin (DCP) is an established HCC tumor marker, but the precise mechanism of its production is still unclear. Recently, we demonstrated that cytoskeletal changes during epithelial-to-fibroblastoid conversion (EFC) or epithelial mesenchymal transition (EMT) induced by chemicals plays a critical mechanistic role in DCP production via impairment in vitamin K uptake. Our proposed mechanism of DCP production is consistent with substantial clinical evidence. Supplementary vitamin K2 analogues reduced serum DCP levels in hepatocellular carcinoma (HCC) patients. HCC patients with high serum DCP are associated with vascular invasion, metastasis and tumor recurrence. On the other hand, hypoxia has been reported to induce EMT or cytoskeletal changes. Therefore, we examined whether hypoxia induced DCP production during EFC or EMT in HCC cells. Indeed, hypoxic stimulation induced hepatoma cell lines (HepG2 or PLC/PRF/5 cells) to undergo EFC or EMT and these cells produced DCP. Immunofluorescence study demonstrated that hypoxic stimulation impaired labeled low-density lipoprotein uptake, which was a surrogate for vitamin K uptake. In addition, fine filamentous actin network, which has crucial role for clathrin-mediated endocytosis of vitamin K, was disrupted in DCP producing cells by hypoxic stimulation. Thus, hypoxic stimulation induced HCC cells to produce DCP in the same mechanism as chemicals. Furthermore, immunohistochemical study using surgically resected HCC samples showed that a positive staining of nuclear hypoxia inducible factor (HIF)-1alpha was more frequently observed in HCC cells with stronger staining intensity of DCP. Importantly, clinical observations that DCP as an HCC tumor marker was more useful in larger tumors, which is likely to be exposed with hypoxia during tumor development, support our results.

Tissue distribution of K-vitamers under different nutritional regimens in the rat.

Two forms of vitamin K [phylloquinone (K1) and menaquinone-4 (MK-4)] were added to vitamin K-deficient rat food in varying amounts. These diets were given as the sole source of nutrition to rats for one week. The minimal dietary requirements (MDR) to attain maximal prothrombin synthesis were determined to be 0.6 and 6-10 microg/g of food for K1 and MK-4, respectively. The difference between both vitamers could be explained by the limited hepatic accumulation of MK-4. Next, vitamin K was offered to rats at concentrations ranging between 0.6 and 3000 microg/g of food, and the tissue distribution of vitamin K was investigated after one week of administration. Accumulation of K1 and MK-4 was found in all tissues investigated, but both the absolute tissue concentration and the ratio between K1 and MK-4 were tissue-dependent. Highest values were found in liver and in heart, but since the heart contains no gamma-glutamylcarboxylase, the function of vitamin K in this tissue remains obscure. High tissue concentrations of MK-4 were also found in pancreas and testis after a diet containing K1 exclusively. The data indicate that this conversion is tissue-specific, but neither the reason nor its mechanism are known.

Overexpression of human prothrombin in permanent cell lines using a dominant selection/amplification fusion marker.

Human prothrombin was overexpressed in transformed eukaryotic cells using a dominant bifunctional selection and amplification marker. The marker consists of the murine wild-type dihydrofolate reductase (dhfr) cDNA and the Escherichia coli hygromycin phosphotransferase gene fused in frame. The gene of interest is connected by the encephalomyocarditis virus 5' untranslated region to the fusion marker gene, forming a dicistronic transcription unit. The human prothrombin gene (FII) was used to monitor expression after initial selection for hygromycin B resistance and DHFR activity. In Chinese hamster ovary (CHO) cells, 5-15 mU prothrombin/10(6) cells per 24 h was obtained; in human 293 kidney cells levels of 20-50 mU/ 10(6) cells per 24 h were obtained. Methotrexate-mediated amplification of the foreign gene in CHO cells resulted in a more than 10-fold increase in FII expression, while in the presence of methotrexate, 293 cells expressed 200-250 mU/10(6) cells per 24 h. The use of this fusion marker within a dicistronic transcription unit allowed efficient dominant selection of cell clones and amplification of the gene of interest. Stably transfected cell lines that were able to secrete high levels of processed gamma-carboxylated human prothrombin were thus obtained.

Biosynthesis of prothrombin: intracellular localization of the vitamin K-dependent carboxylase and the sites of gamma-carboxylation.

Prothrombin is a vitamin K-dependent blood coagulation protein that undergoes posttranslational gamma-carboxylation and propeptide cleavage during biosynthesis. The propeptide contains the gamma-carboxylation recognition site that directs gamma-carboxylation. To identify the intracellular sites of carboxylation and propeptide cleavage, we monitored the synthesis of prothrombin in Chinese hamster ovary cells stably transfected with the prothrombin cDNA by immunofluorescent staining. The vitamin K-dependent carboxylase was located in the endoplasmic reticulum and Golgi complex. Antibodies specific to prothrombin processing intermediates were used for immunocytolocalization. Anti-des-gamma-carboxyprothrombin antibodies stained only the endoplasmic reticulum whereas antiproprothrombin antibodies (specific for the propeptide) and antiprothrombin:Mg(II) antibodies (which bind the carboxylated forms of proprothrombin and prothrombin) stained both the endoplasmic reticulum and the Golgi complex. Antiprothrombin:Ca(II)-specific antibodies (which bind only to the carboxylated form of prothrombin lacking the propeptide) stained only the Golgi complex and secretory vesicles, and colocalized with antimannosidase II and anti-p200 in the juxtanuclear Golgi complex. These results indicate that uncarboxylated proprothrombin undergoes complete gamma-carboxylation in the endoplasmic reticulum and that gamma-carboxylation precedes propeptide cleavage during prothrombin biosynthesis.

Construction, properties and specific fluorescent labeling of a bovine prothrombin mutant engineered with a free C-terminal cysteine.

To define the role of phosphatidylserine-induced conformational changes in prothrombin activation during blood coagulation, a recombinant bovine prothrombin was constructed, characterized and shown to have a globally native-like conformation. We introduced a cysteine to replace the penultimate residue (Gly581) of a previously constructed active site mutant, and expressed the double mutant in Chinese hamster ovary cells at the level of 0.6 microgram/ml of cell culture medium. Specific labeling with fluorescein maleimide was accomplished by limited reduction with dithiothreitol to free the engineered cysteine while maintaining the native-like functional properties of the molecule. The average stoichiometry of labeling was 0.84 probe/protein. The location of the probe at the C-terminus was confirmed by proteolysis by native thrombin, by Taipan venom, and by carboxypeptidase Y. Both the double mutant and labeled prothrombin could be activated by snake venoms and the prothrombinase but, as expected, the double mutant meizothrombin did not autolyze as does native meizothrombin. Thus, for the first time, a native-like but specifically labeled prothrombin has been constructed. This molecule will be an essential tool for elucidating the structural role of membranes during prothrombin activation. In addition, the methods described might be usefully applied to labeling of an odd, engineered cysteine in other disulfide bond-containing proteins.

Vitamin K-dependent carboxylase activity, prothrombin mRNA, and prothrombin production in two cultured rat hepatoma cell lines.

The presence of under-gamma-carboxylated forms of plasma prothrombin is a marker for human primary hepatocellular carcinoma. A rat hepatoma cell line (7777) which was previously shown to secrete undercarboxylated prothrombin when grown as a solid tumor has now been grown in monolayer culture. This cell line has a decreased activity of the microsomal vitamin K-dependent carboxylase when compared to a control (H4IIEC3) hepatoma line, does not increase intracellular prothrombin concentrations in response to vitamin K depletion, and secretes undercarboxylated prothrombin even when grown in vitamin K supplemented media. Prothrombin gene expression in the 7777 cell line, as measured by prothrombin mRNA levels, was not altered in the 7777 cell line. This cell line appears to be a model for assessing the cellular alterations responsible for undercarboxylated prothrombin excretion by human hepatocellular tumors.

Nucleotide sequence of prothrombin gene in abnormal prothrombin-producing hepatocellular carcinoma cell lines.

A protein induced by vitamin K absence or antagonist II, PIVKA-II is synthesized in the liver and possesses a structure similar to prothrombin except that ten glutamic acid residues in amino-terminal Gla domain are not completely gamma-carboxylated and are functionally inactive. This protein can be detected in the plasma of patients with hepatocellular carcinoma (HCC) and used as a new tumor marker. To analyze the mechanism of PIVKA-II production in HCC tissue, the prothrombin gene of PIVKA-II-secreting HCC cell lines was sequenced to detect the mutation in the Gla domain and carboxylase recognition site of leader sequence located on exons I and II that may cause the inhibition of carboxylation. Exons I and II and donor and acceptor site of intron I of the prothrombin gene in two HCC cell lines, PLC/PRF/5 and huH-2, were analyzed by polymerase chain reaction (PCR), and the product was sequenced directly. In addition, RNA samples of these cell lines were used for complementary DNA synthesis, followed by PCR and sequencing. The nucleotide sequences of the Gla domain in both HCC cell lines were conserved. One nucleotide change was detected at nt.554 (adenine to guanine), but this did not influence the amino acid sequence. Splicing sites between exons I and II, the leader sequence of the precursor prothrombin, and protease target sites also were conserved as the reported prothrombin gene, and mutations reported for other des-gamma-carboxy coagulation factors were not detected. These results also were confirmed by DNA analysis of seven human fresh-frozen samples (three PIVKA-II-positive HCC samples and four control specimens). The mechanism of PIVKA-II production in HCC is still unclear, but it is not caused by mutation in the prothrombin gene.

The interaction of salicylate and vitamin K in synthesis of Factor II (prothrombin).

The effects of sodium salicylate on synthesis of a vitamin K-dependent plasma coagulation protein, Factor II (prothrombin) was studied using the isolated rat liver perfused for 10 hours in vitro. Cumulative synthesis of Factor II was measured by a standard coagulation assay, by activation with E. carinatus venom, and by rocket immunoelectrophoresis. When sodium salicylate 5 mg or 25 mg was added to the liver perfusate (volume 100 ml) at the outset of the perfusion, cumulative synthesis of both coagulation activity and immunoreactive protein was significantly less than that seen in control perfusions containing no salicylate. The inhibitory effect of salicylate was prevented by pretreatment of rat liver donors with supplemental vitamin K injected 24 hours before sacrifice. Although some interaction between salicylate and vitamin K was apparent from these experiments, the results from vitamin K-deficient rat liver donors were quite different from those containing salicylate. There was no assayable Factor II coagulation activity produced in 10 hours of perfusion of vitamin K-deficient rat livers, but cumulative synthesis of immunoreactive Factor II was quite comparable to that seen in control perfusions.

[Vitamin K deficiency in the newborn]. / Vitamin-K-Mangel bei Neugeborenen.

Newborns have low levels of the vitamin K dependent clotting factors. Early studies were suggestive of vitamin K deficiency. Recently these findings were questioned by studies that failed to detect signs of vitamin K deficiency in the clotting system of newborns using more specific methods, while other studies did find signs of vitamin K deficiency using the same methods. The question was finally solved by direct measurement of vitamin K showing very low levels of the vitamin in the serum of newborns immediately after birth. Whether vitamin K supplementation to the mother reduces the incidence of vitamin K induced changes in the clotting system of newborns remains to be elucidated. In the meantime it seems prudent to administer parenteral vitamin K prophylactically to all newborns immediately after birth.

Comparative pharmacokinetics of coumarin anticoagulants XXXV: Examination of possible pharmacokinetic interaction between (R)-(+)- and (S)-(--)-warfarin in humans.

The elimination kinetics and anticoagulant effect produced by single 1.5-mg/kg doses of (R)-(+(-, (S)-(--)-, and racemic warfarin were determined in 10 healthy men. The results obtained in experiments with the individual enantiomers were used to predict the elimination kinetics and anticoagulant effect of racemic warfarin, assuming that there is no interaction between the two enantiomers. These predictions were compared to experimental results, and no significant differences were observed. This finding suggests that there are no pronounced pharmacokinetic or pharmacodynamic interactions between single large doses of (R)-(+)- and (S)-(--)-warfarin in humans.