Magdalena Carlberg

Mevalonic acid is limiting for N-linked glycosylation and translocation of the insulin-like growth factor-1 receptor to the cell surface. Evidence for a new link between 3-hydroxy-3-methylglutaryl-coenzyme a reductase and cell growth.

Depletion of mevalonic acid (MVA), obtained by inhibition of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase using lovastatin, depressed the biosynthesis of dolichyl-phosphate and the rate of N-linked glycosylation and caused growth arrest in the melanoma cell line SK-MEL-2. The growth arrest was partially prevented by addition of high concentrations of insulin-like growth factor-1 (IGF-1) to the cells, indicating that MVA depletion may inhibit cell growth through decreasing the number of IGF-1 receptors (IGF-1R) at the cell surface. Such a decrease in receptor number might be a result of a lowered translocation of de novo synthesized receptors to the cell membrane which in turn might be a result of a decreased N-linked glycosylation of the receptor proteins. We could also demonstrate that IGF-1R became underglycosylated and that the amount of de novo synthesized IGF-1R proteins at the cell membrane was drastically decreased upon MVA depletion. Analysis of receptor proteins cross-linked with IGF-1, as well as binding assays and immunocytostaining confirmed that the number of functional membrane-bound IGF-1R was substantially reduced. The N-linked glycosylation and the expression of de novo synthesized IGF-1R proteins at the cell surface as well as the number of IGF-1 binding sites were completely restored upon replenishment of MVA. These effects of MVA were efficiently abrogated by the glycosylation inhibitor tunicamycin. The translocation of IGF-1R to the cell membrane was shown to take place just prior to initiation of DNA synthesis in arrested cells stimulated with MVA. Additionally, there was a clear correlation between IGF-1 binding and initiation of DNA synthesis with regard to the MVA dose requirement. It was confirmed that inhibition of HMG-CoA reductase activity and N-linked glycosylation also depressed the expression of functional IGF-1R in other cell types (i.e. hepatoblastoma cells and colon cancer cells). Our data suggest that this mechanism is involved in MVA-regulated cell growth.

Evidence for protein dolichylation

Labeling of human colon carcinoma cells with [3H]dol, followed by extensive delipidation and removal of dol‐P oligosaccharides, showed that dol are bound to cellular proteins with sizes of 5, 10, 27, 75 and >140 kDa. HPLC purification of proteolytic products of [3H]dol‐ and [35S]cys‐labeled proteins revealed a hydrophobic peak containing both dol and cysteine. The dol/cys‐labeled products were clearly separated from GG‐cys, and exhibited a hydrophobicity between that of dol‐P and dol. In another set of experiments delipidated proteins were treated with methyl iodide, which cleaves thioether bonds. After HPLC purification of released dol‐like lipids, these were subjected to mass spectrometry. This demonstrated molecular ions with the same mass as that of dol. Taken together our data provide evidence for the existence of proteins covalently modified with dol.

Anthraquinone-induced cell injury : acute toxicity of carminomycin, epirubicin, idarubicin and mitoxantrone in isolated cardiomyocytes

Acute toxic effects of the antineoplastic anthraquinones carminomycin, epirubicin, idarubicin and mitoxantrone were studied in primary cultures of cardiomyocytes, which were isolated from adult rats. Both time- and concentration-dependent changes of cell structure and viability (trypan blue exclusion) following incubation of myocytes with subclinical, clinical and toxic concentrations of the anthraquinones were examined by light microscopy. The area under the decay curve of viable and rod-shaped myocytes was used to express cytotoxicity of the drugs. Mitoxantrone was found to reduce cell viability and number of rod-shaped cells to the greatest extent, followed by carminomycin, idarubicin and epirubicin. A significantly lower accumulation in cardiomyocytes was obtained with epirubicin and idarubicin compared with carminomycin. An inhibitory effect on oxygen consumption by the cells occurred already at 0.1 microM with epirubicin, whereas inhibition caused by other anthraquinones was less pronounced. Our data indicate a weak association of net accumulation and the toxicity parameter IC50 for carminomycin and idarubicin. In contrast to these results, a more significant correlation of cytotoxicity and anthraquinone lipophilicity was found, which suggests that the lipophilic character of a particular anthraquinone may be an important factor in drug-induced acute cardiotoxicity.

Doxorubicin- and daunorubicin-induced energy deprivation and nucleotide degradation in isolated cardiomyocytes.

Cytotoxic mechanisms of the antitumor agents daunorubicin and doxorubicin were elucidated in isolated cardiomyocytes from adult rats. Incubation with daunorubicin resulted in a concentration-dependent loss of cell viability and changes of the cell structure. Only the highest concentration of doxorubicin (1 mM) caused similar effects. Doxorubicin was found to stimulate oxygen consumption by cardiomyocytes (about 20%), while the opposite effect was observed after daunorubicin treatment. A rapid decrease of the mitochondrial ATP content (more than 40%) and elevation of the cytosolic ADP level (doxorubicin 2-fold and daunorubicin 6-fold) was followed by increased release of adenosine and inosine to the surrounding medium. When myocytes were exposed to an anthracycline concentration lower than plasma levels measured clinically (0.15 microM), doxorubicin and daunorubicin significantly decreased the intracellular ADP and NAD levels. Isolated cardiomyocytes were found to be able to form daunorubicinol from daunorubicin. In contrast, no conversion of doxorubicin was detected in our experiments. In conclusion, our data demonstrate that decreased ATP production and increased nucleoside formation are major events in the toxicity induced by daunorubicin and doxorubicin in isolated cardiomyocytes. The results also suggest that the toxic effects may be caused by separate mechanisms.

Dolichol-like lipids with stimulatory effect on DNA synthesis: Substrates for protein dolichylation?

Substantial evidence has suggested that a nonsterol product of mevalonic acid (MVA) is essential for the initiation of DNA synthesis in mammalian cells. Several possible isoprenoid candidates have been suggested, but the identity of this compound still remains unknown. In this study we have isolated and purified MVA products from SV40‐transformed human fibroblasts and identified fractions with a growth‐stimulatory effect. The cells were labelled with [14C]MVA in the presence of inhibitors of 3‐hydroxy‐3‐methylglutaryl coenzyme A (HMG‐CoA) reductase. After lipid extraction, the [14C]MVA‐labelled lipids were subjected to high performance liquid chromatography and size‐exclusion chromatography, and the effect of the fractionated eluate on the DNA synthesis of arrested MVA‐depleted target cells was tested. Thereby we found a fraction of [14C]MVA‐labelled lipids with a substantial stimulatory effect on DNA synthesis. The chromatographic behavior suggested that the growth‐stimulating fractions contained dolichol‐20. This was confirmed by mass spectrometric analysis. Similar results were obtained when lipids from hepatocellular carcinoma cells and a sample from breast tumor were isolated and analyzed by the same procedure. The mechanisms by which these compounds induce DNA synthesis are unknown. Recent data obtained in our laboratory have provided evidence that dolichyl groups are covalently linked to tumor cell proteins, which implicates a new biological function for long‐chain polyisoprenoid alcohols (Hjertman et al. [1997] FEBS Lett 416:235–238). In this study we demonstrate that tumor cells containing dolichol‐like growth‐stimulatory lipids also contained dolichylated proteins. This raises the question whether the growth‐stimulatory dolichol‐like lipids serve as substrates for the dolichylation reaction. J. Cell. Biochem. 71:502–514, 1998.