Neuroblastoma: inhibition of progression (Part II). Basic science in pediatric surgery.

In neuroblastoma, amplification of the protooncogene N-myc is the most important molecular characteristic predicting a bad outcome for the patients. Despite the importance of the N-myc gene, little is known about the mechanisms regulating its expression. We found evidence that insulin-like growth factor II stimulates the growth of neuroblastoma in a paracrine fashion. Two neuroblastoma cell lines predominantly expressed IGF-II whereas two other cell lines expressed the IGF-receptor. In a receptor-positive cell line, N-myc expression was enhanced by stimulation with IGF-II. As the growth-stimulating signals of the IGF receptor are transmitted via Ras proteins, inactivation of Ras is one promising tool to prevent the induction of N-myc expression by IGF-II. Treatment of neuroblastoma cells with an inhibitor of the farnesyl-protein-transferase (FPTase) inactivated H-ras protein completely and N-ras protein by more than 50 %. Cell growth of neuroblastoma cells in serum containing medium was clearly diminished by inhibition of FPTase. The growth-promoting effect of IGF-II was reduced to exactly half the amount observed in non-inhibited cells.

Neuroblastoma: induction of differentiation (Part I). Basical science in pediatric surgery.

Neuroblastoma is the second-most common solid tumor in childhood. The majority of patients have a very poor outcome due to aggressive growth and metastatic spread. In contrast, in rare cases, spontaneous regression or differentiation towards a benign ganglioneuroma are observed. The mechanism leading to differentiation of neuroblastoma is of particular therapeutic interest. In this paper we report the results of our attempts to induce the expression of genes necessary for differentiation of neuroblastoma cells. TrkA codes for the high affinity receptor of NGF, a neurotrophin known to promote differentiation. Treatment with retinoic acid caused a 3-fold increase of the trkA expression in neuroblastoma cell lines. Neurofibromin, the gene product of the NF-1 gene, is involved in downregulation of the activity of ras-proteins. In contrast to immature neuronal tissues in mature brain, the type II isoform of neurofibromin is predominantly expressed. Retinoic acid was able to raise the proportion of type II NF-1 expressed in neuroblastoma cells.

Apoptosis of cultured neuroblastoma cells is induced by ceramide and not by ligation of the Fas/Apo-1/CD95 receptor.

The expression of the Fas/APO-1/CD95 receptor on the neuroblastoma cell lines IMR-32, Kelly, SK-N-SH, LS and SiMa was investigated. The induction of apoptosis was attempted by incubation with Fas antibody IgM and IgG, Fas ligand and with ceramide. All neuroblastoma cell lines proved to be positive for Fas/APO-1/CD95 receptor expression by FACS. However, propidium iodide staining in FACS showed that neither incubation with the Fas antibody nor with the Fas ligand resulted in convincing apoptosis of the neuroblastoma cells. On the other hand, incubation with ceramide rapidly led to all signs of apoptosis morphologically and in the FACS. Therefore, other pathways than Fas/Apo-1/CD95 ligation must be of significance for the apoptosis of these neuroblastoma cell lines.

Expression analysis of multidrug resistance associated genes in neuroblastomas.

In a series of 40 neuroblastomas we analyzed the relative mRNA levels of the MDR associated genes encoding MDR1/P-glycoprotein (MDR1), multidrug resistance associated protein (MRP), lung cancer resistance related protein (LRP) and topoisomerase IIalpha (TOPO IIalpha) by cDNA-PCR. Cyclin A (CYCA) was included to examine cellular proliferation activity. MYCN gene expression was analyzed as it was recently shown to be associated with enhanced MRP gene expression in neuroblastomas. We found that tumors with MYCN gene amplification exhibit significantly increased MYCN and MRP gene expression levels. Tumors with an allelic loss of the chromosomal 1p region showed significant (P<0.05) lower MDR1 gene expression (MDR1: 50+/-29, n=4) than tumors without (MDR1: 117+/-81, P<0.05, n=36). Moreover, significant positive correlations were found for MYCN/TOPO IIalpha (P<0.0001), MYCN/CYCA (P<0.05), TOPO IIalpha/CYCA (P<0.01), MRP/CYCA (P<0.0001) and MRP/LRP (P<0.05). Our results give evidence that MDR in neuroblastomas might be caused by multiple resistance factors and that a higher proliferation rate of neuroblastoma cells possibly based on altered MYCN gene expression is associated with enhanced MRP, CYCA and TOPO IIalpha gene expression.

Growth inhibition of neuroblastoma cells by lovastatin and L-ascorbic acid is based on different mechanisms.

Hydroxymethyl-glutaryl-CoA-reductase (HMG-CoA-reductase), the key enzyme for cholesterol synthesis and essential for the synthesis of the precursor for p21ras farnesylation, was inhibited in neuroblastoma cells by lovastatin or L-ascorbic acid. Both compounds inhibited clonogenic colony formation of neuroblastoma cells in soft agar. However, while the addition of mevalonate, the product of HMG-CoA-reductase, circumvented the inhibition by lovastatin it had no reversing effect on the inhibition by L-ascorbic acid. The role of reactive oxygen compounds generated by the degradation of catecholamines, and the pro-oxidative effects of L-ascorbic acid are discussed as mechanisms of action of L-ascorbic acid.

Inhibition of farnesyl-protein-transferase in neuroblastoma cells by alpha-hydroxyfarnesylphosphonate.

Oncogenic Ras is responsible for malignant transformation in a variety of tumors. Farnesylation of Ras by farnesyl-protein-transferase (FPTase) is necessary for membrane localization of Ras-proteins, a prerequisite for its biological activity. Although mutations in ras genes are rare in neuroblastoma inactivation of Ras by inhibition of the FPTase is of interest in neuroblastoma. In this tumor, amplification of N-myc is frequently observed and expression of N-myc is induced via Ras signaling. Farnesyl-protein-transferase of neuroblastoma cells is inhibited by alpha-hydroxyfarnesylphosphonate. In homogenates of the cell line SK-N-AS an ID50 = 6.5 microM is estimated, in SK-N-SH the ID50 is 3.4 microM. The consequences of the inhibition of FPTase on the membrane localization was examined by immunoblots. Western blots of membrane proteins analysed with H-ras and N-ras specific antibodies revealed that H-ras protein is more sensitive to the inhibition of FPTase than N-ras protein. After culturing neuroblastoma cells for 24 hrs in the presence of 20 microM alpha-hydroxyfarnesylphosphonate H-ras protein completely dissappeared from the membrane fraction whereas N-ras protein was only affected by 50%. K-ras was not detectable on Western blots of three neuroblastoma cell lines. The experiments showed that FPTase inhibitors are effective in neuroblastoma cells but for complete inactivation of N-ras stronger conditions are required than for H-ras.

SiMa, a new neuroblastoma cell line combining poor prognostic cytogenetic markers with high adrenergic differentiation.

We describe the establishment and characterization of a new neuroblastoma (Nb) cell line, SiMa, carrying the major recurrent chromosome changes associated with poor prognosis Nb, including amplification of N-MYC by formation of double minutes (dmin), der(1)t(1;17)(p35;q12) and der(22)t(17;22)(q22;p13), and loss of chromosome 11, documented at both initiation and late passage. In contrast to these cytogenetic stigmata of poor prognosis, analysis of catecholamine synthesis by high pressure liquid chromatography (HPLC) measurement revealed an advanced degree of adrenergic differentiation with high rates of 3,4-Dihydroxyphenylalanine (DOPA), noradrenaline, homovanillic acid (HVA), and vanillylmandelic acid (VMA) production. Contrastingly advanced differentiation and poor prognostic genetic markers combine to render SiMa a unique instrument for investigating the pathology and therapy of Nb.

Cholesterol based antineoplastic strategies.

Manipulation of cholesterol metabolism open several possibilities of interfering with the growth of malignant cells. Deprivation of cholesterol decreases the velocity of growth and alters the composition of the cell membrane. The high requirement for LDL of malignant cells can be utilized for drug targeting. Proliferation assays were performed with neuroblastoma cells and cell lines of acute myeloid leukemia deprived of cholesterol by inhibition of HMG-CoA-reductase or culture in LDL-deficient medium. The cholesterol content of the cell membrane when reduced to 50% had no effect on the toxicity of LAK-cells but the toxicity of the fluorescent dye merocyanine MC 540 was enhanced two-fold. LDL-mediated drug targeting to AML cells was performed with oxidized LDL and showed toxic reactions. These results proved that cholesterol deprivation could be used to support some therapeutic approaches.

Regulation of expression of two different transcripts of the NF-1 gene in neuroblastoma.

Neurofibromin, the gene product of the NF-1 gene is expressed in two isoforms. The m-RNA of NF-1 type II contains an insertion of 63 bases in the so called GAP-related domain, that distinguishes it from the type I transcript. By sequence homology neurofibromin is supposed to have a similar function in regulating activity of ras in intracellular signal transduction as the GTPase activating protein (GAP). Both transcripts of NF-1 are simultaneously expressed in different molar ratio in neuroblastoma tumors. 1) We examined 9 different neuroblastoma cell lines for the ratio of expression of NF-1 type I and type II. For quantification of the two transcripts we performed RT-PCR of the m-RNA of the neuroblastoma cells using primers designed to cover the GAP-related domain. We found values ranging from a more than 3-fold excess of type I transcript in the cell line Kelly to a slight excess of the type II transcript (I/II = 0.6) in the cell line IMR 5. 2) As there are indications that expression of NF-1 type II is related to the state of differentiation, we tried to shift expression of NF-1 from type I to type II by treatment of the neuroblastoma cells with retinoic acid. Treatment of Kelly cells with 5 microM retinoic acid for 24 h already lowered the excess of the type I transcript from 3-fold to an only 1.6-fold excess. An inversion of the molar ratio from an excess of the type I transcript to an excess of type II transcript would enable to investigate the different role of the two transcripts in the regulation of ras-activity and differentiation.

Toxic reactions of oxidized LDL on cells of acute myeloid leukemia.

In AML patients LDL concentration of the serum is reduced due to the high LDL receptor activity of the AML cells. This phenomenon enables the use of LDL particles as vehicles for drug targeting. Toxic lipid peroxides and aldehydes were introduced into LDL particles by the simple but effective oxidation with 10 microM CuSO4. Up to 250 nmol peroxides and 6 nmol malondialdehyde were formed per mg LDL protein within 30 h of oxidation. This oxidized LDL is effectively taken up by AML cells of the FAB type M3 and M5 indicating the presence of scavenger receptors on these cells. Within 96 h 61-84% of the AML cells are killed by the oxidized LDL. Our results open a possibility to achieve specificity for targeting lipophilic antineoplastic drugs towards AML cells using oxidized LDL as vehicles. The use of oxidized LDL as drug carrier is recommended for purging of AML bone marrow because hematopoietic stem cells that don't possess scavenger receptors are protected from toxic action.

[Molecular biology markers of neuroblastoma as the basis for a therapy concept]. / Molekularbiologische Merkmale des Neuroblastoms als Grundlage eines Therapiekonzepts.

Molecular biological markers that influence prognosis should be considered in the optimization of therapy for disseminated forms of neuroblastoma. Some of these markers include: N-myc amplification, deletion of 1p36, ploidy, and trk expression. Neuroblastoma without N-myc amplification and high trk expression show a favorable outcome and would enable a more gentle chemotherapy treatment. Neuroblastomas with N-myc amplification and lack of trk expression behave extremely unfavorably and require an even more intensive treatment.

Cytotoxicity of ether phospholipid BM 41.440 on neuroblastoma cells.

The thioether lysophospholipid BM 41.440 proved to be toxic against cells of two neuroblastoma cell lines in a dose- and time-dependent manner. The ID50 estimated in three different in vitro test systems declined from about 10 micrograms/ml after 24 h to 1 microgram/ml after a 1-week treatment of the neuroblastoma cells. These values are comparable to the ID50 found for neoplastic cells derived from other tissues. In comparison, hematopoietic progenitor cells (granulocyte/monocyte-colony-forming units) proved to be less sensitive to short-term treatment with BM 41.440. After long exposure to this drug the selectivity towards neuroblastoma cells decreased. This observation makes it unlikely that BM 41.440 can be used for treatment of neoplasia such as neuroblastoma, because only short-term treatment is acceptable considering the high bone marrow toxicity.

Inhibition of the membrane localization of p21 ras proteins by lovastatin in tumor cells possessing a mutated N-ras gene.

Mutated ras genes are found in a variety of human tumors. For biological activity the gene product p21 ras needs to be bound to the cell membrane by a farnesyl residue. Treatment of tumor cells with lovastatin reduces the availability of farnesyl pyrophosphate for the modification of the ras proteins. The membrane localization of p21 ras has been reduced by 30-36% after the tumor cells have grown in the presence of 10 microM lovastatin for 7 days. The extent of the inhibition depends on the growth kinetics of the cell lines.

Cytotoxic effects of m-[131I]- and m-[125I]iodobenzylguanidine on the human neuroblastoma cell lines SK-N-SH and SK-N-LO.

As we have reported recently, the human neuroblastoma cell line SK-N-SH is able to take up and store m-iodobenzylguanidine (mIBG). This is in contrast to several other neuroblastoma cell lines, among which are SK-N-LO cells. Both cell lines were used in cell killing experiments with unlabeled and radioactive-labeled mIBG. Using 1-200 microCi m-[131I]IBG (1 h incubation time), only SK-N-SH cells could to a large extent be destroyed in a dose-dependent manner. This effect is completely caused by the radioactive labeling of the molecule, because unlabeled mIBG proved not to be toxic in the concentration range used in experiments with radiolabeled mIBG (30 nM-3 microM). The killing effect was strongly reduced when m-[131I]IBG with low specific activity (0.2-0.3 mCi/mg) was used instead of 20-30 mCi/mg. Similar effects in both cell lines were obtained using m-[131I]-and m-[125I]IBG. SK-N-SH cells that survived a first treatment with m-[131I]IBG were less sensitive to a second treatment. SK-N-LO cells were more sensitive against m-[131I]- and m-[125I]IBG than SK-N-SH cells if both cell lines are exposed to these radioactive compounds over a long period of time (24 h). The reason that only SK-N-SH cells could be destroyed in short-term incubation experiments is that mIBG is stored for approximately 7 days in these cells only. SK-N-LO cells could only be destroyed to a significant degree if m-[131I]IBG was permanently present in the test system. Bone marrow stem cells (CFU-c) also proved to be sensitive against m-[131I]IBG, although the effects were less pronounced than on SK-N-SH cells.

Clinical experiences in the treatment of neuroblastoma with 131I-metaiodobenzylguanidine.

Treatment of neuroblastoma is an unsolved problem of pediatric oncology. In spite of highly intensified chemotherapy, the long-term survival rate of children with a metastatic neuroblastoma is below 10%. We therefore used 131I-metaiodobenzylguanidine (MIBG) for the first time to treat children with a neuroblastoma in relapse or primary unresponsiveness to chemotherapy. We had previously demonstrated that MIBG is useful for the scintigraphic imaging of neuroblastoma lesions and had investigated the cytotoxicity and uptake of MIBG in various neuroblastoma cell lines. We treated 6 children with neuroblastoma in a total of 19 courses. Three of the children suffered from a relapse of neuroblastoma; 3 had never gained a remission. Four of the 6 children lost their bone pain and fever during the first 3 days. In 5 of the 6 children the solid tumor as well as the bone marrow infiltration responded to MIBG treatment, with responses ranging from transitory decrease of the tumor mass to complete disappearance of abdominal tumors. We also witnessed a stabilization of osteolytic lesions, a decrease in elevated serum catecholamines, and a decrease in bone marrow infiltration. Five of the 6 children died of tumor progression 55-249 days after the first MIBG treatment.

Specific uptake of m-[125I]iodobenzylguanidine in the human neuroblastoma cell line SK-N-SH.

The uptake of m-[125I]iodobenzylguanidine (mIBG), a compound structurally analogous to the antihypertensive drug guanethidine, was examined in various human cell lines. Of three neuroblastoma lines, SK-N-LO, IMR-32, and SK-N-SH, only the last showed specific uptake of the compound. In contrast, only a nonspecific uptake could be demonstrated for the other neuroblastoma lines, as well as for an osteogenic sarcoma line (SAOS-2) and a melanoma line (IgR 3). Based on analyses of uptake characteristics from Lineweaver-Burk plots it is evident that two different transport mechanisms are responsible for mIBG uptake into SK-N-SH cells: a nonspecific diffusion mechanism, and a specific, active uptake system. The latter was dramatically reduced at 4 degrees compared to 37 degrees, as well as in the presence of ouabain or the absence of oxygen. A competitive inhibition of the transport of mIBG by norepinephrine was observed. When drug-treated SK-N-SH cells were incubated in fresh medium, 20 to 30% of mIBG was still retained in the SK-N-SH cells 24 h after the end of incubation with mIBG, whereas no mIBG was detectable in SK-N-LO cells already after 1 h.

The role of reactive oxygen compounds derived from 6-hydroxydopamine for bone marrow purging from neuroblastoma cells.

6-Hydroxydopamine(6-OHDA), a specific neurotoxin against sympathetic nerve cells, is a drug already used for purging of bone marrow from neuroblastoma cells before autologous bone marrow transplantation. However, we could not detect significant differences in the toxicity of 6-OHDA against neuroblastoma and other tumor cells under the purging conditions clinically used. In contrast, bone marrow stem cells were much more resistant. The unspecific toxic effect of 6-OHDA is caused by H2O2 or H2O2-derived products which are generated by auto-oxidation in the incubation medium before a significant amount of 6-OHDA is taken up by the cells. Withdrawal of oxygen during the incubation period and subsequent incubation with an oxygen containing medium led to a more specific destruction of neuroblastoma cells which can take up 6-OHDA selectively.