Potentiation of antitumour activity of docetaxel by combination with trastuzumab in a human prostate cancer xenograft model and underlying mechanisms.

Antitumour activity of docetaxel (Taxotere) in hormone-dependent (HD) and hormone-independent (HID) prostate cancer PAC120 xenograft model was previously reported, and its level was associated with HER2 protein expression. In the present study, we evaluate the antitumour effects of docetaxel combined with trastuzumab (Herceptin), an anti-HER2 antibody. Although trastuzumab alone had no effect on tumour growth, it potentiated the antitumour activity of docetaxel in HD tumours and more strongly in HID variants. Using the HID28 variant, we show that docetaxel treatment of tumour-bearing mice induces an increased HER2 mRNA expression of the tyrosine kinase receptor of 25-fold 24 h after docetaxel treatment, while HER2 protein and p-AKT decreased. This was followed by an increase of HER2 protein 3 days (two-fold) after docetaxel treatment and by a strong HER2 release in the serum of treated mice; expression of phospho-ERK, p27, BCL2 and HSP70 concomitantly increased. Similar molecular alterations were induced by docetaxel plus trastuzumab combination, except for that there was a transient and complete disappearance of AR and HSP90 proteins 24 h after treatment. We show that in addition to its known effects on tubulin and mitotic spindles, docetaxel induces complex signalisation pathway mechanisms in surviving cells, including HER2, which can be pharmacologically targeted. This study suggests that the docetaxel/trastuzumab combination may prove an effective therapeutic approach for HER2-expressing hormone-refractory prostate cancer.

Mucinous differentiation features associated with hormonal escape in a human prostate cancer xenograft.

Many theories mention hypersensitive, promiscuous, outlaw or bypass signalling pathways to explain the acquisition of hormone independence in prostate cancer. Hormonal escape of prostate tumours is marked by many biological changes, including mucinous and neuroendocrine differentiation. Since expression of several mucins has been linked to carcinoma tumour progression, we have characterised the expression of mucins at both RNA and protein levels in an in vivo model of prostate cancer in hormonal escape. Using PAC120, a xenograft of a human hormone-dependent prostate tumour, and its hormone-independent variants, we analysed the expression of mucins (MUC1, MUC2, MUC4, MUC5AC, MUC5B, MUC6) by immunohistochemistry or reverse transcriptase (RT)-PCR. While the parental PAC120 tumour was a compact poorly-differentiated tumour of Gleason score 9 (5+4), hormone-independent variants displayed mucinous, neuroendocrine-like or mixed histological changes; these changes were stable through serial transplantations or after testosterone supply. MUC1 mRNA was expressed in both PAC120 and the hormone-independent variants, although at variable levels. All tumours displayed a high and constant expression of MUC2 and no expression of MUC4 mRNA. While MUC1 was expressed in all xenografts whatever their hormone dependence status, MUC2, MUC5B and MUC6 were preferentially expressed in hormone-independent variants. The loss of hormone dependence in this prostate cancer xenograft model is therefore marked by irreversible histological alterations, mucinous or neuro-endocrine, associated with an expression of secretory MUC2, MUC5B and MUC6, independent of the histological differentiation subtype. These data point to mucinous differentiation as an important step in the acquisition of hormone independence in this cancer, and suggest that secretory mucins might participate in an unknown pathway of hormonal escape in prostate cancer.

Molecular epidemiology of human herpesvirus 8 in africa: both B and A5 K1 genotypes, as well as the M and P genotypes of K14.1/K15 loci, are frequent and widespread.

We have studied 52 new HHV8 strains by sequencing the complete hypervariable K1 gene and genotyping the K14.1/K15 loci located at both sides, respectively, of the viral genome. The samples originated from 49 patients with Kaposi's sarcoma (KS; 32 patients), multicentric Castleman's disease (MCD; 12 patients), or primary effusion lymphoma (PEL; 5 patients). Among these patients, 32 were of African origin (West and Central African countries and Creoles from French Guiana) and the 17 others were mostly French homosexuals. Comprehensive phylogenetic studies allowed the identification of distinct groups within the three already known main subtypes. Interestingly, two new sequences that did not cluster within a known subtype or group could be considered as prototypes of early/ancient variants of the C subtype and A/C set, respectively. Among the 32 African strains, the majority were either of the B subtype (13 cases) or of the A5 group (11 cases), indicating that this latter genotype is frequent and widespread in Africa. In contrast, a subtype C strain infected most of the 17 other patients. PCR-based genotyping of the K14.1/K15 loci revealed an overall predominance of P subtype, except in the A5 and B K1 groups, in which the P and M alleles were equally represented. The implications of these data on the evolution and spread of HHV8 among human African populations are discussed.

Virological and molecular characterisation of a new B lymphoid cell line, established from an AIDS patient with primary effusion lymphoma, harbouring both KSHV/HHV8 and EBV viruses.

We report here a new case of primary effusion lymphoma (PEL), occurring in a French homosexual HIV-1 infected male with a pericardial, pleural and mesenteric tumour dissemination, and the establishment from his pleural effusion of a new cell line, Cra-BCBL, dually infected by EBV and KSHV/HHV8. Cra-BCBL cells are of B-cell origin as judged by their clonal immunoglobulin heavy chain (IgH) gene rearrangement, identical to that of the parental tumour. Both the cell line and the lymphoma cells expressed CD38 and CD45 antigens but no classical B-cell or T-cell lineage-restricted antigens. Cra-BCBL harbours a type I EBV virus, expressing a latency type II. Expression of KSHV/HHV8 ORF72 and ORF75 was detected by RT/PCR. In addition, KSHV lytic replication could be induced by treatment by n-butyrate. An equivalent and high copy number of KSHV genomes (20 to 200 copies by cell) was detected both in the primary tumour cells and in the cell line. Southern blot (SB) analysis of EBV terminal repeats (TR) displayed the same unique band in the cell line DNA and in the original tumour cells, consistent with a monoclonal infection of EBV. Furthermore, SB analysis of KSHV/HHV8 TR revealed the same hybridisation pattern between Cra-BCBL and the effusion cells, with a common band at around 30-40 kb corresponding to the fused termini of the viral episomes and a 5 Kb rearranged fragment. The new cell line characterised here could be a useful model to study interactions between two human herpes viruses and their contribution to lymphomagenesis.

Monoclonality or oligoclonality of human herpesvirus 8 terminal repeat sequences in Kaposi's sarcoma and other diseases.

BACKGROUND: Infection with human herpesvirus 8 (HHV8), also termed Kaposi's sarcoma (KS)-associated herpesvirus, is associated with all forms of KS, with primary effusion lymphoma (PEL), and with some forms of multicentric Castleman's disease (MCD), but the pathogenic role of HHV8 in these tumors and the clonal nature of KS are still unclear. The purpose of this study was to examine whether the number of terminal repeats (TRs) contained in the fused TR region of HHV8 could be used as a marker of clonality in HHV8-associated tumors. METHODS: Pulsed-field gel electrophoresis (PFGE) and multiple-probe Southern blot analysis of the HHV8 TR region were performed on high-molecular-weight DNA obtained from tumoral KS, PEL, and MCD lesions. RESULTS: These analysis showed that the fused TR region contains a large but variable number of TR units (ranging from 16 to 75) and that the viral genome is present as extrachromosomal circular DNA in these tumors in vivo, with occasional ladders of heterogeneous linear termini reflecting lytic replication. All PEL tumors and PEL-derived cell lines as well as some KS tumors contained monoclonal or oligoclonal fused TR fragments; however, the TR region appeared polyclonal in MCD tumors and in a few KS lesions. CONCLUSION: Several KS and PEL lesions are monoclonal expansions of a single infected cell, suggesting that HHV8 infection precedes tumor growth and thus supporting an etiologic role of latent HHV8 in these proliferations. Our finding that nodular KS lesions display all possible patterns of clonality supports the model according to which KS begins as a polyclonal disease with subsequent evolution to a monoclonal process.

Molecular characterization of Kaposi's sarcoma-associated herpesvirus/human herpesvirus-8 strains from Russia.

We report the molecular characterization, with subtyping of both K1 and K14.1/K15 genomic regions, of seven new human herpesvirus-8 (HHV-8) strains from Russian patients with classical Kaposi's sarcoma. Phylogenetic studies, based on the complete K1 gene/protein analysis, indicate that six of these strains belong to the A subtype, with one belonging to the A4 group and exhibiting a unique deletion of 19 amino acids in the VR2 region at position 186-204. PCR-based studies of the K14.1/K15 genomic region indicate that four of the new strains were of the M subtype while three belonged to the P subtype. Our study indicates an important genetic diversity of the HHV-8 strains currently present in Russia, including a new peculiar strain possessing a unique deletion in the VR2 segment, and confirms the absence of correlation between the K1 and K14.1/K15 molecular subtypes, as M and P genotypes can be observed in the A K1 subtype.

Sequence variations in EBNA-1 may dictate restriction of tissue distribution of Epstein-Barr virus in normal and tumour cells.

In seropositive individuals Epstein-Barr virus (EBV) establishes a virus reservoir in peripheral blood lymphocytes (PBLs). Transmission from one individual to another occurs via saliva due to a lytic (virion productive) phase of infection in the oropharynx. EBNA-1 is responsible for maintaining viral episomes in the host cell and could, therefore, also affect the persistence of the virus in different cell lineages. Based on sequence analysis of EBNA-1 we now demonstrate that (i) in addition to the prototype EBNA-1 (identical to the B95.8 virus EBNA-1), EBV in normal individuals encompasses multiple EBNA-1 subtypes, both in PBLs and in oral secretions; (ii) although EBV with prototype EBNA-1 is the predominant virus in normal individuals, it is very rarely associated with either nasopharyngeal carcinoma (NPC) or Burkitt's lymphoma (BL); (iii) EBV with an EBNA-1 subtype (V-val) frequently associated with NPC is also selectively detected in oral secretions and not in PBLs; (iv) EBV with the EBNA-1 subtype V-pro is restricted to PBLs, while a mutated version of this subtype is present in BL, but not in NPC. These findings suggest that the variations in EBNA-1 may be relevant to the ability of EBV to persist in different cell types, and hence relevant to its oncogenic potential.

Variation in the sequence of Epstein Barr virus nuclear antigen 1 in normal peripheral blood lymphocytes and in Burkitt's lymphomas.

We have examined sequence variations in the EBNA-1 protein of EBV in normal peripheral blood lymphocytes (PBL) and Burkitt's lymphomas (BL). We find two EBNA-1 strains P (prototype) and V (variant) which differ by 15 amino acids. Each strain has two subtypes defined by the amino acid at position 487 (P-ala, P-thr, V-pro and V-leu). In PBLs from 32 normal individuals, up to three of these subtypes were found in each sample, but the V strain did not occur in the absence of P strain viruses, nor was the V-leu subtype ever observed in normal PBL. In BLs only a single subtype was observed in each tumor. The P-thr and V-leu subtypes were more frequently seen than the P-ala and V-pro subtypes, which occurred in only two and one of the 36 tumor samples respectively. The P-thr was the most commonly observed subtype in peripheral blood of both American and African lymphocytes as well as in African tumors. However, in 11 of 12 American tumors, the EBNA-1 subtype was V-leu. These data indicate that some EBNA-1 subtypes are more likely to lead to oncogenesis, and one subtype, V-leu, appears only to occur in tumors.

Use of Epstein-Barr virus nuclear antigen-1 in targeted therapy of EBV-associated neoplasia.

To target expression of toxic genes to Epstein-Barr virus (EBV)-associated tumor cells, we have developed an EBV-driven enzyme prodrug system (EDEPS) that takes advantage of the trans-activating properties of EBNA1, a latent protein expressed in all EBV-containing cells, to direct expression of cytosine deaminase (CD) at high levels in those cells only. Plasmids were constructed in which the CD gene or a luciferase reporter gene were cloned downstream of the herpes simplex virus thymidine kinase (tk) promoter and the family of repeats (FR) sequence from the oriP region of EBV. Analysis of luciferase activity after transient transfection into a panel of EBV-negative or -positive human cell lines showed that the presence of the FR element enhanced transcription from the tk promoter in all EBV-positive cell lines, whereas transcription from tk was repressed in all EBV-negative cell lines, including B, T, and fibroblast cell lines. In clonogenicity assays following transfection with the CD vector, the presence of 5-fluorocytosine (5-FC) in the culture medium completely abolished cell growth in EBV-positive cell lines, but did not affect the growth of EBV-negative cell lines. This vector system should have wide applicability in that it allows targeted expression of any gene of interest to tumors that carry EBV, irrespective of the role EBV plays in their pathogenesis.

Switching viral latency to viral lysis: a novel therapeutic approach for Epstein-Barr virus-associated neoplasia.

We describe an EBV-driven lytic system (LySED) that can be used to specifically target therapy to EBV- containing tumors. This system takes advantage of the transactivating properties of EBNA-1, a latency protein expressed in all EBV-containing cells, to drive the expression of Zta, a gene sufficient for inducing the EBV lytic cycle. Thus, EBV provides both the target and the executor for mediating tumor-specific cell death, markedly increasing the specificity of the system. Transfection of EBV-positive cell lines with the LySED construct resulted in a switch to lytic cycle and subsequent cell death, even in the presence of an inhibitor of EBV thymidine kinase (acyclovir) without an increase in virion production. In contrast, growth of EBV-negative B-cell lines was not affected.

Metastatic human rhabdomyosarcoma: molecular, cellular and cytogenetic analysis of a novel cellular model.

A new human metastatic rhabdomyosarcoma (RMS) model was established and analyzed for a number of biologic, cytogenetic and molecular parameters. Consistent with previous studies, the metastatic capacity of different RMS cell variants did not correlate with their tumorigenic or proliferative capacities. Interestingly, a highly metastatic variant was diploid, while a nonmetastatic variant was tetraploid, which parallels previous clinical observations. Genes whose expression had been found to be associated with either low- or high-metastatic capacity in carcinoma or melanoma did not show a similar association with different metastatic variants of RMS, derived from a mesenchymal tumor. We also found, in transient reporter gene assays, that several promoters had higher transcriptional activity in highly metastatic than in nonmetastatic RMS cell variants. This novel human RMS metastatic model may be instrumental for a better understanding of the regulatory pathways that control the metastatic phenotype of tumors of mesenchymal origin.

A mutant p21 cyclin-dependent kinase inhibitor isolated from a Burkitt's lymphoma.

The growth arrest mediated by p53 is caused at least in part by the p53 mediated expression of p21 (p21waf1/Cip1). Since only one-third of primary Burkitt's lymphomas (BL) demonstrate mutations in the p53 gene, we examined the structural integrity of the p21 coding region by single-strand conformational polymorphism and DNA sequence analysis to determine the extent to which this gene is mutated in BL. Of 34 BLs analyzed, a frequent change (38%) at codon 31 that replaced Ser with Arg was found in 13 samples, 10 of which were from Africa. This change at codon 31 is also detected in peripheral blood DNA from normal subjects and may thus represent a polymorphism. One BL cell line, DH978, carried a change at codon 63: Phe to Leu. This mutation was heterozygous, and both the wild-type and the mutated p21 mRNA were expressed in the tumor cell line. By transfection experiments, the mutant p21 was less efficient in suppressing clonogenicity than wild-type p21. To our knowledge, this is the only mutation described in p21. The availability of this mutant p21 should further help in functional studies of p21.

Variable IgH chain enhancer activity in Burkitt's lymphomas suggests an additional, direct mechanism of c-myc deregulation.

The deregulation of the c-myc gene in small non-cleaved cell lymphomas (SNCL) with 8;14 translocations is thought to be due to the juxtaposition of this gene with the IgH chain locus, but exactly how the Ig locus contributes to the deregulation is unclear. One widely considered hypothesis is that Ig gene enhancers, when moved near c-myc, might stimulate inappropriate transcription of this gene. To evaluate this hypothesis, we have tested the ability of the two known H chain enhancers, the JH-C mu intronic enhancer and the 3' alpha enhancer, to support transcription from c-myc promoters, by using transient transfection assays in a panel of 32 SNCL cell lines with 8;14 translocations. The activity of the JH-C mu intronic enhancer varied widely among the cell lines tested and correlated with the presence of nuclear factors binding to the E4/octamer regions of the enhancer. The 3' alpha enhancer was much less active than the intronic enhancer in all of our cell lines and seems unlikely to account for the c-myc deregulation (although the enhancer we tested was from the rat, because the human homologue is unidentified at present). A marked difference was also seen in the ability of individual cell lines to support transcription from the unenhanced c-myc constructs. Cell lines supporting the lowest enhancer activity tended to support the highest level of transcription from unenhanced c-myc promoters. The differences in transcription from c-myc promoters were confirmed by stably transfecting constructs containing c-myc promoters in SNCL cell lines. Our data strongly suggest that the importance of Ig enhancers for c-myc deregulation varies markedly in different cell lines and that, for many, trans-acting regulators of the c-myc promoters are as important or more important, than Ig enhancers in the deregulation of the c-myc gene.

Characterization of the human immunoglobulin kappa gene 3' enhancer: functional importance of three motifs that demonstrate B-cell-specific in vivo footprints.

Using a combination of in vivo footprinting and site-directed mutagenesis, we have functionally characterized an enhancer located 12 kb downstream of the human immunoglobulin kappa constant-region gene. The core enhancer region is highly homologous to the murine 3' kappa enhancer. However, in addition to two regulatory elements homologous to the functional motifs of the murine enhancer, we find a third positive regulatory element in the human enhancer. This element is associated with an 11/12-bp direct repeat (DR) that is well conserved in the murine locus but was not recognized as functionally important in the murine enhancer. Mutation of any of the three motifs of the human enhancer decreases its activity to 3 to 20% of the wild-type level, indicating cooperative interaction between these elements. The DR motif does not resemble any known enhancer element and does not appear to function as a transcriptional activator on its own when present in multiple copies. Interestingly, nuclear extracts from both B- and T-cell lines contain factors binding to DR in vitro, but in vivo footprinting shows no evidence of protein-DNA binding in the T-cell line. This finding suggests that an additional regulatory mechanism, such as the effect of chromatin configuration on accessibility, may be involved in the B-cell-restricted activity of the human 3' kappa enhancer.

Biochemical analysis of the role of transmethylation in the methionine dependence of tumor cells.

The effects on methionine metabolism of the substitution of homocysteine for methionine in vitro were investigated in normal and tumor cell lines differing in their ability to utilize homocysteine for growth. The major finding of this study was that methionine-independent (Met-Indep) cell lines had much lower basal transmethylation rates than methionine-dependent (Met-Dep) cell lines. This was particularly evident in the parent SP1 cell line and its Met-Indep revertant, SP1-R. SP1-R compensated for a lack of methionine by reducing both its transmethylation and growth rates. An analysis of other potential differences in methionine metabolism between Met-Dep and Met-Indep cell lines failed to demonstrate any consistent abnormalities in all but the absolutely Met-Dep MDAY-D2 cell line. Thus, protein, S-adenosylmethionine, and polyamine synthesis were the same in Met-Dep and Met-Indep cell lines. These results indicate that the major regulatory step in determining the Met-Dep phenotype is an inherent increase in the rate of transmethylation reactions. Cell lines with high basal transmethylation rates cannot compensate for a relative deficiency of methionine and either cease growing (MDAY-D2) or generate revertants (SP1-R) for which the basal rate of transmethylation is considerably reduced.

Genotypic and phenotypic evidence of clonal interactions in murine tumor cells.

The stability of mixed tumor cell populations has been described in terms of phenotypic characteristics such as metastatic potential, immunogenicity, and drug resistance. We have extended these analyses to the molecular level in a model that uses transfection of the hemagglutinin antigen (HA) gene of influenza virus into murine CT-26 colorectal carcinoma cells. Transfection was followed by fluorescence-activated cell sorting (FACS) to select a parent population with expression of high levels of HA. We characterized this population (FACS-3) and four derived clones (5, 6, 9, and 18) over time with regard to phenotypic characteristics: immunogenicity and cross-protection against tumor challenge, cell surface expression of HA, evidence of HA gene amplification, and levels of HA mRNA. During 6 months in culture, the FACS-3 parent cells remained stable, but the individual clones varied for all of the parameters assessed. Among the clones, all possible molecular variations occurred, including changes in HA gene copy number (increased in clone 5 and decreased in clone 18), gene rearrangement (clone 5), decrease in HA mRNA (clones 6, 9, and 18), increase in HA mRNA (clone 5), and an abnormality in translational control or a posttranslational error. In all cases, the molecular changes correlated with cell surface HA expression, immunogenicity, and cross-protective potential. We conclude that in vitro clonal interactions play a role in stabilizing heterogeneity in this system. These studies show that even in the absence of selection, clonal interactions may alter the phenotype of tumors by increasing malignant progression or impeding tumor growth.

Patterns of methionine auxotrophy in normal and neoplastic cells: the methionine independence of lymphocyte mitogenesis and low frequency of the methionine-dependent phenotype in human tumors.

Seven murine and 17 human tumor-derived cell lines were tested for their ability to grow in methionine-free medium containing the methionine precursor homocysteine. Three murine tumors, SP1, MDAY-D2, and L1210, failed to grow in this medium and were therefore methionine dependent (Met-Dep). In contrast, all human tumors, including 8 recently established cell lines, were methionine independent (Met-Indep). Concanavalin A-induced lymphocyte mitogenesis was also Met-Indep but required 3 to 4 times the amount of homocysteine needed for the growth of normal fibroblasts or Met-Indep tumors. In addition, lymphocyte mitogenesis was also supported by exogenous 5'-methylthioadenosine, another methionine precursor formed during polyamine synthesis. In contrast, Met-Dep tumors did not respond to increasing homocysteine concentration, nor was their growth supported by 5'-methylthioadenosine. These findings demonstrate that Met-Dep can occur by varied mechanisms relating to such parameters as homocysteine concentration and the ability of cells to generate Met-Indep revertants or to grow in 5'-methylthioadenosine. In general, we found the Met-Dep phenotype to be more common in murine tumor cells and to occur infrequently in human tumors. This may imply a species difference in methionine metabolism.

Inhibition of rat natural killer cell function by carcinogenic nickel compounds: preventive action of manganese.

The effects of carcinogenic nickel [(Ni) CAS: 7440-02-0] and Ni compounds on the natural killer (NK) cell activity of rat peripheral blood mononuclear cells (PBMCs) were studied. Rhabdomyosarcomas were locally induced by one im injection of Ni or Ni subsulfide [(Ni3S2) CAS: 12035-72-2] dust in the hind leg of WAG rats. A weakly tumorigenic dose of 5 mg Ni3S2 (tumor incidence, 2%) induced a transient decrease of PBMC NK activity against YAC-1 cells in vitro (from the 17th to the 23d wk after Ni3S2 inoculation), which could be restored by in vivo injections of partially purified rat fibroblastic interferon (IFN). Injection of 20 mg Ni (tumor incidence, 47.5%) produced a long-lasting depression of NK cell activity (from the 8th to the 23d wk). In vivo chronic IFN treatment of the Ni-injected rats neither restored NK cell activity nor affected the tumor incidence. However, NK cells of Ni-treated animals responded normally to IFN in vitro. Prospective analysis of individual NK cell responses showed that a persistent depression of basal NK cell activity was restricted to rats that subsequently developed a tumor. In these animals the time between carcinogen treatment and clinical detection of the primary tumor was positively correlated with the mean level of NK cell activity (3-4 determinations/rat). Admixture of manganese to Ni inhibited the development of tumors and also prevented the depression of NK cell activity produced by Ni alone. Noncarcinogenic Ni oxide stimulated NK cell activity. These results point out the possible involvement of NK cells in resistance to Ni-induced carcinogenesis.

Influence of the uremic state on the development of malignancy. An experimental study in the rat.

The effect of chronic uremia on the development of subcutaneously injected malignant tumoral cells was evaluated in 213 male Wistar AG rats made chronically uremic by simultaneous right nephrectomy and partial ligation of the left renal artery. The tumoral cells injected were stemming from a parental rhabdomyosarcoma (9-4/0) induced by intramuscular injection of 20 mg of colloidal nickel suspended in oil to a male Wistar rat. 54 sham-operated rats and 43 nonoperated animals served as control-groups. Renal function and tumoral growth were checked weekly up to the 60th postoperative day, at which time the surviving rats were sacrificed and submitted to autopsy. At day 15 after cell grafting, a tumoral lump could be felt by finger touch in 68% of the uremic rats, but in only 11% of the sham-operated and 14% of the nonoperated controls (p less than 0.0001). Throughout the study, the tumoral lumps which developed in the uremic animals were of significantly larger size than in the nonuremic controls. Pulmonary tumoral metastases were evidenced at autopsy in 95% of the uremic rats, but in only 50% of the sham-operated and in 54% of the nonoperated controls (p less than 0.005). These results indicate an apparently accelerating and amplifying effect of uremia on the development of a malignant tumor in the rat, for which a decrease in cell-mediated immunity associated with the uremic state still remains a questionable hypothesis.

Electron microprobe in vitro study of interaction of carcinogenic nickel compounds with tumour cells.

The effect of various nickel salts on cultured rhabdomyosarcoma cells was studied. Certain of these compounds, e.g., nickel subsulfide (Ni3S2) and nickel itself, induce tumours in muscle, while others have no effect, e.g., nickel monoxide (NiO). It has been suggested that the carcinogenicity of nickel is related to its penetrating power (phagocytosis) in transformed cells. We have used electron microscopy and microanalysis to study the ultrastructure and intracellular localization of nickel in ultra-thin sections. Nickel subsulfide and nickel monoxide penetrate into cells and are concentrated in vacuoles, exhibiting a particular affinity for membrane structures. They subsequently appear to be eliminated in the extracellular medium. Colloidal nickel and iron carbonyl, on the other hand, do not penetrate cells. Various tumoral and normal cell lines were compared for their ability to phagocytose nickel subsulfide and it was found that the compound penetrated only macrophages and impregnated the membranes of polynuclear cells. These results suggest that the phagocytosis of nickel compounds is not directly related to the eventual induction of a tumour. No nuclear localization could be detected, but we did demonstrate a mechanism for the concentration and elimination of these compounds in certain tumour cells.