Segmentation genes provide the signals for the activation and regulation of homeotic genes in Drosophila but cannot maintain the resulting pattern of expression because their activity ceases halfway through embryogenesis. Maintenance of the pattern is due to the Polycomb group of genes (Pc‐G) and the trithorax group of genes (trx‐G), responsible for the persistence of the active or repressed state of homeotic genes. We have identified a regulatory element in the Ubx gene that responds to Pc‐G and trx‐G genes. Transposons carrying this element create new binding sites for Pc‐G products in the polytene chromosomes. This Pc‐G maintenance element (PRE), establishes a repressive complex that keeps enhancers repressed in cells in which they were originally repressed and maintains this state through many cell divisions. The trx‐G products stimulate the expression of enhancers in cells in which they were originally active. This mechanism is responsible for the correct regulation of imaginal disc enhancers, which lack themselves antero‐posterior positional information. The PRE also causes severe variegation of the mini‐white gene present in the transposon, a phenomenon very similar to heterochromatic position‐effect variegation. The significance of this mechanism for homeotic gene regulation is discussed.

Polycomb group genes are necessary for maintaining homeotic genes repressed in appropriate parts of the body plan. Some of these genes, e.g. Psc, Su(z)2 and E(z), are also modifiers of the zeste‐white interaction. The products of Psc and Su(z)2 were immunohistochemically detected at 80–90 sites on polytene chromosomes. The chromosomal binding sites of these two proteins were compared with those of zeste protein and two other Polycomb group proteins, Polycomb and polyhomeotic. The five proteins co‐localize at a large number of sites, suggesting that they frequently act together on target genes. In larvae carrying a temperature sensitive mutation in another Polycomb group gene, E(z), the Su(z)2 and Psc products become dissociated from chromatin at non‐permissive temperatures from most but not all sites, while the binding of the zeste protein is unaffected. The polytene chromosomes in these mutant larvae acquire a decondensed appearance, frequently losing characteristic constrictions. These results suggest that the binding of at least some Polycomb group proteins requires interactions with other members of the group and, although zeste can bind independently, its repressive effect on white involves the presence of at least some of the Polycomb group proteins.

Purpose: Advanced melanoma is a highly drug-refractory neoplasm representing a significant unmet medical need. We sought to identify melanoma-associated cell surface molecules and to develop as well as preclinically test immunotherapeutic reagents designed to exploit such targets. Experimental Design and Results: By transcript profiling, we identified glycoprotein NMB (GPNMB) as a gene that is expressed by most metastatic melanoma samples examined. GPNMB is predicted to be a transmembrane protein, thus making it a potential immunotherapeutic target in the treatment of this disease. A fully human monoclonal antibody, designated CR011, was generated to the extracellular domain of GPNMB and characterized for growth-inhibitory activity against melanoma. The CR011monoclonal antibody showed surface staining of most melanoma cell lines by flow cytometry and reacted with a majority of metastatic melanoma specimens by immunohistochemistry. CR011alone did not inhibit the growth of melanoma cells. However, when linked to the cytotoxic agent monomethylauristatin E (MMAE) to generate the CR011-vcMMAE antibody-drug conjugate, this reagent now potently and specifically inhibited the growth of GPNMB-positive melanoma cells in vitro. Ectopic overexpression and small interfering RNA transfection studies showed that GPNMB expression is both necessary and sufficient for sensitivity to low concentrations of CR011-vcMMAE. In a melanoma xenograft model, CR011-vcMMAE induced significant dose-proportional antitumor effects, including complete regressions, at doses as low as 1.25 mg/kg. Conclusion: These preclinical results support the continued evaluation of CR011-vcMMAE for the treatment of melanoma.

Medulloblastoma is the most malignant pediatric brain tumor. It is believed to originate from the undifferentiated external granule layer cells in the cerebellum, but the mechanism of tumorigenesis remains unknown. Here we studied three types of human medulloblastoma cells that express markers corresponding to different levels of neuronal differentiation. They expressed the neuronal repressor element 1 (RE1) silencing transcription factor/neuron-restrictive silencer factor (REST/NRSF; refs. 7-10) at very high levels compared with either neuronal progenitor NTera2 (NT2) cells or fully differentiated human neuron teratocarcinoma (hNT cells). To counter the effect of REST/NRSF, we used a recombinant transcription factor, REST-VP16, constructed by replacing repressor domains of REST/NRSF with the activation domain of viral protein (VP16). Transient expression of REST-VP16 in medulloblastoma cells was able to compete with the endogenous REST/NRSF for DNA binding and stimulate neuronal promoters. High-efficiency expression of REST-VP16 mediated by adenovirus vectors (Ad.REST-VP16) in medulloblastoma cells was able to counter REST/NRSF-mediated repression of neuronal promoters, stimulate expression of endogenous neuronal genes and trigger apoptosis through the activation of caspase cascades. Furthermore, intratumoral injection of Ad.REST-VP16 in established medulloblastoma tumors in nude mice inhibited their growth. Therefore, REST/NRSF may serve as a new target for therapeutic interventions for medulloblastoma through agents such as REST-VP16.

Maleless (mle) is essential in Drosophila melanogaster males both in somatic cells and in germ cells. In somatic cells mle is necessary for X-chromosome dosage compensation. The role of mle in the germline is unknown. We have analyzed the expression pattern and localization of MLE, the other MSLs and acetylated isoforms of histone H4 in male germ cells to address whether dosage compensation and/or X inactivation occur in the Drosophila germline. We observed that MLE is the only MSL expressed in the male germ cells and it is not localized to the X chromosome. We conclude that in the germline mle is not involved in chromosomal dosage compensation but may be involved in post-transcriptional gene regulation. We also observed that the acetylation pattern of histone H4 is very dynamic during spermatogenesis. While the pattern is not compatible with dosage compensation or X inactivation, it is consistent with all premeiotic chromosomes being in an active configuration that is replaced in post-meiotic stages with an inactive chromatin constitution.