Mitochondria as targets for established and novel anti-cancer agents.

Chemoresistant cells have acquired the ability to evade the action of multiple classes of anti-neoplastic compounds. One mechanism by which tumor cells survive in the presence of chemotherapy is by increasing their apoptotic threshold. Since mitochondria are central players in drug-induced apoptosis, recent efforts to eradicate chemorefractory cells have focused on the identification of compounds that directly affect mitochondrial function. A number of reports indicate that mitochondria are direct targets for multiple classes of experimental compounds. A few clinically available anticancer agents like DNA damaging compounds and anti-microtubule agents are also reported to act directly on mitochondria. The purpose of this mini-review is to discuss recent advances in the interactions between anti-cancer agents and mitochondria, and highlight potential mitochondrial targets for novel chemotherapeutic interventions.

Ascorbic acid enhances arsenic trioxide-induced cytotoxicity in multiple myeloma cells.

Multiple myeloma (MM) is a clonal B-cell malignancy characterized by slow-growing plasma cells in the bone marrow (BM). Patients with MM typically respond to initial chemotherapies; however, essentially all progress to a chemoresistant state. Factors that contribute to the chemorefractory phenotype include modulation of free radical scavenging, increased expression of drug efflux pumps, and changes in gene expression that allow escape from apoptotic signaling. Recent data indicate that arsenic trioxide (As(2)O(3)) induces remission of refractory acute promyelocytic leukemia and apoptosis of cell lines overexpressing Bcl-2 family members; therefore, it was hypothesized that chemorefractory MM cells would be sensitive to As(2)O(3). As(2)O(3) induced apoptosis in 4 human MM cell lines: 8226/S, 8226/Dox40, U266, and U266/Bcl-x(L). The addition of interleukin-6 had no effect on cell death. Glutathione (GSH) has been implicated as an inhibitor of As(2)O(3)-induced cell death either through conjugating As(2)O(3) or by sequestering reactive oxygen induced by As(2)O(3). Consistent with this possibility, increasing GSH levels with N-acetylcysteine attenuated As(2)O(3) cytotoxicity. Decreases in GSH have been associated with ascorbic acid (AA) metabolism. Clinically relevant doses of AA decreased GSH levels and potentiated As(2)O(3)-mediated cell death of all 4 MM cell lines. Similar results were obtained in freshly isolated human MM cells. In contrast, normal BM cells displayed little sensitivity to As(2)O(3) alone or in combination with AA. Together, these data suggest that As(2)O(3) and AA may be effective antineoplastic agents in refractory MM and that AA might be a useful adjuvant in GSH-sensitive therapies. (Blood. 2001;98:805-813)

The androgen receptor (AR) amino-terminus imposes androgen-specific regulation of AR gene expression via an exonic enhancer.

Androgen and glucocorticoid receptor (AR, GR), two closely related members of the nuclear receptor superfamily, can recognize a similar cis-acting DNA sequence, or hormone response element (HRE). Despite this apparent commonality, these receptors regulate distinct target genes in vivo. The AR gene itself is regulated by AR but not GR in a variety of cell types, including osteoblast-like cells, as shown here. To understand this specificity, we first identified the DNA sequences responsible for androgen-mediated up-regulation of AR messenger RNA. A 6.5-kb region encompassing exon D, intron 4, and exon E of the AR gene contains four exonic HREs and exhibits cell type-specific, AR-mediated transcriptional enhancement when placed upstream of a heterologous promoter and reporter gene. A 350-bp fragment consisting of just exons D and E exhibits the same cell- and androgen-specificity as the 6.5-kb region, as well as the native AR gene. Consistent with a role for the exonic HREs, androgen regulation via this intragenic enhancer requires the HREs as well as a functional receptor DNA binding domain. A panel of AR/GR chimeric receptors was used to test which AR domains (amino-terminal, DNA binding or ligand binding) confer androgen-specific regulation of the 350-bp enhancer. Only chimeric receptors containing the amino-terminus of AR induced reporter gene activity from the AR gene enhancer. Further, a constitutively active AR consisting of only the AR amino-terminus and DNA binding domain (AA phi) retained the capacity to activate the internal responsive region, unlike a constitutively active chimera harboring the GR amino-terminus and AR DNA binding domain (GA phi). Thus, the AR amino terminus is the sole determinant for androgen-specific regulation of the AR gene internal enhancer. These findings support a model in which the amino termini of ARs bound to HREs within the AR gene interact with an exclusive auxiliary factor(s) to elicit androgen-specific regulation of AR messenger RNA. This is the first example of androgen-specific response in which the necessary and sufficient distinguishing capacity resides within the AR amino terminus.

Regulation of Bcl-xL: a little bit of this and a little bit of STAT.

The Bcl-2 family of proteins are key regulators of apoptosis. Bcl-xL, is an anti-apoptotic protein with a high degree of homology to Bcl-2; however, the signals that regulate Bcl-xL and Bcl-2 appear to be different. Levels of Bcl-xL, but not Bcl-2, are increased in response to various survival signals. Furthermore, an inverse correlation between the levels of Bcl-2 and Bcl-xL has been reported for a number of cancers. Although the precise molecules that control Bcl-xL activity are unclear, the STAT, Rel/NF-kappaB, and Ets transcription factor families have recently been reported to directly regulate the bcl-x gene. Activated Ras, integrin, vitronectin, and hepatocyte growth factor signaling cascades have also been linked to changes in Bcl-xL expression. Bcl-xL can also be affected by post-translational mechanisms. Here we review recent advances in identifying the signaling pathways and factors involved in regulation of the bcl-x gene.

Multiple androgen response elements and a Myc consensus site in the androgen receptor (AR) coding region are involved in androgen-mediated up-regulation of AR messenger RNA.

The androgen receptor (AR) gene is transcriptionally regulated by AR (autoregulation); however, the androgen response elements (AREs) required for this process have not been found in the AR promoter or in the 5'-flanking region. We previously showed that the AR cDNA contains AREs involved in AR mRNA autoregulation and that auto(up)regulation is reproduced in PC3 cells (a human prostate cancer cell line) expressing the human AR cDNA driven by a heterologous promoter. A 350-bp fragment of the AR cDNA contains the requisite AREs (ARE-1 and ARE-2) and, when linked upstream of a reporter gene, confers androgen inducibility in a cell-specific manner. Here we report that, although an AR cDNA harboring silent mutations of ARE-1 and ARE-2 produces a transcriptionally active AR, AR mRNA encoded by this mutant cDNA is not up-regulated in androgen-treated PC3 cells. Thus, ARE-1 and ARE-2 are essential for androgen-mediated up-regulation of AR mRNA in this model. Since ARE-1 and ARE-2 are located on separate exons (exons D and E) in the AR gene, we evaluated these AREs in their native context, a 6.5-kb AR genomic fragment. Androgen regulated the 6.5-kb AR genomic fragment and the 350-bp region of the AR cDNA at comparable levels, suggesting that sequences in exons D and E are likely to be involved in androgen-mediated up-regulation of the native AR gene. Furthermore, androgen regulated both responsive regions in U2OS cells, a human osteoblastic cell line that exhibits androgen-mediated up-regulation of native AR mRNA. DNAse I footprinting of the 350-bp region with recombinant AR (DNA- and ligand-binding domains) suggested the presence of additional AREs. Gel shift analyses and mutational studies showed that maximal androgen regulation and AR binding were dependent on the integrity of four AREs (ARE-1, ARE-1A, IVSARE, and ARE-2). While the presence of multiple, nonconsensus AREs is common among other androgen-regulated enhancers, the androgen-responsive region of the AR gene is unique because it contains exonic AREs. DNA binding studies with nuclear extracts were performed to determine whether non-AR transcription factors contribute to androgen regulation of the 350-bp region. These studies, in conjunction with mutational analysis and reporter gene assays with dominant negative Myc and Max expression vectors, showed that Myc and Max interaction with a Myc consensus site is required for androgen regulation of the 350-bp fragment. These results represent a novel interaction between AR and the Myc family of proteins and support a model of androgenic control of AR mRNA via AR and Myc family interaction with a unique internal androgen-responsive region harboring multiple exonic regulatory sequences.