Potential for modulation of the fas apoptotic pathway by epidermal growth factor in sarcomas.

One important mechanism by which cancer cells parasitize their host is by escaping apoptosis. Thus, selectively facilitating apoptosis is a therapeutic mechanism by which oncotherapy may prove highly advantageous. One major apoptotic pathway is mediated by Fas ligand (FasL). The death-inducing signaling Ccmplex (DISC) and subsequent death-domain aggregations are created when FasL is bound by its receptor thereby enabling programmed cell death. Conceptually, if a better understanding of the Fas pathway can be garnered, an oncoselective prodeath therapeutic approach can be tailored. Herein, we propose that EGF and CTGF play essential roles in the regulation of the Fas apoptotic pathway in sarcomas. Tumor and in vitro data suggest viable cells counter the prodeath signal induced by FasL by activating EGF, which in turn induces prosurvival CTGF. The prosurvival attributes of CTGF ultimately predominate over the death-inducing FasL. Cells destined for elimination inhibit this prosurvival response via a presently undefined pathway. This scenario represents a novel role for EGF and CTGF as regulators of the Fas pathway in sarcomas.

Potential for functional redundancy in EGF and TGFalpha signaling in desmoid cells: a cDNA microarray analysis.

Genes that replace or duplicate the function of other genes are considered functionally redundant. In this cDNA microarray study, using an Agilent microarray platform and GeneSifter analysis software, we evaluated (1) the degree of downstream transcriptional redundancy and (2) the level of genetic uniqueness apparent in desmoid tumor cells stimulated in vitro for 3 h or for 24 h with 100 ng/ml of exogenous recombinant human EGF (rhEGF) or with recombinant human transforming growth factor alpha (rhTGFalpha). Our intent was to identify genes costimulated, or genes unique to, desmoid cells stimulated in vitro with rhEGF and rhTGFalpha. This experimental approach demonstrated a 55% transcriptional redundancy in the number of desmoid genes significantly upregulated or downregulated following 3 h of stimulation with rhEGF or with rhTGFalpha, and a 65% transcriptional redundancy following 24 h of growth factor stimulation. Approximately 150 genes costimulated by rhEGF and rhTGFalpha were identified. This study suggests that EGF and TGFalpha retain some level of functional redundancy, possibly resulting from their divergence from a common ancestral gene.

FAP-associated desmoid invasiveness correlates with in vitro resistance to doxorubicin.

Desmoid tumors are locally invasive myofibroblastic lesions that arise predominantly in the abdominal wall or shoulder girdle and are prone to aggressive local recurrences without metastases. We hypothesized the intrinsic invasiveness and drug resistance displayed by cells derived from a familial adenomatous polyposis (FAP)-associated desmoid tumor would surpass the response shown by cells derived from sporadic desmoid tumors. In vitro cell motility and expression of motility-associated genes were quantified using Boyden Chambers and Enzyme-Linked ImmunoSorbent Assays, respectively. Doxorubicin resistance was quantified by Trypan Blue dye exclusion. cDNA microarrays identified genes responsive to doxorubicin. FAP-associated tumor cells were significantly more invasive and refractory to doxorubicin than were cells extracted from sporadic tumors. Pro-MMP1 protein predominated over MMP3 in FAP-associated cell culture supernatants, while MMP3 was the dominant antigen in sporadic tumor cell supernatants. Three genes associated with apoptosis were identified by microarray, two prosurvival genes overexpressed in FAP-associated cell cultures (NTN1, TNFRSF10C) and one proapoptosis gene overexpressed in sporadic tumor cell cultures (FOXL2).

Desmoid cell motility is induced in vitro by rhEGF.

Desmoid tumors are benign but locally invasive myofibroblastic lesions that arise predominantly in the abdominal wall or shoulder and are prone to aggressive local recurrences. A perceived association between desmoid activity and the expression of growth factors during pregnancy or following trauma suggests a cause-and-effect relationship between growth factor stimulation and desmoid invasiveness. We used Boyden Chambers to quantify cell motility in order to determine the effect of growth factor stimulation on desmoid cell migration. Desmoid cell cultures were treated under serum-free conditions with epidermal growth factor (rhEGF) or transforming growth factor alpha (rhTGFalpha). Additional cell cultures were pretreated under serum-free conditions with the EGF receptor (EGFR) inhibitor AG1478, alone or in combination with the TGFbeta1 receptor inhibitor SB431542, and then stimulated with growth factor prior to being assayed for cell motility. The experiments demonstrated a direct dose-dependent relationship between rhEGF stimulation and desmoid motility. In contrast, rhTGFalpha was less effective at inducing cell migration. rhEGF-induced cell migration could be diminished, but not reduced to control levels, by inhibiting EGFR. When EGF and TGFbeta1 receptors were inhibited simultaneously, the level of rhEGF-induced cell migration was reduced significantly beyond the level of cell migration generated by inhibition of EGFR alone.

Fas death pathway in sarcomas correlates with epidermal growth factor transcription.

Modulation of apoptosis may influence sarcoma pathogenesis and/or aggressiveness. The Fas death pathway, mediated by FasL or TGFbeta, is one of two apoptotic pathways. Recent studies report that EGF can modulate TGFbeta and/or FasL expression/activity; thus, EGF has the potential to influence activation of the Fas pathway. EGF is not always detectable in mesenchymal tumors; therefore, we hypothesized EGF would define which Fas ligand predominates. We assayed 57 surgically removed human sarcomas for 10 genes involved in the Fas pathway. Skeletal muscle biopsies from eight patients served as controls. Sample transcripts were detected by real-time RT-PCR. We attempted to identify relevant predictor variables. The 57 sarcomas were segregated into two categories defined by EGF mRNA content: (1) 23 tumors with EGF concentrations that approximated muscle EGF transcript levels (high-EGF tumors); and (2) 34 tumors that either lacked EGF mRNA, or whose mRNA levels were very low and frequently undetected by PCR (low-EGF tumors). TGFbeta1 expression best predicted Fas transcript concentrations in the 34 low-EGF sarcomas, while FasL predicted Fas mRNA levels in the remaining 23 high-EGF sarcomas. The results suggest ligand activity in the Fas death pathway correlates with EGF transcription in sarcomas.

Oncogene coexpression in mesenchymal neoplasia correlates with EGF transcription.

Epidermal growth factor (EGF) is a potent mitogenic factor for cells of mesodermal and ectodermal origin, and its over-expression is associated with a variety of cancers. We asked whether oncogene coexpression occurs in mesenchymal neoplasms, if coexpression correlates with EGF transcription, and whether coexpression can be attributed to the EGF-induced overexpression of oncogenes. We quantified the mRNA concentrations of EGF and 14 oncogenes in 42 primary sarcomas, 31 benign tumors, and 10 skeletal muscle controls, and compared mRNA concentrations and gene pair correlations in EGF positive (EGF+) tumors to transcript concentrations and correlations in EGF negative (EGF-) tumors. Transcripts were detected by real time reverse transcription-polymerase chain reaction. Pearson's correlation coefficients identified gene associations, and gene synchrony associated with EGF expression was evaluated using chi square. Transcript concentrations in tumors were compared graphically and with t tests. Gene correlations predominated in EGF+ benign tumors and in EGF- primary sarcomas. The dichotomy in oncogene coexpression evident in benign and malignant tumors could not be attributed to statistical differences in mRNA content between EGF+ and EGF- tumors. EGF may enhance, or may indicate the presence of, oncogene coexpression in benign mesenchymal lesions, but counters gene synchronization in sarcomas.

Doxorubicin induces cell senescence preferentially over apoptosis in the FU-SY-1 synovial sarcoma cell line.

Surgical resection coupled with adjuvant radiotherapy and/or doxorubicin based chemotherapy are the mainstays of synovial sarcoma (SS) treatment. Although effective as a SS adjuvant, the proposed mechanism of action of doxorubicin remains controversial. Current opinion supports DNA damage-induced apoptosis. This in vitro study used cDNA gene expression profiling to investigate whether apoptosis, alone or in combination with cell senescence, is induced by doxorubicin in SS cells. Cell cultures of the FU-SY-1 SS, the pleomorphic SW982 sarcoma, and a primary dermal fibroblast (NHDF), were exposed to 500 nM doxorubicin, and then processed for cDNA microarray analysis. The one class response option of SAM (Significance Analysis of Microarrays) was used to test for significant overexpression of 15 apoptosis-related genes and nine senescence-related genes. Drug-induced cell senescence was quantified by measuring beta-galactosidase activity. None of 15 apoptosis-related genes and only two of nine senescence-related genes were identified by SAM as significantly overexpressed in doxorubicin-treated cultures. Drug-induced senescence as reflected by beta-galactosidase activity was significantly increased (p < 0.05) only in FU-SY-1 SS cultures. Apoptosis does not appear to be a major determinant of doxorubicin-induced mortality in FU-SY-1 SS or NHDF cultures, but may impact SW982 cells via the overexpression of BAX relative to Bcl-2. Doxorubicin-induced cell senescence was prominent in FU-SY-1 SS cultures, but negligible in SW982 and NHDF cultures. Likely, both apoptosis and cell senescence contribute to doxorubicin-induced cell death in this synovial sarcoma cell line.

G3139 antisense oligonucleotide directed against antiapoptotic Bcl-2 enhances doxorubicin cytotoxicity in the FU-SY-1 synovial sarcoma cell line.

Synovial sarcoma (SS) is a highly aggressive, periarticular soft tissue sarcoma that causes death in more than half of affected children, adolescents, and young adults. Five- and 10-year survival rates are as low as 36 and 20%, respectively. Bcl-2, a negative regulator of apoptosis, is overexpressed in up to 90% of SS. Increased Bcl-2 expression not only leads to the development of cancer, but also to resistance of many anticancer chemotherapeutic agents. We hypothesized reducing Bcl-2 expression in SS should enhance doxorubicin cytotoxicity. Cell cultures representing two human sarcomas (FU-SY-1 SS and the pleomorphic SW982) and a primary human dermal fibroblast comparator (NHDF) were exposed in vitro to doxorubicin, or to doxorubicin preceded by Bcl-2 (G3139) antisense oligonucleotides, and assayed for cell survival, apoptosis, and modulations in Bcl-2 and Bcl-xL mRNA and protein content. SW982 sarcoma cells proved most susceptible to doxorubicin, while NHDF mesenchymal cells were least sensitive to doxorubicin. Treatment of FU-SY-1 SS with G3139 reduced Bcl-2 mRNA and protein levels, which enhanced doxorubicin-induced cell killing. There was a concurrent reduction in Bcl-xL mRNA following G3139 application in FU-SY-1 and NHDF cultures, but not in SW982. G3139 anti-Bcl-2 intervention sensitized the FU-SY-1 SS to doxorubicin, due to increased apoptosis. G3139 intervention was ineffective in the two non-SS cell lines.

Diagnosis and management of synovial sarcoma.

Synovial sarcoma is a unique tumor with substantial promise for biologically targeted therapy. Although it demonstrates moderate chemosensitivity, with approximately 50% response rates to ifosfamide- and doxorubicin-containing regimens, it has a diagnostic translocation and a potentially informative chimeric protein product. Although surgical management remains the cornerstone to effect local control, therapeutic advancements are unlikely to occur by continuing to include advanced cases of synovial sarcomas in trials with other soft tissue sarcomas. Rather, attention should be turned toward prospective molecular targets and development of novel agents to exploit them. Research should be directed at understanding the fusion protein of the X,18 translocation and further validating the role of overexpressed proteins in synovial sarcoma. Meanwhile, carefully designed clinical trials of these agents, with translational correlates, will provide in vivo data to complement the preclinical experience.

Transit tumor retrieval preserves RNA fidelity and obviates snap-freezing.

UNLABELLED: Genetic expression profiling is enabling investigators to discover new diagnostic and possibly therapeutic pathways in sarcoma biology. To draw substantial conclusions from these molecular analyses, adequate tissue samples must be accrued. Beyond cohort size, the most variable and limiting aspect of doing gene expression analyses on fresh human tissue is the preservation of labile ribonucleic acids extracted from clinical specimens. We have developed a novel retrieval protocol that is readily amenable to the clinical constraints placed on surgeons and pathologists that minimizes variables that can corrupt ribonucleic acid fidelity. We evaluate critically genomic message integrity of mesenchymal tumors derived from transcontinental inter-institutional collaboration. Intact total ribonucleic acid was isolated and assessed for quality and quantity. Ribosomal RNA integrity was quantified using a bioanalyzer. Ribonucleic acid from 42 mesenchymal tumors was isolated and quantified, with selected samples amplified. The mean ribosomal ratios for collaborative institutions ranged from 1.0 to 1.18. Samples remained at 4 degrees C before processing from 1 to 17 days. Tumors stabilized using this protocol retained total ribonucleic acid integrity suitable for amplification and genomic expression analysis regardless of the institutional source or preprocessing duration, enabling a potential consortium of investigators to collaborate in the expression profiling of sarcomas. LEVEL OF EVIDENCE: Diagnostic study, Level III-3 (no consistently applied gold standard). See the Guidelines for Authors for a complete description of levels of evidence.

Discriminate gene lists derived from cDNA microarray profiles of limited samples permit distinguishing mesenchymal neoplasia ex vivo.

BACKGROUND: Mesenchymal neoplasia comprises a heterogeneous group of tumors with over 200 benign neoplasms and 100 sarcomas. Currently, tumors are classified using histologic and immunocytologic characteristics, with diagnostic error rates reported as high as 40% of cases. As a feasibility study, our goal was to generate a preliminary discriminatory gene list for selected mesenchymal tumors, including sarcomas. This technique may enable an eventual molecular classification schema based on expression profiles that can complement current clinical and pathologic diagnostic procedures in mesenchymal tumors. METHODS: cDNA microarray analyses were preformed on connective tissue tumors obtained at time of surgical resection or biopsy. Messenger RNA (mRNA) from four general tumor classes was competitively hybridized against a human dermal fibroblast cell line comparator and the resulting gene expression profiles processed by ANOVA and linear discriminate analysis. RESULTS: The tissue classification involved 18 patients with malignant peripheral nerve sheath tumors, giant cell containing tumors, benign spindle cell lesions, or Ewing's family of tumors. Lymph nodes from two patients served comparative purposes. Twenty-five differentially regulated genes considered most variable among the five tissue classes were identified. The tissues were segregated into five classes by linear discriminate analysis. CONCLUSIONS: Linear discriminate analysis of cDNA gene expression profiles partitioned mesenchymal tumor classes, even when constrained by limited sample sizes.

Validation of cDNA microarray analysis to distinguish tumor type ex vivo.

Tissue biopsy alone is diagnostic in less than 10% of cases of metastatic bone disease with an unknown primary tumor. Accordingly, to delineate the neoplastic process in cases of metastatic bone disease, where the primary tumor cannot be determined with conventional modalities, new tools are attractive. Complementary deoxyribonucleic acid cDNA microarrays may be such a tool. Verification and validation of this technology for analysis of specimens ex vivo is mandatory if it is to be used to analyze the complexity of heterogenous biologic tissues. To convey the issues involved in the clinical application of cDNA microarray analysis of tumor specimens ex vivo, a set of experiments analyzing surgically resected human specimens is presented. The specific aims of the analysis were to assess cDNA microarray reproducibility of serial tissue preparations of tumor samples and to confirm that heterogeneity within a given clinical tumor specimen does not preclude this technique as a tool to discern tumor types. Complementary deoxyribonucleic microarray analysis, as it applies to heterogeneous clinical tumor samples in limited numbers, based on preliminary data, seems to be an appropriate but still experimental method for distinguishing tumor types.