Development and characterization of a preclinical ovarian carcinoma model to investigate the mechanism of acquired resistance to trastuzumab.

Trastuzumab (Herceptin®) is a humanized monoclonal antibody designed to bind and inhibit the function of the human epidermal growth factor receptor 2 (HER2)/erbB2 receptor. Trastuzumab has demonstrated clinical activity in several types of HER2-overexpressing epithelial tumors, such as breast and metastatic gastric or gastroesophageal junction cancer. Relapse and therapeutic resistance, however, still occur in a subset of patients treated with regimens including trastuzumab, despite significant improvements in response rates, survival and quality of life. To investigate the potential mechanisms of acquired therapeutic resistance to trastuzumab, we developed a preclinical model of human ovarian cancer cells, SKOV-3 Herceptin-resistant (HR), and examined the corresponding changes in gene expression profiles. SKOV-3 HR cells were developed by in vivo serial passaging of parental trastuzumab-sensitive SKOV-3 cells. Following four rounds of serial transplantation of 'break-through' xenograft tumors under trastuzumab treatment, significant and reproducible differences in the effects of trastuzumab treatment between SKOV-3 HR and SKOV-3 cells in vivo and in vitro were revealed. SKOV-3 HR cells retained HER2 protein expression but were unaffected by the antiproliferative effects of trastuzumab. The trastuzumab binding affinity for SKOV-3 HR cells was diminished, despite these cells having more binding sites for trastuzumab. Microarray expression profiling (MEP) was performed to determine the genes involved in the resistance mechanism. Functional analysis revealed the differential expression of genes potentially involved in angiogenesis, metastasis, differentiation and proliferation, such as mucin1 (MUC1). Immunohistochemical staining of SKOV-3 HR cells demonstrated a marked overexpression of MUC1. Based on these data, we hypothesize that the overexpression of MUC1 may hinder trastuzumab binding to HER2 receptors, abrogating the antitumor effects of trastuzumab and thus could contribute to resistance to therapy. Moreover, the resultant MEP preclinical gene signature in this preclinical model system may provide the basis for further investigation of potential clinical mechanisms of resistance to trastuzumab.

A selective phosphatase of regenerating liver phosphatase inhibitor suppresses tumor cell anchorage-independent growth by a novel mechanism involving p130Cas cleavage.

The phosphatase of regenerating liver (PRL) family, a unique class of oncogenic phosphatases, consists of three members: PRL-1, PRL-2, and PRL-3. Aberrant overexpression of PRL-3 has been found in multiple solid tumor types. Ectopic expression of PRLs in cells induces transformation, increases mobility and invasiveness, and forms experimental metastases in mice. We have now shown that small interfering RNA-mediated depletion of PRL expression in cancer cells results in the down-regulation of p130Cas phosphorylation and expression and prevents tumor cell anchorage-independent growth in soft agar. We have also identified a small molecule, 7-amino-2-phenyl-5H-thieno[3,2-c]pyridin-4-one (thienopyridone), which potently and selectively inhibits all three PRLs but not other phosphatases in vitro. The thienopyridone showed significant inhibition of tumor cell anchorage-independent growth in soft agar, induction of the p130Cas cleavage, and anoikis, a type of apoptosis that can be induced by anticancer agents via disruption of cell-matrix interaction. Unlike etoposide, thienopyridone-induced p130Cas cleavage and apoptosis were not associated with increased levels of p53 and phospho-p53 (Ser(15)), a hallmark of genotoxic drug-induced p53 pathway activation. This is the first report of a potent selective PRL inhibitor that suppresses tumor cell three-dimensional growth by a novel mechanism involving p130Cas cleavage. This study reveals a new insight into the role of PRL-3 in priming tumor progression and shows that PRL may represent an attractive target for therapeutic intervention in cancer.

Discovery of promiscuous HLA-II-restricted T cell epitopes with TEPITOPE.

TEPITOPE is a prediction model that has been successfully applied to the in silico identification of T cell epitopes in the context of oncology, allergy, infectious diseases, and autoimmune diseases. Like most epitope prediction models, TEPITOPE's underlying algorithm is based on the prediction of HLA-II peptide binding, which constitutes a major bottleneck in the natural selection of epitopes. An important step in the design of subunit vaccines is the identification of promiscuous HLA-II ligands in sets of disease-specific gene products. TEPITOPE's user interface enables the systematic prediction of promiscuous peptide ligands for a broad range of HLA-binding specificity. We show how to apply the TEPITOPE prediction model to identify T cell epitopes, and provide both a road map and examples of its successful application.

Unique clinical and pathological features in HLA-DRB1*0401-restricted MBP 111-129-specific humanized TCR transgenic mice.

Amino acid residues 111-129 represent an immunodominant epitope of myelin basic protein (MBP) in humans with human leukocyte antigen (HLA)-DRB1*0401 allele(s). The MBP 111-129-specific T cell clone MS2-3C8 was repeatedly isolated from a patient with multiple sclerosis (MS), suggesting an involvement of MS2-3C8 T cells in the pathogenesis. To address the pathogenic potential of the MS2-3C8 T cell clone, we generated transgenic (Tg) mice expressing its T cell receptor and restriction element, HLA-DRB1*0401, to examine the pathogenic characteristics of MS2-3C8 Tg T cells by adoptive transfer into HLA-DRB1*0401 Tg mice. In addition to the ascending paralysis typical of experimental autoimmune encephalomyelitis, mice displayed dysphagia due to restriction in jaw and tongue movements and abnormal gait. In accordance with the clinical phenotype, infiltrates of MS2-3C8 Tg T cells and inflammatory lesions were predominantly located in the brainstem and the cranial nerve roots in addition to the spinal cord and spinal nerve roots. Together, these data suggest a pathogenic role of MBP-specific T cells in inflammatory demyelination within the brainstem and cranial nerve roots during the progression of MS. This notion may help to explain the clinical and pathological heterogeneity of MS.

The use of bioinformatics for identifying class II-restricted T-cell epitopes.

An important step in the design of subunit vaccines is the identification of promiscuous T helper cell epitopes in sets of disease-specific gene products. Most of the epitope prediction models are based on HLA-II peptide binding, which constitutes a major bottleneck in the natural selection of epitopes. Here we describe a computer model, TEPITOPE, that enables the systematic prediction of promiscuous peptide ligands for a broad range of HLA binding specificity. We show how to apply the TEPITOPE prediction model to identify T-cell epitopes, and provide examples of its successful application in the context of oncology, allergy, and infectious and autoimmune diseases.

Cascades of transcriptional induction during dendritic cell maturation revealed by genome-wide expression analysis.

Dendritic cells (DC) are central regulators of immunity. Signal-induced maturation of DCs is assumed to be the starting point for specific immune responses. To further understand this process, we analyzed the alteration of transcript profiles along the time course of CD40 ligand-induced maturation of human myeloid DCs by Affymetrix GeneChip microarrays covering >6800 genes. Besides rediscovery of genes already described as associated with DC maturation proving reliability of the methods used, we identified clusterin as novel maturation marker. Looking across the time course, we observed synchronized kinetics of distinct functional groups of molecules whose temporal coregulation underscores known cellular events during dendritic cell maturation. For example, an early-peaking wave of inflammatory chemokines was followed by a sustained increase of constitutive chemokines and accompanied by slow but continuous induction of survival proteins. After an immediate but transient induction of cytokine-responsive transcripts, there was an increased expression of a group of genes involved in not only the regulation of cytokine effects, but also of transcription in general. Our results demonstrate that microarray studies along time courses combined with real-time PCR not only discover new marker molecules with functional implications, but also dissect the molecular kinetics of biological processes identifying complex pathways of regulation.

Computational dissection of tissue contamination for identification of colon cancer-specific expression profiles.

Microarray profiles of bulk tumor tissues reflect gene expression corresponding to malignant cells as well as to many different types of contaminating normal cells. In this report, we assess the feasibility of querying baseline multitissue transcriptome databases to dissect disease-specific genes. Using colon cancer as a model tumor, we show that the application of Boolean operators (AND, OR, BUTNOT) for database searches leads to genes with expression patterns of interest. The BUTNOT operator for example allows the assignment of "expression signatures" to normal tissue specimens. These expression signatures were then used to computationally identify contaminating cells within conventionally dissected tissue specimens. The combination of several logic operators together with an expression database based on multiple human tissue specimens can resolve the problem of tissue contamination, revealing novel cancer-specific gene expression. Several markers, previously not known to be colon cancer associated, are provided.