The significance of epithelial differentiation in mixed mesodermal tumors of the uterus. A clinicopathologic and immunohistochemical study

Mixed mesodermal tumors and carcinosarcomas of the uterus are classified as sarcomas. However, in other sites, malignant biphasic tumors may be classified as carcinomas, mesotheliomas, or sarcomas. In order to clarify their behavior and patterns of differentiation, we performed a clinicopathologic and immunohistochemical study of 22 cases aimed at analyzing the pattern of spread and histologic appearance of the metastasis, as well as the distribution of intermediate filaments in the primary tumor and the metastasis. Four monoclonal antibodies (Mabs) were used to detect epithelial lineage, three that recognize keratin (AE1/AE3, CAM5.2, MAK6) and one that recognizes epithelial membrane antigen (EMA). A Mab against vimentin was also used. Metastases involved the omentum, pelvic peritoneum, ovaries, fallopian tubes, pelvic or para-aortic lymph nodes, liver parenchyma, and tonsil. These metastases were composed of carcinoma only. Lymphatic/vascular invasion was identified in 11 cases; it consisted exclusively of carcinoma. In all 12 cases evaluated immunohistochemically, keratin and EMA were identified in the majority of the cells in the epithelial component and in a more focal distribution in the spindle cell component in 11 (92%). Vimentin was detected in the majority of spindle cells in nine cases (75%) and in a more focal distribution in the epithelial component in six cases (50%). In the spindle cell component, keratin and EMA were present in widely scattered individual spindle-shaped and rounded cells, within solid clusters of rounded cells, and in nests of cells with small lumens. The distribution of keratin, EMA, and vimentin in the metastases (carcinoma in all instances) was similar to the epithelial component in the primary tumor. Our findings indicate that the epithelial component of these tumors invades lymphatic/vascular spaces and metastasizes, whereas the spindle cell component has limited metastatic potential, if any. Since the behavior of these neoplasms is dictated by the epithelial element, we believe that mixed mesodermal tumors of the uterus should be classified as carcinomas rather than sarcomas.     

Competitive control of the self-renewing T cell repertoire

We develop a mathematical model for the self-renewing part of the T cell repertoire. Assuming that self-renewing T cells have to be stimulated by immunogenic MHC-peptide complexes presented on the surfaces of antigen-presenting cells, we derive a model of T cell growth in which competition for MHC-peptide complexes limits T cell clone sizes and regulates the total number of self-renewing T cells in the animal. We show that for a sufficient diversity and/or degree of cross-reactivity, the total T cell number hardly depends upon the diversity of the T cell repertoire or the diversity of the set of presented peptides. Conversely, for repertoires of lower diversity and/or cross-reactivity, steady-state total T cell numbers may be limited by the diversity of the T cells. This provides a possible explanation for the limited repertoire expansion in some, but not all, mouse T cell re-constitution experiments. We suggest that the competitive interactions described by our model underlie the normal T cells numbers observed in transgenic mice, germ-free mice and various knockout mice.      

Localization of a gene for otosclerosis to chromosome 15q25-q26

Among white adults otosclerosis is the single most common cause of hearing impairment. Although the genetics of this disease are controversial, the majority of studies indicate autosomal dominant inheritance with reduced penetrance. We studied a large multi- generational family in which otosclerosis has been inherited in an autosomal dominant pattern. Five of16 affected persons have surgically confirmed otosclerosis; the remaining nine have a conductive hearing loss but have not undergone corrective surgery. To locate the disease- causing gene we completed genetic linkage analysis using short tandem repeat polymorphisms (STRPs) distributed over the entire genome. Multipoint linkage analysis showed that only one genomic region, on chromosome 15q, generated a lod score >2.0. Additional STRPs were typed in this area, resulting in a lod score of 3.4. STRPs FES (centromeric) and D15S657 (telomeric) flank the 14. 5 cM region that contains an otosclerosis gene.      

IMPT1, an imprinted gene similar to polyspecific transporter and multi- drug resistance genes

Human chromosome 11p15.5 and distal mouse chromosome 7 include a megabase-scale chromosomal domain with multiple genes subject to parental imprinting. Here we describe mouse and human versions of a novel imprinted gene, IMPT1 , which lies between IPL and p57 KIP2 and which encodes a predicted multi-membrane-spanning protein similar to bacterial and eukaryotic polyspecific metabolite transporter and multi- drug resistance pumps. Mouse Impt1 and human IMPT1 mRNAs are highly expressed in tissues with metabolite transport functions, including liver, kidney, intestine, extra-embryonic membranes and placenta, and there is strongly preferential expression of the maternal allele in various mouse tissues at fetal stages. In post-natal tissues there is persistent expression, but the allelic bias attenuates. An allelic expression bias is also observed in human fetal and post-natal tissues, but there is significant interindividual variation and rare somatic allele switching. The fact that Impt1 is relatively repressed on the paternal allele, together with data from other imprinted genes, allows a statistical conclusion that the primary effect of human chromosome 11p15.5/mouse distal chromosome 7 imprinting is domain-wide relative repression of genes on the paternal homolog. Dosage regulation of the metabolite transporter gene(s) by imprinting might regulate placental and fetal growth.      

Comparison of the hybrid capture tube test and PCR for detection of human papillomavirus DNA in cervical specimens.

The strong association of human papillomavirus (HPV) and cervical cancer makes it important to study HPV detection methods that may play a role in cervical cancer screening. We compared two DNA methods that are commonly used for HPV research in the United States: the MY09/MY11 L1 consensus primer PCR-based test and the first-generation Hybrid Capture tube method (HCT). Laboratory assays by each method were performed with 596 cervicovaginal specimens collected from participants in a large cohort study conducted in Portland, Oreg. Included were 499 specimens from women whose cytology was normal and 97 specimens from women with squamous intraepithelial lesions (SILs). The overall HPV DNA positivity for known types was 22.5% by PCR compared to 13.6% by HCT. When the analysis was restricted to the 14 HPV types detectable by both methods, the sensitivity of HCT, with PCR used as the standard for HPV status, was higher for specimens from women with concurrent SILs (81.0%) than for specimens from women with normal cytology (46.7%). Among specimens testing positive by both methods, 97.2% of the time the two methods agreed on whether specimens were positive for cancer-associated HPV types. Both of these HPV test methods provide information that supplements the information provided by the Pap smear. The PCR method has higher analytic sensitivity than HCT in detecting HPV, but HCT may be helpful in identifying women with concurrent SILs.