Faculty Bibliography

BACKGROUND:Remedial surgery for patients with persistent or recurrent primary hyperparathyroidism (1 degrees HPT) remains a significant challenge. Cervical reexploration is technically difficult; reoperative neck anatomy is distorted by fibrosis and, as a result, remedial 1 degrees HPT patients carry an increased risk of injury to the recurrent (RLN) and superior laryngeal nerve(s) as well as to normal residual parathyroid tissue. Causative hyperfunctioning parathyroid tissue is also more frequently ectopic in the remedial setting and can thus be difficult to localize. METHODS:This report assimilates the current data underlying preoperative, intraoperative and postoperative remedial 1 degrees HPT management and presents an evidence-based algorithm for the management of remedial parathyroid disease. Recommendations are graded according to the quality of supporting data using the system initially developed by Sackett (Chest 95:2S-4S, 1989) and subsequently modified by Heinrich et al. (Ann Surg 243:154-168, 2006). RESULTS:Recent advances in preoperative localization and intraoperative adjuncts have lead to substantial improvements in outcomes after remedial surgery. Preoperative localization techniques, including sestamibi scintigraphy (MIBI), high resolution ultrasound (US), US-guided fine needle aspiration (FNA) and selective venous sampling (SVS), coupled with intraoperative adjuncts such as the rapid parathyroid hormone (PTH) assay have lead to reoperative cure rates as high as 96 percent. Nonetheless, management of remedial 1 degrees HPT varies significantly between surgeons and no formal recommendations standardizing the care of these patients have been published. CONCLUSIONS:Despite the significant challenges associated with remedial surgery for 1 degrees HPT, excellent outcomes can be reproducibly achieved when proper pre-, intra-, and postoperative management is employed.

Several different transcription factors, including estrogen receptor, progesterone receptor, and ETS family members, have been implicated in human breast cancer, indicating that transcription factor-induced alterations in gene expression underlie mammary cell transformation. ESE-1 is an epithelium-specific ETS transcription factor that contains two distinguishing domains, a serine- and aspartic acid-rich (SAR) domain and an AT hook domain. ESE-1 is abundantly expressed in human breast cancer and trans-activates epithelium-specific gene promoters in transient transfection assays. While it has been presumed that ETS factors transform mammary epithelial cells via their nuclear transcriptional functions, here we show (i) that ESE-1 protein is cytoplasmic in human breast cancer cells; (ii) that stably expressed green fluorescent protein-ESE-1 transforms MCF-12A human mammary epithelial cells; and (iii) that the ESE-1 SAR domain, acting in the cytoplasm, is necessary and sufficient to mediate this transformation. Deletion of transcriptional regulatory or nuclear localization domains does not impair ESE-1-mediated transformation, whereas fusing the simian virus 40 T-antigen nuclear localization signal to various ESE-1 constructs, including the SAR domain alone, inhibits their transforming capacity. Finally, we show that the nuclear localization of ESE-1 protein induces apoptosis in nontransformed mammary epithelial cells via a transcription-dependent mechanism. Together, our studies reveal two distinct ESE-1 functions, apoptosis and transformation, where the ESE-1 transcription activation domain contributes to apoptosis and the SAR domain mediates transformation via a novel nonnuclear, nontranscriptional mechanism. These studies not only describe a unique ETS factor transformation mechanism but also establish a new paradigm for cell transformation in general.

ESX is an epithelial-restricted member of a large family of transcription factors known as the Ets family. ESX expression has been shown to be correlated with Her2/neu proto-oncogene amplification in highly aggressive breast cancers and induced by Her2/neu in breast cell lines, but its role in tumorigenesis is unknown. Previously, we have shown that ESX enhances breast cell survival in colony-formation assays. In order to determine whether ESX can act as a transforming gene, we stably transfected MCF-12A human mammary epithelial cells with the ESX expression vector, pCGN2-HA-ESX. The MCF-12A cell line is immortalized, but nontransformed, and importantly, these cells fail to express endogenous ESX protein. We used pCGN2-HA-Ets-2 and pSVRas expression vectors as positive controls for transformation. Like HA-Ets-2 and V12-Ras, stable expression of ESX induced EGF-independent proliferation, serum-independent MAPK phosphorylation and growth in soft agar. Additionally, stable ESX expression conferred increased cell adhesion, motility and invasion in two-dimensional and transwell filter assays, and an epithelial to mesenchymal morphological transition. In three-dimensional cultures, parental and vector control (pCGN2) cells formed highly organized duct-like structures with evidence of cell polarity, ECM adhesion-dependent proliferation and cell survival, and lack of cellular invasion into surrounding matrix. Remarkably, the ESX stable cells formed solid, disorganized structures, with lack of cell polarity, loss of adhesion junctions and cytokeratin staining and loss of dependence on ECM adhesion for cell proliferation and survival. In addition, ESX cells invaded the surrounding matrix, indicative of a transformed and metastatic phenotype. Taken together, these data show that ESX expression alone confers a transformed and in vitro metastatic phenotype to otherwise normal MCF-12A cells.

The DNA in the macronucleus of the stichotrichs like Sterkiella nova (formerly Oxytricha nova) occurs in short molecules ranging from approximately 200 bp to approximately 20,000 bp. It has been estimated that there are approximately 24,500 different sized DNA molecules in the macronucleus. Single genes have been assigned to approximately 130 different sized macronuclear molecules in various stichotrichs (12 in Sterkiella nova) and hypotrichs, suggesting that each of the -24,500 different sized molecules encodes a different gene. To test this proposition we sequenced 31 macronuclear molecules picked randomly from a plasmid library of macronuclear DNA and analyzed them for potential gene content. The open reading frames (ORFs) in three short molecules encode amino acid (aa) sequences that do not match sequences in GenBank. They may or may not encode genes. Twenty-eight of the 31 molecules contain ORFs encoding aa sequences with significant matches to sequences in GenBank. Six molecules contain more than one ORF with a significant match to GenBank. These results indicate that almost all, if not all of the -24,500 different molecules encode one or more genes, yielding an estimate of -26,800 genes in the macronucleus of S. nova.