Faculty Bibliography

Tuberous sclerosis complex (TSC) is a genetically heterogeneous disease caused by mutations of TSC1 or TSC2 genes. It involves multiple organ systems resulting in mild to lethal hamartoma formation due to gene mutation in the germ line and loss of heterozygosity (LOH) in somatic cells. Hamartin (TSC1) and tuberin (TSC2) are expressed broadly. However, little is known about tissue susceptibility to hamartomas when equal or similar amounts of TSC gene expression are present. In this study, we present a 19-week gestational age fetus with pathological features of TSC, which was confirmed by finding LOH of TSC2 in a cardiac rhabdomyoma. Developmental expression of hamartin and tuberin in the TSC fetus, an age-matched non-TSC fetus, and a 26-week gestational age non-TSC fetus were analyzed by immunohistochemistry. We found that in addition to the differential expression of the TSC genes in some normal tissues compared with that in the TSC-affected fetus, the cellular localization and distribution of hamartin and tuberin were dramatically different in different tissues. In general, hamartin and tuberin are mainly expressed in epithelial cells, myocytes, and neural tissues. By comparing the incidence of the hamartomas in early childhood and gene expression in tissues, it appears that tissues with co-expression of hamartin and tuberin are prone to a higher incidence of hamartomas than those expressing only one protein, or two proteins but in different patterns of cellular localization

Gastric fundic gland polyps (FGPs) are considered hamartomas, and various gastrointestinal hamartomas are associated with tuberous sclerosis complex (TSC). The aim of this study was to investigate a possible link between TSC proteins (hamartin and tuberin) and sporadic FGPs. We examined 33 sporadic FGPs and 26 biopsies of normal fundic mucosa by immunohistochemistry. Nuclear immunoreactivity for tuberin was dramatically reduced or lost in most sporadic FGPs, and tuberin unexpectedly accumulated in the cytoplasm in oxyntic glands. About 18% (6/33) of FGPs were immunopositive in an average of 1.7% of oxyntic cell nuclei, compared with 77% (20/26) of controls in an average of 24.4% of oxyntic cell nuclei (P <.01). No change in hamartin was noted. We further examined the tuberin-associated proteins glucocorticoid receptor (GCR) and p27. Nuclear immunoreactivity for GCR was lost in most sporadic FGPs, but p27 distribution was normal. Sporadic FGPs had a low frequency of staining for Ki-67 except for some cells from cystic components, which is consistent with their slow growth. Our results are consistent with the hypothesis that tuberin may play an important role in pathogenesis of sporadic FGPs. First, an altered cellular localization of tuberin may lead to the deregulation of cell proliferation by interrupting its interaction with hamartin. Second, altered cellular localization of tuberin may preclude its negative regulation of gene transcription mediated by GCR

INTRODUCTION: The neuronal cell death that occurs after ischemia-induced cerebral infarction (stroke) contains elements of apoptosis and necrosis, an intermediary form of the two, and a distinct excitotoxic process. We previously developed a photochemical model of stroke in the rat. We have now adapted this model for use in the mouse. The present manuscript describes the mouse model. METHODS: Minimal beam intensity (0.1 W/cm(2)) cold white light (8 min exposure) was used to evoke discrete infarcts in the parietal lobes of 11 mice sensitized by the administration of fresh Rose Bengal (10 mg/kg by rapid iv infusion). RESULTS: At 2 h, five out of five mice and at 6 h, six out of six mice demonstrated light microscopic histologic features like those in the rat model. These included a superior ischemic zone with shrunken and pyknotic nuclei, a middle transition zone of edematous vacuolated neuropil but normal neurons with open chromatin and retained Nissl granules, and an inferior zone with normal neurons. There was widespread nuclear terminal deoxynucleotidyl-transferase-mediated dUTP nick end labeling (TUNEL) in the superior infarct zone in 11/11 mice. However, in the edematous vacuolated transition zone, 11/11 mice had TUNEL positive and negative nuclei randomly mixed. Light microscopic analysis of that same transition zone showed no pyknosis or chromatin bodies in the TUNEL positive or negative cells. DISCUSSION: In mice, photoactivation of Rose Bengal evoked similar infarct and transition zone patterns found previously in rats, with TUNEL evidence of apoptotic and nonapoptotic events. Thus, it will be possible to use this model for further quantitative study of apoptotic and excitotoxic events in wild and transgenic mice

An objective method for the analysis of tissue section is described that uses the chemical composition of the tissue, rather than cell morphology, as an indicator for the state of health of the cells in the tissue. The chemical composition of cells and tissue, and small variations therein, are determined by an objective, quantitative spectral measurement carried out in the infrared spectral region. This method does not utilize any stains or chemical treatment of the sample, but uses an inherent optical property of all materials. The spectral information is converted to false color images by unsupervised mathematical methods. The false color maps reveal the same anatomical features of the tissue that can be confirmed using a variety of common histopathological procedures, and may be used to differentiate between normal and diseased areas of the tissue

The IR microspectra for individual normal and cancerous cells are reported at a spatial resolution that permits a distinction between the nuclear and cytoplasmic regions. The observed spectra reveal large differences in the spectral contributions of RNA, DNA, and phospholipids: metabolically inactive cells show spectral patterns of proteins only, whereas actively dividing cells also show spectral signatures of nucleic acids and phospholipids. These spectral variations are independent of the state of health of a cell

BACKGROUND:The diagnosis of melanoma can be difficult because of shared cytomorphology with other malignant neoplasms. The most commonly used melanocytic markers, anti-S-100 protein and HMB-45 antigen, have limited specificity and sensitivity, respectively. Microphthalmia transcription factor (Mitf) is a nuclear transcription factor critical for the development and survival of melanocytes and has been shown as a sensitive and specific marker for melanoma in histologic specimens. METHODS:To evaluate the efficacy of Mitf as a marker for melanoma in cytologic preparations, 81 cell blocks from 44 patients with melanoma and 37 patients with nonmelanoma malignancies (29 patients with carcinoma, 4 patients with mesotheliomas, 2 patients with lymphoma, and 2 patients with islet cell tumors) were stained with monoclonal antibodies against Mitf (clone D5), S-100 protein, and HMB-45 antigen. The staining was evaluated blindly by three independent observers. The presence of nuclear staining for Mitf and cytoplasmic staining for S-100 protein or HMB-45 antigen in > 10% of tumor cells was considered positive staining for each antigen. RESULTS:Forty-four melanomas (100%), including all 3 spindle-cell melanomas, were positive for Mitf. All nonmelanoma neoplasms were negative with only one exception: One mammary carcinoma showed rare (< 10%), weak nuclear staining with Mitf. The sensitivity and specificity of Mitf as a marker for melanoma were both 100%, whereas the sensitivity of HMB-45 antigen was 90.4%, and the specificity of S-100 protein was 70.3%. CONCLUSIONS:Mitf is a sensitive and specific marker for malignant melanoma, including the spindle-cell variant, in cytologic specimens and may be superior to the current standard melanocytic markers, S-100 protein and HMB-45 antigen.

HYPOTHESIS: Increased cell motility is a hallmark of cancer cells. Proteins involved in cell motility may be used as molecular markers to characterize the malignant potential of tumors. METHODS: Molecular biology and immunohistochemistry techniques were used to investigate the expression of a selected panel of motility-related proteins (Rho A, Rac 2, Cdc42, PI3K, 2E4, and Arp2) in normal, premalignant, and squamous cell cancer cell lines of human head and neck origin. To assess the clinical potential of these proteins as molecular markers for cancer, immunohistochemistry was performed on paraffin-fixed head and neck cancer specimens (n = 15). RESULTS: All six motility-associated proteins were overexpressed in the premalignant and squamous cell cancer cell lines relative to normal keratinocytes. Immunohistochemistry with Rho A and Rac 2 showed increased staining in areas of cancer but not in normal tissue. CONCLUSION: Proteins involved in cell motility can be used as markers for head and neck squamous cell carcinoma. The head and neck cell lines used in this study may be used as a model to further investigate cell motility. Molecular markers of motility could have a significant impact on the diagnosis and staging of cancers originating from differentiated non-motile cells