Tumor markers are defined as antigens specifically expressed in tumor cells or tissues that can be utilized for pathological diagnosis. Here we propose that host immune and/or inflammatory cells could be used as tumor markers because they could predict the biological malignancy of tumors. First we showed that CD8+ T cells infiltrated within cancer cell nests in colorectal cancer could be a novel prognostic factor as revealed by both uni− and multivariate analyses; the more CD8+ cells, the better the survival rate. This was also a clinicopathological demonstration of anti−tumor immunity. Next we showed that macrophages along the invasive margin could function to suppress hematogenous metastasis because these cells were smaller in number in cases with simultaneous or metachronous hematogenous metastasis. These macrophages expressed costimulatory molecules B7−2, HLA−DR, CD40, and CD11c, sharing a phenotype with dendritic cells, representative antigen presenting cells. This suggests their involvement in tumor immunity. Our data indicate that lymphocytes and macrophages could influence the biological malignancy of colorectal cancer.
We examined nitric oxide synthase (NOS)−containing neurons in thiamine−deficient mice histochemically using NADPH−diaphorase (NADPH−d). Decreases in the number of cell bodies and the densities of fibers in NADPH−d−containing neurons were detected in selective areas of the thalamus, the lateral habenular nucleus, the dorsal raphe nucleus and the medial vestibular nucleus, and decreases in the density of the fibers were also detected in the caudal part of the medial portion of the medial mammillary nucleus in thiamine−deficient mice. A marked loss of NADPH−d−containing neurons was also noted in the hippocampus. The results suggest that NOS is associated with neuropathological vulnerability of the brain in experimental thiamine deficiency.
Calponin (CALP) is a 34kDa protein which interacts with actin, tropomyosin and calmodulin, and regulates contractive roles of smooth muscle cells. The present study deals with the expression of CALP and actin in myoepithelial cells (MECs) during postnatal development of rat salivary glands. In the postnatal development of MECs, the specimens obtained from submandibular (SMG) or sublingual glands (SLG) on 0 to 84days old rats were examined by immunocytochemistry.MECs detected by CALP and actin were observed from day 0 to mature salivary glands, and the expressions of actin immunoreaction in MECs of salivary glands on days 5 or 7 were higher than those of CALP, while in MECs of salivary glands of rats over 5 weeks old, the expressions of CALP were a little higher than in actin. MECs in SMGs and SLGs during the postnatal development were characterized by existence of CALP immunoreactivity, whereas actin immunostaining was strongly localized in the smooth muscle cells of the blood vessels. No MECs detected with CALP and actin were observed in ductal segments including striated ducts and granular convoluted tubules (GCT), suggesting that the secretion from GCT cells was not related to CALP and actin. The CALP immunoreactivity is a useful marker for MECs in salivary glands in the rat.
We localized types I, III, IV and VI collagen in the rat anterior pituitary using immunofluorescence staining and identified collagen synthesizing cells using double immunostaining of primary cultured anterior pituitary cells. Collagen types I, III and VI were located around the basement membrane and type IV was found at the basement membrane of the anterior pituitary gland. However, none of these extracellular matrices were observed in glandular cells. Primary cultured cells formed small clusters, although several individual cells remained isolated. Type I collagen was identified at the cytoplasm. Type III collagen appeared at the cytoplasm or along cell profiles as short lines or dots. Types IV and VI collagen appeared as continuous or dotted lines along cell profiles. Double immunostaining revealed that type I collagen was synthesized by GH, PRL, LH, TSH, ACTH and folliculo−stellate cells and that type III collagen was synthesized by PRL, LH, TSH and folliculo−stellate cells. These results show that anterior pituitary cells can synthesize some collagens, and suggest that the types I, III, IV and VI collagen in the anterior pituitary gland may be synthesized by glandular cells in vivo.
VacA cytotoxin secreted by Helicobacter pylori causes cytoplasmic vacuolation in gastric epithelial cells. To elucidate whether VacA directly affects the cell surface, changes in the membrane structure after VacA treatment were examined in a human gastric epithelial cell culture (AZ−521) at an ultrastructural level. When the time−course changes in the localization of VacA were analyzed by immunoelectron microscopy in AZ−521 cells that had been exposed to VacA, VacA was detected predominantly on the plasma membrane at an early stage of the exposure, and later in endocytotic vesicles and/or tubulo−vesicles. Five to 24hr after exposure, the VacA was distributed in a limestone cavern like−structure, which was formed by elongated surfaces of plasma membranes, as shown by scanning electron microscopy. Therefore, the binding of VacA to the plasma membrane and the subsequent internalization into the enlarged cavern like−plasma membranes may be involved in the induction of vacuolation of gastric epithelial cells by VacA.
Astroglial elements in the suprachiasmatic nucleus (SCN) in golden hamsters and rats were investigated by GFAP immunocytochemistry and a computer−assisted analyzing system. Serial frontal sections, including the SCN region, were prepared and GFAP immunoreactivity in the following areas were measured as optical density (O.D.) per area (O.D./μm2); (1) the whole SCN region, (2) the ventrolateral portion (VLSCN), where vasoactive intestinal peptide (VIP) neurons were mainly distributed, (3) the dorsomedial portion, where arginine vasopressin (AVP) neurons were mainly distributed, and (4) as determined by unilateral intraocular injection of cholera toxin B subunit, the area where retinohypothalamic tract (RHT) fibers were distributed. In the rat, GFAP immunoreactivity was observed throughout the SCN, particularly in the VLSCN, while in the golden hamster, astroglial elements enclosed the whole region of the SCN. Inaddition, quantitative analysis revealed that the O. D. of the GFAP immunoreactivity in the VLSCN, where VIP neurons and the RHT fibers were distributed, were significantly higher than those in other regions of the SCN. These results strongly suggested that the astroglial elements, VIP neurons and RHT fibers have close morphological interactions, although delicate differences were demonstrated in both species.
Adenosine 5′−triphosphate (ATP) is known to have various effects on cells as an extracellular nucleotide. It decreases the degree of edema and the requirement for intravenous hydration in critically ill patients. We examined the direct effects of ATP on the permeability of a vascular endothelial cell monolayer using an in vitro co−culture chamber system. Human umbilical vein endothelial cells were cultured on filter membranes of co−culture chambers to form a confluent monolayer. FITC-labeled 70k dalton dextran and ATP at various concentrations were added to the upper chamber. The concentrations of FITC−dextran, which permeated to the lower chamber through the endothelial cell monolayer, were determined in 4hr. A significant decrease in the permeability of FITC−dextran was observed with 10-6 to 10-3M of ATP. One, 10, 100, and 1000μM ATP decreased the permeability from 1.00±0.05 to 0.75±0.05, 0.66±0.04, 0.61±0.07, and 0.58±0.08 respectively (mean±SD, n=5, p<0.05). Adenosine, an ATP metabolite, did not change the permeability. Phorbol 12−myristate 13−acetate (PMA), a protein kinase C stimulator, decreased the permeability. Treatment with staurosporine, a protein kinase inhibitor, and myrystoylated protein kinase C (19-27), a protein kinase C specific inhibitor, suppressed the decrease of the permeability by ATP. These results indicated that ATP decreases the permeability of endothelial cell monolayers through protein kinase C (PKC) in a signal transduction system.
DNA degradation induced in the pancreatic tissues of rats by an administration of a diabetogenic dose of alloxan was studied with TUNEL assay. TUNEL−positive nuclei first appeared in pancreatic islet cells. Up to 3hr after an alloxan injection, the reaction products of TUNEL assay were found only in the nucleus of islet cells. However, 7hr after alloxan administration, some nuclei of cells in the exocrine portion, such as acinar cells, ductal epithelial cells, and some other cells were also labeled by TUNEL reaction in addition to the islet cells. Moreover, 7hr after alloxan administration, the TUNEL reaction products were found not only in the nucleus but also in the cytoplasm of insulin−producing cells injured by alloxan, as well as in the area containing amorphous substances that seemed to be the cell debris derived from destructed insulin−producing cells. The TUNEL−positive amorphous substances seemed to be finally absorbed by macrophages. These phenomena, the diffusion and mixing of the cytoplasmic and nuclear components of insulin−producing cells, are considered characteristics of islet cell necrosis induced by alloxan.
A simple method of in−situ hybridization (ISH) without using any bacteriological or recombinant procedures was exemplified by examining the localization of the two isoforms of calcineurin catalytic subunits, Aα and Aβ, in rat brain tissue. Total RNA was extracted from rat brain tissue and utilized as a template to perform probe labeling during reverse transcription and subsequent polymerase chain reaction steps (RT−PCR) by incorporating digoxigenin−11−dUTP. These PCR probes were employed for ISH. To confirm the validity of this method, the expression of mRNAs of alpha-fetoprotein and albumin in rat liver tissue was also examined using the same procedure. The distribution of calcineurin Aα and Aβ mRNAs was clearly detected in brain tissue. The entire procedure took less than a few days from starting the preparation of the probes to the light microscopic observation of the specimens. Moreover, the sensitivity of the ISH using these probes was higher than that of the end−labeled probe. This method has significant advantages as it allows laboratories without any recombinant facilities to accomplish ISH, and has the potential for widespread application.
Morphological observations have indicated that there are two mechanisms in the human carcinogenesis: de novo carcinogenesis and multistep carcinogenesis through preneoplastic lesions. Recent molecular pathological techniques have made it possible to precisely analyze the step of the genetic event in the carcinogenetic process. Here, we report molecular pathological results suggesting multistep carcinogenesis in ovarian mucinous tumors. Specimens of 44 ovarian mucinous tumors, including 15 mucinous cystadenomas, 19 mucinous cystic tumors of borderline malignancy and 10 mucinous adenocarcinomas, were collected from our surgical files. Immunohistochemical analyses, including p53, Ki−67, p21/WAF1, cyclin D1 and c−erbB2, were performed. Loss of heterozygosity using 7 microsatellite markers and K−ras point mutation were examined. Furthermore, the correlation between the histological and genetic heterogeneities was analyzed. It was found that p53, Ki−67, and c−erbB2 expressions increased according to the histological grade as well as the frequency of genetic abnormalities. Histological heterogeneities were relatively well−correlated with the genetic heterogeneities. These results suggest that there is a subset of ovarian mucinous tumors in which there is an accumulation of genetic abnormalities.
The bladder hypertrophy caused by infravesical obstruction due to benign prostatic hyperplasia (BPH) often results in bladder dysfunction. Here we attempted to study morphometrically the histological changes in bladder detrusor with reference to extracellular matrix and to find the correlation of those changes with bladder hypertrophy. Ultrasound estimated bladder weight (UEBW) and urodynamic parameters were evaluated in 34 BPH patients (71.4±7.9 years) before surgery. Bladder wall samples were obtained at surgery and stained using Masson trichrome method and immunohistochemistry for collagen types I and III and fibronectin, followed by morphometric color image analysis system. UEBW, ratios of connective tissue−to−smooth muscle (c/m), collagen type III-immunoreactive area−to−collagen type I-immunoreactive area (collagen III/I) and fibronectin−immunoreactive area−to−non−immunoreactive area (f/n) ranged from 21.0 to 126.1g (47.9±25.4g), 12.9 to 53.3% (27.6±9.2%), 31.9 to 69.7% (46.5±8.0%) and 40.2 to 125.9% (59.8±15.9%), respectively. UEBW correlated with c/m (r=0.744, p<0.0001), collagen III/I (r=0.698, p<0.0001) and f/n (r=0.733, p<0.0001). Among urodynamic parameters, bladder compliance was the only one that correlated with c/m (r=0, 673, p<0.0001), collagen III/I (r=0.475, p<0.05) and f/n (r=0.590, p<0.001). These results suggest that an abnormal increase of connective tissue accompanied with increased levels of collagen type III and fibronectin could contribute to advanced bladder hypertrophy with a loss of elasticity of the bladder wall in patients with infravesical obstruction.
It has been proposed that a difference between the length of the cell cycle and the S−phase in cells in the rhombomere boundary and those in the center of each rhombomere (interboundary) produces a driving force for hindbrain segmentation. To assess the validity of this hypothesis, we examined the cell kinetics of rhombomeric cells in the chick embryo using two kinds of labeling procedures. Cumulative labeling indices with bromodeoxyuridine (BrdU) revealed that the ratio of the number of S−phase nuclei in the boundary tended to be greater than in the interboundary area, whereas the time intervals required for reaching the plateau phase of labeling indices were similar (6hr) in the two areas through all stages examined with the exception of r4 at stage 15. By employing a double labeling method using BrdU and iododeoxyuridine that enables the comparison of the length of the S−phase in cells in the boundary and interboundary areas, we demonstrated that the ratio of the number of double−labeled nuclei in the two areas was not significantly different at stage 15. The present studies have indicated that the cell kinetic parameters of the cell populations in the boundary are not significantly different from those in the interboundary during hindbrain segmentation and suggest that molecular and cellular interactions other than the differential growth rate may play more crucial roles in hindbrain segmentation.