Dying enterocytes downregulate signaling pathways converging on Ras: rescue by protease inhibition.

Organ and cell cultures of the small intestine serve as excellent in vitro models for programmed cell death (PCD). Cells cultured in serum-free, minimal medium rapidly died, as evidenced by histological changes, internucleosomal DNA cleavage, and TdT-mediated dUTP nick end labeling. Cell death was pervasive, although nonepithelial cells within the fibrovascular villus core were spared. PCD did not require a functional p53 gene. Serine and cysteine protease inhibitors, but not FCS, suppressed it. Relative to structural and functional proteins, dying enterocytes rapidly downregulated Ras-convergent proteins, including epidermal growth factor receptor, Erb-B2, and the son of sevenless guanine nucleotide exchangers. Reductions in the steady-state levels of both protein and mRNA were observed. These reductions were prevented by a combination of death-defying serine and caspase inhibitors, indicating a requirement for the initiation of death. Thus, during catastrophic PCD, intestinal epithelial cells delete cell surface signaling pathways responsible for Ras activation.

Differential processing of guanylyl cyclase C along villus-crypt axis of rat small intestine.

Many strains of enterotoxigenic Escherichia coli produce a heat-stable peptide enterotoxin (STa) that binds to the intestinal receptor guanylyl cyclase C (GC-C). STa receptors are structurally heterogeneous, but the molecular events causing this heterogeneity remain obscure. We examined the influence of cell position along the villus-crypt axis on STa receptor heterogeneity by fractionating EDTA-dissociated cells that detached in a villus-to-crypt direction. STa affinity labeling experiments revealed that the initially released villus "tip" fraction had four major STa binding proteins (STBPs), with relative molecular weight (M(r)) of 150,000, 135,000, 125,000, and 95,000, that did not react with a GC-C carboxy-terminal antibody. Yet succeeding villus cell fractions had major immunoreactive STBPs with M(r) of 275,000 and 250,000. Limited proteolysis of these larger GC-C isoforms produced 1) smaller STBPs that had M(r) similar to those in the initial villus fraction, 2) a 65,000 M(r) protein GC-C isoform that did not bind STa, and 3) elevated basal and STa-induced cyclase activity. Our data show that STBP structural heterogeneity in the intact intestine arises largely from multisite proteolytic processing of GC-C.

Structure, glycosylation, and localization of rat intestinal guanylyl cyclase C: modulation by fasting.

Guanylyl cyclase C (GC-C), an intestinal receptor guanylyl cyclase, binds diarrhea-producing bacterial ligands such as the Escherichia coli heat stable enterotoxin. We examined the regulatory influence of feeding and fasting on the expression, structure, and biochemical properties of GC-C. When solubilized at 4 degrees C under nonreducing conditions, GC-C from both fed and fasted rats migrated on 7% sodium dodecyl sulfate-polyacrylamide electrophoretic gels as two extremely large aggregates that barely penetrated the stacking and resolving gels. Chemical reduction of disulfide linkages disaggregated GC-C in fed but not fasted rat samples, causing it to migrate as smaller forms (approximately 220 and 240 kDa). Although GC-C aggregates from fasted rats resisted this disaggregating effect of chemical reduction, they rapidly acquired it within 90 min of refeeding. When solubilized at denaturing temperatures (95 degrees C) under reducing conditions, GC-C aggregates largely disassembled into four smaller proteins (relative molecular weight approximately 140,000, 131,000, 85,000, and 65,000). However, the 131-kDa glycoprotein was disproportionately increased in fasted rat membranes. This unit and the 220-kDa unit were sensitive to endoglycosidase H. Subcellular fractionation and immunohistochemical studies revealed a major redistribution of GC-C from surface to intracellular enterocyte sites during fasting.

Characterization of the fundal reflectance of infants.

Pupillary images recorded by a linear knife-edge photoscreener for infants accommodating more or less within the bounds of chromatic aberration show three smooth offset pupillary irradiance ramps when analyzed with a color video camera and frame-grabbing hardware. Both the shape of the ramps and the lack of a second image in a shadowing experiment support the view that the retina acts as an angularly diffuse partial reflector in photorefraction. We propose a lateral image-spreading characteristic to account for observed color-dependent ramp heights and offsets and consider possible physical mechanisms for lateral spreading.

Photorefraction of the living eye: a model for linear knife edge photoscreening.

A computer model is developed to explain colored artifacts in pupil images of the living human eye formed by the linear knife edge photoscreener. An approximate expression for the RGB irradiances recorded by the photoscreener is derived by using the statistical properties of the line spread functions that characterize the source, the aberrations of the eye, and the reflecting properties of the retina. The pupil image for each color is shown to be very much a map of the deviations perpendicular to the knife edge imposed on rays by an equivalent aberration plate, and the overall response of the photoscreener to specific aberrations of the eye is cataloged for errors of accommodation and chromatic aberration, spherical aberration, astigmatism, and tear bands.

[Plasma CK-MB activity and QRS score in estimating acute myocardial infarct size].

Comparison of the peak value of plasma isoenzyme MB of creatine kinase (CK-MB) and the modified Selvester QRS score system from the standard 12-lead electrocardiogram in estimating acute myocardial infarct size and predicting hospital prognosis was performed in 52 patients with initial acute myocardial infarction (AMI). A correlation coefficient (C.C.) of 0.51 in all these patients was found. The C.C. in the patients with anterior AMI (n = 22, r = 0.64) is larger than that in inferior AMI (n = 30, r = 0.34). The prognostic value of the two methods are different. Peak CK-MB activity could predict hospital mortality and morbidity (serious arrhythmia or/and Killip AMI classification more than class III) for both anterior and inferior AMI; however, the QRS score system was significant only for anterior AMI. The clinical significance of a high ratio of peak CK-MB activity to QRS score was discussed briefly.