Predicting the prevalence of a disease in a cohort at risk.

A non-homogeneous Markov chain model is used to describe the evolution of a progressive chronic disease. Transition probabilities are estimated from data on coronary heart disease in Australia. By varying the transition probabilities, the effects of changes in medical management or in the natural history of the disease are examined. For reducing rates of fatal and non-fatal acute myocardial infarction a multifaceted strategy of primary and secondary prevention and improved acute care is found most effective. Primary prevention reducing the risk of early stages of the disease has the potential to yield the highest proportion of disease-free people. By contrast, intensive medical treatment of acute disease would reduce mortality but increase the total burden of disease in the population.     

A novel 37-Kd adhesive membrane protein from cloned murine bone marrow stromal cells and cloned murine hematopoietic progenitor cells

The adhesion of hematopoietic progenitor cells to bone marrow stromal cells is critical to hematopoiesis and involves multiple effector molecules. Stromal cell molecules that participate in this interaction were sought by analyzing the detergent-soluble membrane proteins of GBI/6 stromal cells that could be adsorbed by intact FDCP-1 progenitor cells. A single-chain protein from GBI/6 cells having an apparent molecular weight of 37 Kd was selectively adsorbed by FDCP-1 cells. This protein, designated p37, could be surface-radiolabeled and thus appeared to be exposed on the cell membrane. An apparently identical 37- Kd protein was expressed by three stromal cell lines, by Swiss 3T3 fibroblastic cells, and by FDCP-1 and FDCP-2 progenitor cells. p37 was selectively adsorbed from membrane lysates by a variety of murine hematopoietic cells, including erythrocytes, but not by human erythrocytes. Binding of p37 to cells was calcium-dependent, and was not affected by inhibitors of the hematopoietic homing receptor or the cell-binding or heparin-binding functions of fibronectin. It is proposed that p37 may be a novel adhesive molecule expressed on the surface of a variety of hematopoietic cells that could participate in both homotypic and heterotypic interactions of stromal and progenitor cells.     

Plasma cell granuloma of the lacrimal sac

A 31-year-old man with a 2-year history of recurrent dacryocystitis underwent dacryocystorhinostomy. An unexpected operative finding was a polyp in the lacrimal sac. Histopathological examination revealed a plasma cell granuloma. The nature of this uncommon inflammatory pseudotumour of the lacrimal sac is discussed.     

A nonrandom interneuronal pattern in the developing frog spinal cord

In the developing spinal cord of the frog, Xenopus laevis, a population of interneurons assumes a pattern that represents a previously undescribed level of organization. Glyoxylic acid treatment and immunocytochemistry show that the neurons contain catecholamines and their synthetic enzyme, tyrosine hydroxylase. Cells are located within the ependymal layer of the floor plate region of the larval spinal cord. The cells have several processes including a long one that projects toward the brain without fasciculating with other labeled processes. In addition, the cytoplasm of the catecholaminergic cells extends into the central canal, showing that they are a population of cerebrospinal fluid-contacting neurons. The spatial domain of catecholaminergic neurons starts abruptly at the boundary between the hindbrain and spinal cord and continues to the tip of the tail. The neurons occupy two longitudinal columns within the sheet of floor plate cells, which includes cells that do not exhibit the catecholaminergic phenotype. Unlabeled cells are intercalated between catecholaminergic cells in each column, giving the labeled cells the appearance of being spaced along the length of the spinal cord. This general arrangement is evident at the time of hatching. Spatial analysis showed that the position of cells along a column is not random. The nonrandom behavior is due to cells being excluded from the area immediately surrounding other catecholaminergic cells. Further analysis showed that the cellular pattern lacks segmental or other periodic repeats. Ultimately, the location of a cell within a column depends upon the position of its closest catecholaminergic neighbor. © 1993 Wiley-Liss, Inc.