PathMaster: content-based cell image retrieval using automated feature extraction.

OBJECTIVE: Currently, when cytopathology images are archived, they are typically stored with a limited text-based description of their content. Such a description inherently fails to quantify the properties of an image and refers to an extremely small fraction of its information content. This paper describes a method for automatically indexing images of individual cells and their associated diagnoses by computationally derived cell descriptors. This methodology may serve to better index data contained in digital image databases, thereby enabling cytologists and pathologists to cross-reference cells of unknown etiology or nature. DESIGN: The indexing method, implemented in a program called PathMaster, uses a series of computer-based feature extraction routines. Descriptors of individual cell characteristics generated by these routines are employed as indexes of cell morphology, texture, color, and spatial orientation. MEASUREMENTS: The indexing fidelity of the program was tested after populating its database with images of 152 lymphocytes/lymphoma cells captured from lymph node touch preparations stained with hematoxylin and eosin. Images of "unknown" lymphoid cells, previously unprocessed, were then submitted for feature extraction and diagnostic cross-referencing analysis. RESULTS: PathMaster listed the correct diagnosis as its first differential in 94 percent of recognition trials. In the remaining 6 percent of trials, PathMaster listed the correct diagnosis within the first three "differentials." CONCLUSION: PathMaster is a pilot cell image indexing program/search engine that creates an indexed reference of images. Use of such a reference may provide assistance in the diagnostic/prognostic process by furnishing a prioritized list of possible identifications for a cell of uncertain etiology.

Involvement of a pertussis toxin-sensitive G protein in the mitogenic signaling pathways of sphingosine 1-phosphate.

Sphingosine 1-phosphate, a sphingolipid metabolite, was previously reported to increase DNA synthesis in quiescent Swiss 3T3 fibroblasts and to induce transient increases in intracellular free calcium (Zhang, H., Desai, N. N., Olivera, A., Seki, T., Brooker, G., and Spiegel, S. (1991) J. Cell Biol. 114, 155-167). In the present study, pretreatment of Swiss 3T3 fibroblasts with pertussis toxin reduced sphingosine 1-phosphate-induced DNA synthesis. Sphingosine 1-phosphate decreased cellular cAMP levels and also caused a drastic decrease in isoproterenol- and forskolin-stimulated cAMP accumulation. Pertussis toxin treatment prevented the inhibitory effect of sphingosine 1-phosphate on cAMP accumulation, suggesting that a pertussis toxin-sensitive Gi or Gi-like protein may be involved in sphingosine 1-phosphate-mediated inhibition of cAMP accumulation. Mitogenic concentrations of sphingosine 1-phosphate stimulated production of inositol phosphates which was inhibited by pertussis toxin, while the response to bradykinin was not affected. Furthermore, calcium release induced by sphingosine 1-phosphate, but not by bradykinin, was also attenuated by pertussis toxin treatment. However, sphingosine 1-phosphate-induced phosphatidic acid accumulation was unaffected by pertussis toxin. The increase in specific DNA binding activity of activator protein-1, which was induced by treatment of quiescent Swiss 3T3 fibroblasts with sphingosine 1-phosphate, was also inhibited by pertussis toxin. These results suggest that some of the sphingosine 1-phosphate-induced signaling pathways are mediated by G proteins that are substrates for pertussis toxin.

Stereospecificity of sphingosine-induced intracellular calcium mobilization and cellular proliferation.

Sphingosine is a positive regulator of cell growth in Swiss 3T3 fibroblasts (Zhang, H., Buckley, N. E., Gibson, K., and Spiegel, S. (1990) J. Biol. Chem. 265, 76-81). The present study investigated the stereospecificity of sphingosine-induced cell proliferation and its mitogenic signal transduction mechanisms. D-(+)-erythro Stereoisomers (cis and trans) stimulated DNA synthesis, whereas neither L-(-)-threo-sphingosine (cis or trans) nor DL-threo-dihydrosphingosine had any effect. Previously, we have shown that sphingosine-1-phosphate may mediate the mitogenic effect of sphingosine (Zhang, H., Desai, N. N., Olivera, A., Seki, T., Brooker, G., and Spiegel, S. (1991) J. Cell Biol. 114, 155-167). However, no major differences were found in the formation of D-(+)-erythro- and L-(-)-threo- sphingosine-1-phosphate derived from the respective sphingosine isomers in intact cells. Thus, the stereospecificity of the response to sphingosine may reside at the level of specific intracellular targets for sphingosine-1-phosphate. Sphingosine-1-phosphate triggers dual signal transduction pathways of activation of phospholipase D leading to increases in the levels of phosphatidic acid and mobilization of calcium from internal stores. Both D-(+)-erythro- and L-(-)-threo-sphingosine isomers induced similar increases in phosphatidic acid concomitant with identical decreases in phosphatidylcholine levels. In contrast, only the D-(+)-erythro-stereoisomers (cis and trans) were effective in releasing calcium from intracellular stores. Our results suggest that the formation of phosphatidic acid is not sufficient to mediate sphingosine-stimulated DNA synthesis. However, the stereospecificity of the sphingosine-induced mobilization of calcium from internal stores seems to correlate with the induction of DNA synthesis by sphingosine stereoisomers.

Signaling pathways for sphingosylphosphorylcholine-mediated mitogenesis in Swiss 3T3 fibroblasts.

Sphingosylphosphorylcholine (SPC), or lysophingomyelin, a wide-spectrum growth promoting agent for a variety of cell types (Desai, N. N., and S. Spiegel. 1991. Biochem. Biophys. Res. Comm. 181: 361-366), stimulates cellular proliferation of quiescent Swiss 3T3 fibroblasts to a greater extent than other known growth factors or than the structurally related molecules, sphingosine and sphingosine-1-phosphate. SPC potentiated the mitogenic effect of an activator of protein kinase C, 12-O-tetradecanoylphorbol 13-acetate, and did not compete with phorbol esters for binding to protein kinase C in intact Swiss 3T3 fibroblasts. However, downregulation of protein kinase C, by prolonged treatment with phorbol ester, reduced, but did not eliminate, the ability of SPC to stimulate DNA synthesis, indicating that SPC may act via both protein kinase C-dependent and -independent signaling pathways. SPC induced a rapid rise in intracellular free calcium ([Ca2+]i) in viable 3T3 fibroblasts determined with a digital imaging system. Although the increases in [Ca2+]i were observed even in the absence of calcium in the external medium, no increase in the levels of inositol phosphates could be detected in response to mitogenic concentrations of SPC. Furthermore, in contrast to sphingosine or sphingosine-1-phosphate, the mitogenic effect of SPC was not accompanied by increases in phosphatidic acid levels or changes in cAMP levels. SPC, but not sphingosine or sphingosine-1-phosphate, stimulates the release of arachidonic acid. Therefore, the ability of SPC to act an extremely potent mitogen may be due to activation of signaling pathway(s) distinct from those used by sphingosine or sphingosine-1-phosphate.

Sphingosine-1-phosphate, a metabolite of sphingosine, increases phosphatidic acid levels by phospholipase D activation.

Sphingosine and sphingosine-1-phosphate, metabolites of membrane sphingolipids, have recently been shown to stimulate release of calcium from internal sources and to increase proliferation of quiescent Swiss 3T3 fibroblasts (Zhang, H., Desai, N. N., Olivera, A., Seki, T., Brooker, G., and Spiegel, S. (1991) J. Cell Biol. 114, 155-167). The present study demonstrates that mitogenic concentrations of sphingosine induce early increases in sphingosine-1-phosphate levels which precede the increase in the potent mitogen, phosphatidic acid. Sphingosine-1-phosphate itself induces a more rapid increase in phosphatidic acid, thus suggesting that it may mediate the effects of sphingosine on phosphatidic acid accumulation. The concentration dependence for the formation of phosphatidic acid induced by sphingosine-1-phosphate correlates with its effect on DNA synthesis. Similar to sphingosine, sphingosine-1-phosphate also stimulates the activity of phospholipase D, although a significant effect is observed at a much lower concentration. However, in contrast to previous reports with sphingosine, sphingosine-1-phosphate does not inhibit the phosphatidic acid phosphohydrolase activity in cell homogenates. Thus, in addition to its effect on mobilization of calcium, sphingosine-1-phosphate can increase the level of phosphatidic acid, most likely via activation of phospholipase D. We suggest that sphingosine-1-phosphate mediates the effect of sphingosine on phosphatidic acid accumulation in Swiss 3T3 fibroblasts and may regulate cellular proliferation by affecting multiple transmembrane signaling pathways.

Measuring morphine's capacity to establish a place preference.

A series of experiments are described providing an assessment of the procedures of conditioned place preference (CPP) testing involving an automated system having 12 separate chambers. Experiment 1 provides data to demonstrate (a) that in these chambers no initial preferences for one side over the other exists among rats, (b) that this neutrality of sides is not affected by session lengths between 15 and 60 min, and (c) that the optimal session length for tests in these chambers is on the order of 30 min. Experiment 2 demonstrates the stability of control groups' scores across a number of conditioning and testing sessions. Experiments 3 and 4 provide data to demonstrate (a) that a positive CPP can be established in our chambers using injections of morphine, (b) that a regimen of dosing with unequal numbers of days of putative and alternate conditioning is a reliable and conservative test of the opioid's ability to establish a CPP, and (c) that although the activity of rats decreases across a session, the general activity of rats before and after conditioning procedures is the same. Experiment 5 replicates the procedures employed by Scoles and Siegel (25) and demonstrates that the tendency for rats to explore novel environments is strong, and care must be taken to provide an opportunity for rats to pair different experiences with each side of the chamber in order for a CPP to emerge.

PCP and conditioned place preferences.

Phencyclidine (PCP), in doses of 0.25, 0.35, and 0.45 mg/kg, was administered systemically to male Sprague-Dawley rats in order to determine if a positive conditioned place preference (CPP) could be achieved. Other subjects received systemic injections of morphine, 4.0 mg/kg, as a standard for comparison. At testing, rats receiving 0.45 mg/kg PCP showed a positive CPP compared to controls, as did rats receiving morphine. Previous research had shown that larger doses of PCP and prolonged times after PCP administration produced aversion as indexed by CPP testing. The narrow dose range and short time span in which PCP's positively reinforcing properties are apt to emerge may be related to PCP's psychotomimetic potential and to its ability to sustain its own intake even though aversive effects are often manifest.

Ethanol with small doses of morphine establishes a conditioned place preference.

Conditioned place preference (CPP) testing is a way of indexing the reinforcing efficacy of drugs among rats. CPP testing involves using an alley with two distinctive sides. Typically, rats have drug experiences on one side and placebo experiences on the other. At testing, without drugs, their preference for side is tabulated. Rats' (6 groups of 12 each) place preferences were assessed before and after they were placed, once a day for 9 days, in the putative side of conditioning, and on 3 interspersed days, in the other side. During putative conditioning, one group received saline prior to being placed in both sides (a control group). Two groups had either morphine (2.0 mg/kg) or ethanol (0.5 g/kg) with the putative side of conditioning and saline with the other side. Three groups received morphine plus ethanol before being placed in the putative side of conditioning and either saline, morphine, or ethanol in the other side. At testing, rats that received morphine plus ethanol on side of putative conditioning showed a strong CPP whereas others did not. Results are compatible with the idea that ethanol's reinforcing effect is enhanced when there is a surfeit of opioidergic activity.