Inositol phosphates are well-known signaling molecules, whereas the inositol pyrophosphates, such as diphosphoinositol pentakisphosphate (InsP7/IP7) and bis-diphosphoinositol tetrakisphosphate (InsP8/IP8), are less well characterized. We demonstrate physiologic regulation of Dictyostelium chemotaxis by InsP7 mediated by its competition with PtdIns(3,4,5)P3 for binding pleckstrin homology (PH) domain-containing proteins. Chemoattractant stimulation triggers rapid and sustained elevations in InsP7/InsP8 levels. Depletion of InsP7 and InsP8 by deleting the gene for InsP6 kinase (InsP6K/IP6K), which converts inositol hexakisphosphate (InsP6/IP6) to InsP7, causes rapid aggregation of mutant cells and increased sensitivity to cAMP. Chemotaxis is mediated by membrane translocation of certain PH domain-containing proteins via specific binding to PtdIns(3,4,5)P3. InsP7 competes for PH domain binding with PtdIns(3,4,5)P3 both in vitro and in vivo. InsP7 depletion enhances PH domain membrane translocation and augments downstream chemotactic signaling activity.

The ability to sense and respond to shallow gradients of extracellular signals is remarkably similar in Dictyostelium discoideum amoebae and mammalian leukocytes. Chemoattractant receptors and G proteins are fairly evenly distributed along the cell surface. Receptor occupancy generates local excitatory and global inhibitory processes that balance to control the chemotactic response. Uniform stimuli transiently recruit PI3Ks to, and release PTEN from, the plasma membrane, while gradients of chemoattractant cause the two enzymes to bind to the membrane at the front and back of the cell, respectively. Interference with PI3Ks alters chemotaxis, and disruption of PTEN broadens PI localization and actin polymerization in parallel. Thus, counteracting signals from the upstream elements of the pathway converge to regulate the key enzymes of PI metabolism, localize these lipids, and direct pseudopod formation.

In chemotaxing cells, localization of phosphatidylinositol 3,4,5-trisphosphate (PI(3,4,5)P 3 ) to the leading edge of the cell sets the direction and regulates the formation of pseudopods at the anterior. We show that the lipid phosphatase activity of PTEN mediates chemotaxis and that the sharp localization of PI(3,4,5)P 3 requires localization of PTEN to the rear of the cell. Our data suggest that a phosphatidylinositol 4,5-bisphosphate (PI(4,5)P 2 ) binding motif at the N terminus of PTEN serves the dual role of localizing the enzyme to the membrane and regulating its activity. Mutations in this motif enhance catalytic activity but render the enzyme inactive in vivo by preventing membrane association. The key role of this motif may explain the heretofore puzzling tumor-suppressing mutations occurring within the PI(4,5)P 2 binding motif. On the other hand, the localization of PTEN does not depend on its phosphatase activity, the actin cytoskeleton, or the intracellular level of PI(3,4,5)P 3 , suggesting that events controlling localization are upstream of phosphoinositide signaling.

Recent studies have demonstrated that PH domains specific for PI(3,4,5)P3 accumulate at the leading edge of a number of migrating cells and that PI3Ks and PTEN associate with the membrane at the front and back, respectively, of chemotaxing Dictyostelium discoideum cells. However, the dependence of chemoattractant induced changes in PI(3,4,5)P3 on PI3K and PTEN activities have not been defined. We find that bulk PI(3,4,5)P3 levels increase transiently upon chemoattractant stimulation, and the changes are greater and more prolonged in pten- cells. PI3K activation increases within 5 s of chemoattractant addition and then declines to a low level of activity identically in wild-type and pten- cells. Reconstitution of the PI3K activation profile can be achieved by mixing membranes from stimulated pi3k1-/pi3k2- cells with cytosolic PI3Ks from unstimulated cells. These studies show that significant control of chemotaxis occurs upstream of the PI3Ks and that regulation of the PI3Ks and PTEN cooperate to shape the temporal and spatial localization of PI(3,4,5)P3.

The directional movement of cells in chemoattractant gradients requires sophisticated control of the actin cytoskeleton. Uniform exposure of Dictyostelium discoideum amoebae as well as mammalian leukocytes to chemoattractant triggers two phases of actin polymerization. In the initial rapid phase, motility stops and the cell rounds up. During the second slow phase, pseudopodia are extended from local regions of the cell perimeter. These responses are highly correlated with temporal and spatial accumulations of PI(3,4,5)P3/PI(3,4)P2 reflected by the translocation of specific PH domains to the membrane. The slower phase of PI accumulation and actin polymerization is more prominent in less differentiated, unpolarized cells, is selectively increased by disruption of PTEN, and is relatively more sensitive to perturbations of PI3K. Optimal levels of the second responses allow the cell to respond rapidly to switches in gradient direction by extending lateral pseudopods. Consequently, PI3K inhibitors impair chemotaxis in wild-type cells but partially restore polarity and chemotactic response in pten- cells. Surprisingly, the fast phase of PI(3,4,5)P3 accumulation and actin polymerization, which is relatively resistant to PI3K inhibition, can support inefficient but reasonably accurate chemotaxis.

Directed cell migration involves signaling events that lead to local accumulation of PI(3,4,5)P(3), but additional pathways act in parallel. A genetic screen in Dictyostelium discoideum to identify redundant pathways revealed a gene with homology to patatin-like phospholipase A(2). Loss of this gene did not alter PI(3,4,5)P(3) regulation, but chemotaxis became sensitive to reductions in PI3K activity. Likewise, cells deficient in PI3K activity were more sensitive to inhibition of PLA(2) activity. Deletion of the PLA(2) homolog and two PI3Ks caused a strong defect in chemotaxis and a reduction in receptor-mediated actin polymerization. In wild-type cells, chemoattractants stimulated a rapid burst in an arachidonic acid derivative. This response was absent in cells lacking the PLA(2) homolog, and exogenous arachidonic acid reduced their dependence on PI3K signaling. We propose that PLA(2) and PI3K signaling act in concert to mediate chemotaxis, and metabolites of PLA(2) may be important mediators of the response.

The role of PI(3,4,5)P 3 in Dictyostelium signal transduction and chemotaxis was investigated using the PI3-kinase inhibitor LY294002 and pi3k-null cells. The increase of PI(3,4,5)P 3 levels after stimulation with the chemoattractant cAMP was blocked >95% by 60 M LY294002 with half-maximal effect at 5 M. This correlated well with the inhibition of the membrane translocation of the PH-domain protein, PHcracGFP. LY294002 did not reduce cAMP-mediated cGMP production, but significantly reduced the cAMP response up to 75% in wild type and completely in pi3k-null cells. LY294002-treated cells were round, not elongated as control cells. Interestingly, cAMP induced a time and dose-dependent recovery of cell elongation. These elongated LY294002-treated wild-type and pi3k-null cells exhibited chemotactic orientation toward cAMP that is statistically identical to chemotactic orientation of control cells. In control cells, PHcrac-GFP and F-actin colocalize upon cAMP stimulation. However, inhibition of PI3-kinases does not affect the first phase of the actin polymerization at a wide range of chemoattractant concentrations. Our data show that severe inhibition of cAMPmediated PI(3,4,5)P 3 accumulation leads to inhibition of cAMP relay, cell elongation and cell aggregation, but has no detectable effect on chemotactic orientation, provided that cAMP had sufficient time to induce cell elongation.

In the conduct of multicenter clinical trials, comparisons among the clinical centers are often made with respect to such measures as incidence of diagnosis of various events, operative mortality, and patients adherence to the treatment regimen. Slippage and outlier tests for statistically identifying clinics that are outliers - that is, truly different from the rest of the clinics - with respect to such measures are described. Critical values obtained for these tests by computer simulation are given for various numbers of clinics, numbers of patients enrolled per clinic, and probabilities (p) of a patient incurring a given event under the null hypothesis of no differences among the clinics. Previously published critical values based on normally distributed response variables are applicable only for p close to 0.5. The slippage and outlier tests are compared with respect to statistical power. Alternative forms of the slippage test based on the angular transformation and on hypergeometric probabilities are considered.