PerspectiveSignal Transduction

Signals to Move Cells

See allHide authors and affiliations

Science  11 Feb 2000:
Vol. 287, Issue 5455, pp. 982-985
DOI: 10.1126/science.287.5455.982

Directed movement along a concentration gradient of chemical attractants is essential for the survival of microbes and for the effectiveness of the immune system's phagocytic cells that pursue and engulf them. This directional motility (called chemotaxis) is governed by chemoattractants such as interleukin-8 (IL-8), C5a, and the formylated peptide N-formyl-Met-Leu-Phe (fMLP), which bind to receptors on the surface of cells. Chemoattractant receptors belong to the seven-transmembrane helix receptor family. After binding their chemoattractant ligands, these receptors become activated and transmit their signals to heterotrimeric G proteins (see the figure). The G protein complex dissociates into the α subunit and the βg subunits, which in turn bind and activate target enzymes such as phospholipase C, phosphoinositide 3-kinase (PI 3-kinase) or adenylyl cyclase. These enzymes generate intracellular messengers that initiate a cascade of events culminating in the biological response to the receptor signal. Five papers in this issue of Science (1-5) investigate how PI 3-kinase is involved in heptahelical receptor signaling and chemotaxis. Together they give us new insights into the intracellular machinery that chemoattractant receptors harness to produce their downstream effects.

Cells on the move are pipped at the post.

Asymmetric signal transduction by the catalytic subunit (p110γ of PI 3-kinase γ. A resting G protein-coupled heptahelical chemoattractant receptor (left) binds its ligand (red) and p110γ becomes activated through an interaction with the G protein βg subunits. The p110γ complex then phosphorylates PIP2, resulting in the production of PIP3 (middle). PIP3 is a target for PH domain-containing proteins, which activate kinases and small GTPases (right). These are important for the conversion of the initial receptor signal into the migration response.

Phosphoinositide 3-kinase converts the membrane phospholipid phosphatidylinositol-4,5-bisphosphate (PIP2) into phosphatidylinositol-3,4,5-trisphosphate (PIP3) (6). Multiple PI 3-kinase isoforms, characterized by their catalytic subunits (p110α, β, δ, γ), catalyze this reaction (6). Of these, the p110γ catalytic subunit is activated by the βg complex of heterotrimeric G proteins (7), which suggests that it has a function downstream of heptahelical chemoattractant receptors. The others (p110α, β, δ) are activated by tyrosine kinase receptors through an adapter molecule that acts as a bridge between the p110 subunit and phosphotyrosine residues on the activated receptor. This classification of the PI 3-kinase isoforms on the basis of their mode of activation is useful but it should not be taken as absolute because tyrosine kinase-activated PI 3-kinases are in some cases also under the control of seven transmembrane helix receptors [(8) and references therein].

The variation in PI 3-kinase isoforms is likely to reflect the different functions of the cells in which they are active. Currently available reagents for investigating p110 function have the drawback that they suffer from crossover effects between isoforms. Li et al. reporting on page 1046 (1), Sasaki et al. on page 1040 (2), and Hirsch et al. on page 1049 (3) overcome this problem by genetically engineering mice to lack the p110γ catalytic subunit of PI 3-kinase γ. With these knockout mice it has been possible to rigorously assign the activation of PI 3-kinase isoforms to different receptor classes and to investigate the role of PI 3-kinase γ in the downstream biological responses to receptor activation.

The three papers convincingly show that leukocytes lacking p110γ are unable to produce PIP3 in response to activation of their fMLP, C5a, or IL-8 receptors. Subsequent signal transduction events known to be regulated by PIP3 are also impaired in p110γ-deficient cells. For instance, phosphorylation and activation of protein kinase B (PKB), which requires PIP3, no longer occurs in response to chemoattractants (2, 3). By contrast, in p110γ-deficient cells both PIP3 production and PKB phosphorylation in response to activation of tyrosine kinase receptors (such as the GM-CSF receptor) were normal. This indicates that these receptors use the α, β, or δ isoforms of p110. Together these observations suggest that seven transmembrane helix receptors engage p110γ (but not the other p110s) to produce PIP3, and that conversely p110γ is not used by tyrosine kinase receptors.

Loss of p110γ results in changes in the cells of the hematopoietic system. Sasaki et al. (2) demonstrate that there is reduced thymocyte survival and defective T lymphocyte activation in p110γ-deficient mice. In contrast, the B cell population appeared to be unaffected by the absence of p110γ (2). The Li (1), Sasaki (2), and Hirsch (3) groups show that in neutrophils, p110γ mediates activation of NADPH oxidase by the fMLP receptor but not by serum opsonin receptors or phorbol esters. The three groups also show that p110γ has a striking effect on the ability of neutrophils and peritoneal macrophages to migrate. Cells deficient in p110γ show a reduction in movement toward a chemoattractant (such as fMLP) and a reduced capacity for migration to the peritoneum in various mouse models of peritonitis. Indeed, the migration responses of cells in vivo seemed more severely compromised than those in vitro implying that p110γ may mediate signals in vivo other than those elicited in vitro.

Two of the groups (2, 3) observed that there were greater numbers of neutrophils in the blood of mice deficient in p110γ. This combination of a higher than normal number of neutrophils in the circulation, coupled with their marked inability to escape into the tissues, is reminiscent of a human leukocyte adhesion deficiency syndrome in which the cells lack the adhesion molecule β2-integrin. Therefore, p110γ might be an important component of the signaling pathways of leukocytes that are activated by selectin or integrin adhesion molecules.

The outcome of an acute infection is to a large extent dependent on the result of the race between the host (to accumulate neutrophils locally) and the microbes (to proliferate). Thus, it is not surprising that the p110γ-deficient animals show a severely reduced ability to clear bacteria from the peritoneal cavity (3). Neutrophils and macrophages are also important instigators of the tissue damage produced at sites of inflammation—for example, in acute respiratory distress syndrome or inflammatory bowel disease. Thus, the p110γ catalytic subunit of PI 3-kinase γ might make an attractive anti-inflammatory drug target.

Although these results indicate that p110γ is an important mediator of chemotactic responses, it should be stressed that this function is not unique to this particular isoform. A recent study showed that microinjection of antibodies to p110β and p110δ into a macrophage cell line resulted in reduced cell migration in response to CSF-1, which binds and activates a receptor tyrosine kinase (9). Because all of these isoforms lead to the production of PIP3, this phospholipid molecule can be seen as a general signal for migration responses (see related story). So, the function of the various PI 3-kinase isoforms is to couple different classes of receptors to their cellular responses through synthesis of PIP3. PIP3 binds to pleckstrin homology (PH) domains (10). Kinases containing PH domains, such as phosphoinositide-dependent kinase (11) and PKB, are known effectors of PI 3-kinase. In the slime mold Dictyostelium discoideum, a homolog of PKB is rapidly activated by heptahelical receptors through a PI 3-kinase, and PKB is required for efficient chemotaxis in response to chemoattractants (12). This suggests that PKB may be an important regulator of chemotaxis in mammalian cells. Other PI 3-kinase effectors include exchange factors for small guanosine triphosphatases (GTPases). These GTPases are important in the regulation of the cytoskeleton (13) and have been implicated in migration and chemotaxis (14).

It is interesting to compare the p110γ-deficient phenotype described in the Li, Sasaki, and Hirsch reports in this issue, with that of the recently described mouse lacking the small GTPase Rac2. The similarities are quite remarkable (15)—Rac2-deficient animals have a higher leukocyte blood count, and their leukocytes are less able to infiltrate the peritoneum in experimental inflammatory models, and to migrate in vitro. Furthermore, they are severely compromised in their capacity for L-selectin-mediated endothelial rolling (something that was not tested in the p110γ knockout animals). The overlap in phenotypes suggests that Rac2 may be in the same leukocyte signaling pathway as p110γ. In vitro studies also imply that Rac is downstream of p110γ because activation of the fMLP receptor results in cytoskeletal rearrangements in a pathway involving Gβg, p110γ, the Rac exchange factor vav, and Rac (16). However, the study by Li et al. (1) suggests that Rac activation still occurs in p110γ-deficient cells. The detailed analysis of possible downstream effector systems in p110γ-deficient mice will no doubt further elucidate the mechanisms involved in the regulation of phagocytic cell migration.


View Abstract

Stay Connected to Science

Navigate This Article