Mark J. DiNubile

High concentrations of phosphatidylinositol-4,5-bisphosphate may promote actin filament growth by three potential mechanisms: Inhibiting capping by neutrophil lysates, severing actin filaments and removing capping protein-β2 from barbed ends1

Cell locomotion requires rapid growth of cortical actin filaments whose barbed ends are capped in the resting cell. Phosphatidylinositol-4,5-bisphosphate (PIP2) may play a critical role as an intracellular messenger in cytoskeletal rearrangement after stimulation. We have examined the effects of PIP2 micelles on the Ca2+-independent actin filament capping activity in high speed supernatants of neutrophil lysates which we had previously demonstrated to be almost entirely due to capping protein-beta2, a homologue of cap Z. High concentrations of PIP2 totally prevented the capping of exogenous spectrin-F-actin seeds by dilute supernatants of neutrophil extracts. Capping could also be inhibited, albeit less effectively, by PIP and PI, but not by other phospholipids. When incubated with filaments in the absence of supernatant, PIP2 increased the number of growing ends. PIP2 also uncapped previously capped actin filaments, as demonstrated by incubating supernatant-capped and uncapped seeds with and without PIP2 and then comparing the initial elongation rates after addition of pyrenyl-G-actin. Incubation of capped seeds with high concentrations of PIP2 increased the number of free barbed ends to a level comparable to that of the uncapped seeds exposed to PIP2. PIP2 caused uncapping to occur too quickly to be explained simply by the off-rate of capping protein-beta2, implying that PIP2 interacted directly with capping protein on the filament ends. In fact, PIP2 transiently uncapped capped seeds in the presence of excess free capping protein. From our data, we estimate that millimolar concentrations of PIP2 (almost 100-fold higher than the amount predicted from the effective concentration in purified systems) would be required to inhibit all the capping protein-beta2 in the cytosol. This discrepancy probably results, in large part, from sequestration of PIP2 by other PIP2-binding proteins in the cytoplasm. If PIP2 mediates differential cytoskeletal growth after chemoattractant stimulation in vivo, very high concentrations may be required subjacent to the plasma membrane for regional severing and uncapping of actin filaments to occur quickly near the perturbed membrane.

CAPPING OF THE BARBED ENDS OF ACTIN FILAMENTS BY A HIGH-AFFINITY PROFILIN-ACTIN COMPLEX

Profilin, a ubiquitous 12 to 15-kDa protein, serves many functions, including sequestering monomeric actin, accelerating nucleotide exchange on actin monomers, decreasing the critical concentration of the barbed end of actin filaments, and promoting actin polymerization when barbed ends are free. Most previous studies have focused on profilin itself rather than its complex with actin. A high-affinity profilin-actin complex (here called profilactin) can be isolated from a poly-(L)-proline (PLP) column by sequential elution with 3 M and 7 M urea. Profilactin inhibited the elongation rate of pyrenyl-G-actin from filament seeds in a concentration- and time-dependent manner. Much greater inhibition of elongation was observed with spectrin-F-actin than gelsolin-F-actin seeds, suggesting that the major effect of profilactin was due to capping the barbed ends of actin filaments. Its dissociation constant for binding to filament ends was 0.3 microM; the on- and off-rate constants were estimated to be 1.7 x 10(3) M-1 s-1 and 4.5 x 10(-4) s-1, respectively. Purified profilin (obtained by repetitive applications to a PLP column and assessed by silver-stained polyacylamide gels) did not slow the elongation rate of pyrenyl-G-actin from filament seeds. Capping protein could not be detected by Western blotting in the profilactin preparation, but low concentrations of gelsolin did contaminate our preparation. However, prolonged incubation with either calcium or EGTA did not affect capping activity, implying that contaminating gelsolin-actin complexes were not primarily responsible for the observed capping activity. Reapplication of the profilactin preparation to PLP-coupled Sepharose removed both profilin and actin and concurrently eliminated its capping activity. Profilactin that was reapplied to uncoupled Sepharose retained its capping activity. Phosphatidylinositol-4,5-bisphosphate (PIP2) was the most potent phosphoinositol in reducing the capping activity of profilactin. Dissociation of the tight profilactin complex may serve as a unique mechanism by which profilin helps regulate actin filament growth.

Nucleation and elongation of actin filaments in the presence of high speed supernate from neutrophil lysates: modulating effects of Ca2+ and phosphatidylinositol-4,5-bisphosphate.

Cell motility depends on the rapid growth of cortical actin filaments whose barbed ends are capped in the resting cell. High speed supernates (HSS) of dilute neutrophil lysates contain actin monomers and/or oligomers that can be induced to polymerize by certain stimuli. We questioned whether some of the actin remaining in the supernate after high speed centrifugation exists as occult nucleation sites which can elongate when uncapped. Phosphatidylinositol-4,5-bisphosphate (PIP2) may play a critical role as an intracellular messenger in cytoskeletal rearrangement after stimulation by removing cappers from barbed filament ends. The experiments reported here examine the separate and interactive effects of PIP2 micelles and micromolar [Ca2+] on the rates of nucleation and elongation of pyrenyl-G-actin in the presence of HSS. HSS slowed the nucleation and elongation rates of gel-filtered pyrenyl-G-actin polymerized at submicromolar [Ca2+]. Under these conditions, PIP2 only slightly increased the number of nucleation sites, but delayed the slowing of the elongation rate in the presence of HSS. Nucleating activity in HSS could be induced by the addition of micromolar [Ca2+] and totally abolished by immunoprecipitation of gelsolin from HSS; incubation of HSS with PIP2 at micromolar [Ca2+] slightly decreased the number of calcium-induced nucleation sites in the supernate. Incubation of HSS with PIP2 before the addition of calcium led to a greater reduction in Ca2+-inducible nucleation sites. HSS possessed more nucleation sites after simultaneous exposure to PIP2 and Ca2+, followed by chelation of Ca2+ with EGTA, than HSS preincubated at micromolar [Ca2+] without PIP2. At submicromolar [Ca2+], PIP2 only generated a few barbed end nucleation sites in the HSS, but lessened the gradual slowing of elongation seen with HSS in the absence of PIP2, presumably by preventing capping by capping protein-beta2 in the supernate. Pointed end nucleating sites in HSS, attributable to gelsolin, could be created by adding micromolar [Ca2+]. The preincubation of HSS with PIP2 in the absence of micromolar [Ca2+] decreased the number of Ca2+-inducible nucleation sites in the HSS. Under conditions mimicking the sequential rise and fall of cytosolic [Ca2+] after stimulation, PIP2 accelerated actin polymerization despite the inhibitory action of HSS by maintaining Ca2+-activated nucleation sites. These observations suggest that a possible role for PIP2 in modulating cytoskeletal growth in vivo may be to regulate nucleation sites activated by sequential changes in cytosolic [Ca2+].