Tracking down the different forms of nuclear actin

Actin is a rather uncommitted protein with a high degree of structural plasticity: it can adopt a variety of structural states, depending on the specific ionic conditions or the interaction with ligand proteins. These interactions lock actin into a distinct conformation, which specifies the oligomeric or polymeric form it can assume. The interplay between monomeric, oligomeric and polymeric forms is used by the cell to execute an enormous variety of motility processes, such as lamellipodium formation during locomotion or intracellular transport of vesicles. In these cytoplasmic events, monomeric G-actin and filamentous F-actin are the prevalent forms. However, there might be other structural states of actin in cells that have so far not received the attention they deserve. Here, we propose that specific, "unconventional" actin conformations might contribute especially to the multitude of functions executed by actin in the nucleus. We present evidence for the existence of different forms of nuclear actin, taken from studies with selected antibodies.     

Filament assembly and regulation of the actin-activated ATPase activity of thymus myosin

The effects of light chain phosphorylation on the actin-activated ATPase activity and filament assembly of calf thymus cytoplasmic myosin were examined under a variety of conditions. When unphosphorylated and phosphorylated thymus myosins were monomeric, their MgATPase activities were not activated or only very slightly activated by actin, but when they were filamentous, their MgATPase activities were stimulated by actin. The phosphorylated myosin remained filamentous at lower Mg2+ concentrations and higher KC1 concentrations than did the unphosphorylated myosin, and the myosin concentration required for filament assembly was lower for phosphorylated myosin than for unphosphorylated myosin. By varying the myosin concentration, it was possible to have under the same assay conditions mostly monomeric myosin or mostly filamentous myosin; under these conditions, the actin-activated ATPase activities of the filamentous myosins were much greater than those of the monomeric myosins. The addition of phosphorylated myosin to unphosphorylated myosin promoted the assembly of unphosphorylated myosin into filaments. These results suggest that phosphorylation may regulate the actomyosin-based motile activities in vertebrate nonmuscle cells by regulating myosin filament assembly.     

Conformational changes of actin induced by calponin.

Calponin, an actin-linked regulatory protein in smooth muscle, caused a remarkable change in the fluorescence intensity of pyrene-labeled actin in the filamentous form. Calponin, an equimolar ratio to actin, decreased the fluorescence intensity of pyrene-labeled F-actin by some 60% to the level near monomeric actin. This change was partially reversed by Ca2+, when calmodulin was present. Thus it appears that calponin causes conformational changes in actin molecules in an actin filament so as to inhibit their interactions with myosin.     

G-actin pool and actin messenger RNA during development of the apical processes of the retinal pigment epithelial cells of the chick.

It has been suggested that during development an increase in the pool of G-actin may drive the elongation of actin-containing processes which occur in several types of epithelial cells. The apical processes of chick retinal pigment epithelial (RPE) cells elongate during the last 7 days of embryonic life (E15-E21) reaching lengths of 20 microns or more by hatching (E21). F-actin bundles form the cores of these processes. We followed the elongation by measuring F-actin in the cells and cytoskeletons. In correlation with this, we studied by DNAse assay the levels of monomeric actin in supernatants of cell extracts from E13, before elongation starts, to E17, when elongation is well underway. Total F-actin increased 1.9-fold over this time period and cytoskeletal actin increased 2.5-fold. In supernatants from extracts of E13 RPE the monomeric actin concentration was 51 +/- 0.5 micrograms/ml. From estimates of cell volume we calculated the cellular monomeric actin concentration at E13 as at least 510 micrograms/ml (13 microM). We compared this with monomeric actin levels in extracts from RPE at E15 and E17. Allowing for the estimated increase in cell volume, our data show little overall change in cellular monomeric actin concentration at these times. Changes in the level of actin mRNA were measured over the same time period. Normalized to equal RNA, we found a twofold increase in beta actin mRNA and a four- to fivefold increase in message for gamma actin at E17 as compared to E13. In summary, we show that (1) there is a substantial pool of monomeric actin in these epithelial cells before elongation starts; (2) process elongation is not associated with a significant change in the size of this pool; and (3) process elongation is associated with a significant increase in actin mRNA.     

Selective assay of monomeric and filamentous actin in cell extracts,using inhibition of deoxyribonuclease I

A simple and selective assay for monomeric and filamentous actin is presented, based on the inhibition of DNAase I by actin. In mixtures of monomeric and filamentous actin, only the monomeric form is measured as DNAase inhibitor. The total amount of actin in a sample can be determined after depolymerization of F actin with guanidine hydrochloride. The assay is rapid enough to detect changes in the polymerization state of actin in vitro over time intervals as short as 3 min. Data characterizing unpolymerized and filamentous actin pools in extracts of human platelets, lymphocytes and HeLa cells are presented.     

Actin depolymerizing factors ADF7 and ADF10 play distinct roles during pollen development and pollen tube growth

An important player in actin remodeling is the actin depolymerizing factor (ADF) which increases actin filament treadmilling rates. Previously, we had prepared fluorescent protein fusions of two Arabidopsis pollen specific ADFs, ADF7 and ADF10. These had enabled us to determine the temporal expression patterns and subcellular localization of these proteins during male gametophyte development. Here we generated stable transformants containing both chimeric genes allowing for simultaneous imaging and direct comparison. One of the striking differences between the two proteins was the localization profile in the growing pollen tube apex. Whereas ADF10 was associated with the filamentous actin array forming the subapical actin fringe, ADF7 was present in the same cytoplasmic region, but in diffuse form. This suggests that ADF7 is involved in the high actin turnover that is likely to occur in the fringe by continuously and efficiently depolymerizing filamentous actin and supplying monomeric actin to the advancing end of the fringe. The possibility to visualize both of these pollen-specific ADFs simultaneously opens avenues for future research into the regulatory function of actin binding proteins in pollen.     

Actin associated with membranes from 3T3 mouse fibroblast and HeLa cells

A protein component of membranes isolated from 3T3 mouse fibroblasts and HeLa cells has been identified as actin by peptide mapping. Extensive but apparently not total coincidence was found between the peptide maps of these two nonmuscle membrane-associated actins compared to chick skeletal muscle actin. Between 2 and 4 percent of the total membrane protein appears in the actin band on sodium dodecyl sulfate polyacrylamide gels of 3T3 membranes while about 4 percent of the membrane protein appears as the actin band from HeLa membranes. These values represent approximately the same proportion of actin to total protein found in the cell homogenates. Treatment of intact cells with levels of cytochalasin B sufficient to cause pronounced morphological changes did not change the amount of actin associated with the membrane in either 3T3 or HeLa cells. However, incubation of isolated membranes under conditions favoring conversion of actin from filamentous to monomeric form resulted in dissociation of approximately 80 and 60 percent of the actin from 3T3 and HeLa membranes, respectively. Thus, approximately 20 percent of 3T3 membrane actin and 40 percent of HeLa membrane actin remained associated with the membrane even under actin depolymerizing conditions.     

Mechanical properties of actin

We used a cone and plate rheometer to evaluate the mechanical properties of actin over a wide range of oscillation frequencies and shear rates. Remarkably, both filamentous and nonfilamentous actin behaved as viscoelastic solids in both oscillatory and shear type experiments, providing that they were given ample time to equilibrate. Actin was purified by gel filtration from rabbit skeletal muscle and Acanthamoeba. Nonfilamentous actin in 2 different buffers had similar properties. In a low ionic strength buffer the absence of filaments was confirmed by electron microscopy, ultracentrifugation, and the fluorescence of pyrene-labeled actin. In 0.6 M KI, actin was monomeric by gel filtration. Filamentous actin had similar properties in 2 mM MgCl2 with either 50 mM KC1 or 500 mM KC1. Under all 4 of these conditions, actin required about 1000 min at 25 degrees C for the rheological properties to equilibrate. Under conditions where the oscillation of the rheometer did not affect the mechanical properties, all of the actin preparations had dynamic viscosities that were inverse functions of the frequency and dynamic elasticites that leveled off at low frequencies as expected for viscoelastic solids. For filamentous actin, the values of these parameters were about 2 times higher than for nonfilamentous actin. In shear experiments, both filamentous and nonfilamentous actin exhibited shear rate-dependent yield stresses. When filamentous and nonfilamentous actin structures were disrupted by transient shearing, the dynamic elasticity recovered to 90% in 30 min. Ovalbumin in the low ionic strength buffer also behaved as a viscoelastic material with elasticity and viscosity about 10 times lower than nonfilamentous actin, while cytochrome c behaved as a Newtonian fluid with a viscosity of 0.02 poise.     

Localization of actin messenger RNA during early ascidian development

The spatial distribution of RNA sequences during early development of the ascidian, Styela plicata, was determined by in situ hybridization with poly(U) and cloned DNA probes. Styela eggs and embryos contain three colored cytoplasmic regions of specific morphogenetic fates, the ectoplasm, endoplasm, and myoplasm. These cytoplasmic regions participate in ooplasmic segregation after fertilization and are distributed to different cell lineages during early embryogenesis. n situ hybridization with poly(U) suggests that poly(A)+RNA is unevenly distributed in eggs and embryos, with about 45% in the ectoplasm, 50% in the endoplasm, and only 5% in the myoplasm. In situ hybridization with a histone DNA probe showed that histone RNA sequences were not localized in eggs or embryos and distributed between the three cytoplasmic regions according to their volumes. In situ hybridization with an actin DNA probe showed actin RNA was localized in the myoplasm and ectoplasm of eggs and embryos with about 45% present in the myoplasm, 40% in the ectoplasm, and only 15% in the endoplasm. These results suggest that a large proportion of the egg actin mRNA is localized in the myoplasm, participates in ooplasmic segregation after fertilization, and is differentially distributed to the mesodermal cell lineages during embryogenesis. Analysis of the translation products of egg mRNA suggests that the localized mRNA codes for a cytoplasmic actin isoform.     

Distribution of actin and tropomyosin in Cryptosporidium muris

Actin and tropomyosin of Cryptosporidium muris were localized by immunogold labeling. Two kinds of antibodies for actin labeling were used. The polyclonal antibody to skeletal muscle (chicken back muscle) actin was labeled on the pellicle and cytoplasmic vacuoles of parasites. The feeder organelle has showed a small amount of polyclonal actin antibody labeling as well. Whereas the monoclonal antibody to smooth muscle (chicken gizzard muscle) actin was chiefly labeled on the filamentous cytoplasm of parasites. The apical portion of host gastric epithelial cell cytoplasm was also labeled by smooth muscle actin together. The polyclonal antibody to tropomyosin was much more labeled at C. muris than host cells, so it could be easily identified even with low magnification (×2,000). The tropomyosin was observed along the pellicle, cytoplasmic vacuoles, and around the nucleus also. The skeletal muscle type actin seems to play a role in various cellular functions with tropomyosin in C. muris; on the other hand, the smooth muscle type actin was located mainly on the filamentous cytoplasm and supported the parasites'' firm attachment to host cells. Tropomyosin on the pellicle was thought to be able to stimulate the host as a major antigen through continuous shedding out by the escape of sporozoites or merozoites from their mother cells.     

An antiactin antibody that distinguishes between cytoplasmic and skeletal muscle actins

We elicited antibodies in rabbits to actin purified from body wall muscle of the marine mollusc, Aplysia californica. We found that this antiactin has an unusual specificity: in addition to reacting with the immunogen, it recognizes cytoplasmic vertebrate actins but not myofibrillar actin. Radioimmunoassay showed little or no cross-reaction with actin purified from either chicken gizzard or rabbit skeletal muscle. Immunocytochemical studies with human fibroblasts and L6 myoblasts revealed intense staining of typical cytoplasmic cables. Myofibrils were not stained after treatment of human and frog skeletal muscle with the antibody, although the distribution of immunofluorescence suggested that cytoplasmic actin is associated with membrane systems in the muscle fiber. The antibody may therefore be especially suited for studying the localization of cytoplasmic actin in skeletal muscle cells even in the presence of a great excess of the myofibrillar form.     

Monomeric G‐actin is uniformly distributed in pollen tubes and is rapidly redistributed via cytoplasmic streaming during pollen tube growth

Dynamic assembly and disassembly of the actin cytoskeleton has been implicated in the regulation of pollen germination and subsequent tube growth. It is widely accepted that actin filaments are arrayed into distinct structures within different regions of the pollen tube. Maintenance of the equilibrium between monomeric globular actin (G‐actin) and filamentous actin (F‐actin) is crucial for actin assembly and array construction, and the local concentration of G‐actin thus directly impacts actin assembly. The localization and dynamics of G‐actin in the pollen tube, however, remain to be determined conclusively. To address this question, we created a series of fusion proteins between green fluorescent protein (GFP) and the Arabidopsis reproductive actin ACT11. Expression of a fusion protein with GFP inserted after methionine at position 49 within the DNase I‐binding loop of ACT11 (GFP_Met49–ACT11) rescued the phenotypes in act11 mutants. Consistent with the notion that the majority of actin is in its monomeric form, GFP_Met49–ACT11 and GFP fusion proteins of four other reproductive actins generated with the same strategy do not obviously label filamentous structures. In further support of the functionality of these fusion proteins, we found that they can be incorporated into filamentous structures in jasplakinolide (Jasp)‐treated pollen tubes. Careful observations showed that G‐actin is distributed uniformly in the pollen tube and is rapidly redistributed via cytoplasmic streaming during pollen tube growth. Our study suggests that G‐actin is readily available in the cytoplasm to support continuous actin polymerization during rapid pollen tube growth.     

The cortical actin cytoskeleton of unactivated zebrafish eggs: Spatial organization and distribution of filamentous actin,nonfilamentous actin,and myosin-II

Actin and nonmuscle myosin heavy chain (myosin-II) have been identified and localized in the cortex of unfertilized zebrafish eggs using techniques of SDS-polyacrylamide gel electrophoresis, immunoblotting, and fluorescence microscopy. Whole egg mounts, egg fragments, cryosections, and cortical membrane patches probed with rhodamine phalloidin, fluorescent DNase-I, or anti-actin antibody showed the cortical cytoskeleton to contain two domains of actin: filamentous and nonfilamentous. Filamentous actin was restricted to microplicae and the cytoplasmic face of the plasma membrane where it was organized as an extensive meshwork of interconnecting filaments. The cortical cytoplasm deep to the plasma membrane contained cortical granules and sequestered actin in nonfilamentous form. The cytoplasmic surface (membrane?) of cortical granules displayed an enrichment of nonfilamentous actin. An antibody against human platelet myosin was used to detect myosin-II in whole mounts and egg fragments. Myosin-II colocalized with both filamentous and nonfilamentous actin domains of the cortical cytoskeleton. It was not determined if egg myosin was organized into filaments. Similar to nonfilamentous actin, myosin-II appeared to be concentrated over the surface of cortical granules where staining was in the form of patches and punctate foci. The identification of organized and interconnected domains of filamentous actin, nonfilamentous actin, and myosin-II provides insight into possible functions of these proteins before and after fertilization. © 1996 Wiley-Liss, Inc.     

Association of deoxyribonuclease I with the pointed ends of actin filaments in human red blood cell membrane skeletons

We have characterized the interaction of bovine pancreatic deoxyribonuclease I (DNase I) with the filamentous (F-)actin of red cell membrane skeletons stabilized with phalloidin. The hydrolysis of [3H]DNA was used to assay DNase I. We found that DNase I bound to a homogenous class of approximately equal to 2.4 X 10(4) sites/skeleton with an association rate constant of approximately 1 X 10(6) M-1 S-1 and a KD of 1.9 X 10(-9) M at 20 degrees C. Phalloidin lowered the dissociation constant by approximately 1 order of magnitude. The DNase I which sedimented with the skeletons was catalytically inactive but could be reactivated by dissociation from the actin. Actin and DNA bound to DNase I in a mutually exclusive fashion without formation of a ternary complex. Phalloidin-treated red cell F-actin resembled rabbit muscle G-actin in all respects tested. Since the DNase I binding capacity of the skeletons corresponded to the number of actin protofilaments previously estimated by other methods, it seemed likely that the enzyme binding site was confined to one end of the filament. We confirmed this premise by showing that elongating the red cell filaments with rabbit muscle actin monomers did not appreciably add to their capacity to bind or inhibit DNase I. Saturation of skeletons with cytochalasin D or gelsolin, avid ligands for the barbed end of actin filaments, did not reduce their binding of DNase I. Furthermore, neither cytochalasin D nor DNase I alone blocked all of the sites for addition of monomeric pyrene-labeled rabbit muscle G-actin to phalloidin-treated skeletons; however, a combination of the two agents did so. In the presence of phalloidin, the polymerization of 300 nM pyrenyl actin on nuclei constructed from 5 nM gelsolin and 25 nM rabbit muscle G-actin was completely inhibited by 35 nM DNase I but not by 35 nM cytochalasin D. We conclude that DNase I associates uniquely with and caps the pointed (slow-growing or negative) end of F-actin. These results imply that the amino-terminal, DNase I-binding domain of the actin protomer is oriented toward the pointed end and is buried along the length of the actin filament.     

Proteolysis of acidic calponin by mu-calpain

Acidic calponin is an actin binding protein expressed in smooth muscle and brain. Although the role of smooth muscle calponin (basic calponin) has been well studied, few studies have been performed on acidic calponin. In the present study, we demonstrated that acidic calponin binds to filamentous actin, but not monomeric actin. A co-sedimentation assay indicated that acidic calponin binds to actin with an apparent binding constant of 4 x 10(5) M(-1). In the presence of an excess amount of calmodulin, the binding of acidic calponin to actin was inhibited. The binding of acidic calponin to calmodulin was Ca(2+)-dependent with K(d) of 31 microM. We next investigated whether or not acidic calponin could be a substrate for mu-calpain in vitro, since it has been shown that basic calponin is cleaved by mu-calpain. The results showed that acidic calponin was also cleaved by mu-calpain. Neither the proteolytic pattern nor velocity of acidic calponin was different in the absence or presence of calmodulin. When acidic calponin had bound to actin, however, the susceptibility of the acidic calponin to mu-calpain was significantly reduced, which was reversed by the addition of calmodulin. Our results suggest that acidic calponin might be involved in the mu-calpain-regulated actin cytoskeleton.     

Microinjection of pollen‐specific actin‐depolymerizing factor, ZmADF1, reorientates F‐actin strands in Tradescantia stamen hair cells

Maize actin-depolymerizing factor (ADF) binds both monomeric and filamentous actin and increases actin dynamics in vitro. To test its effects in vivo, recombinant pollen ADF1 was expressed in bacteria and microinjected into Tradescantia stamen hair cells. Initially, all cytoplasmic streaming ceased and the central, longitudinal transvacuolar strands were disrupted. After 20–45 min, streaming resumed but in the form of conspicuous transverse pathways of movement in the cortex. Staining the actin filaments by a second injection of fluorescein-conjugated phalloidin showed that the longitudinal actin cables seen in controls had been replaced by a thickening of the transverse cortical arrays, whose orientation matched the new pattern of streaming. Microinjection of rhodamine–tubulin confirmed that the microtubules also formed a transverse cortical array and it is suggested that the spatial cues for re-modelling the actin after ADF1 injection may be provided by the microtubular system.     

Dependence of collagen remodelling on α-smooth muscle actin expression by fibroblasts

To study the relation between expression of the putative myofibroblast marker α-smooth muscle actin and the remodelling of extracellular matrix, immunocytochemical, gel electrophoresis, and collagen gel contraction studies were performed on two human fibroblast subtypes. Double immunolabelling for total actins and α-smooth muscle (sm) actin as well as affinity labelling of filamentous and monomeric actins in gingival fibroblasts demonstrated that α-sm was colocalized in stress fibres and in regions with high levels of monomeric actin throughout the cytoplasm. α-sm comprised up to 14% of total cellular actin as assessed by 2D gel electrophoresis. Thirteen different gingival and seven different periodontal ligament fibroblast lines constitutively expressed on α-sm actin. These cells exhibited up to 60% inter-line variations of fluorescence due to α-sm actin and up to 70% and 45% inter-line variation in the rate of collagen gel contraction. Quantitative, single cell fluorimetry of α-sm actin immunoreactivity demonstrated a linear relation between gel contraction and α-sm actin (correlation coefficients of 0.71 for gingival and 0.61 for periodontal ligament cells), but there was no detectable relationship between total actin content and gel contraction. In contrast, flow cytometry demonstrated that 99% of the total gated cells from cell lines exhibiting rapid gel contraction showed α-sm actin staining above background fluorescence as compared to only 35% of cells with slow rates of gel contraction. Contracting collagen gels stained with FITC-phalloidin showed cells with well-developed stress fibres that were progressively more compact and elongated during the time of maximal gel contraction. To examine the dependence of gel contraction on assembly of monomeric actin into actin filaments, cells were electroporated in the presence of phalloidin or cytochalasin D. Collagen gels exhibited up to 100% inhibition of gel contraction that was dose-dependent. Gel contraction was inhibited 93% by electroinjection of cells with α-sm actin antibody prior to incubation, but the antibody did not inhibit actin assembly after attachment and spreading on substrates. These data indicate that gel contraction is dependent on α-sm actin expression and that α-sm actin is a functional marker for a fibroblast subtype that rapidly remodels the extracellular matrix. © 1994 wiley-Liss, Inc.     

Actin-filled nuclear invaginations indicate degree of cell de-differentiation

For years the existence of nuclear actin has been heavily debated, but recent data have clearly demonstrated that actin, as well as actin-binding proteins (ABPs), are located in the nucleus. We examined live EGFP-actin-expressing cells using confocal microscopy and saw the presence of structures strongly resembling actin filaments in the nuclei of MDA-MB-231 human mammary epithelial tumor cells. Many nuclei had more than one of these filamentous structures, some of which appeared to cross the entire nucleus. Extensive analysis, including fluorescence recovery after photobleaching (FRAP), showed that all EGFP-actin in the nucleus is monomeric (G-actin) rather than filamentous (F-actin) and that the apparent filaments seen in the nucleus are invaginations of cytoplasmic monomeric actin. Immunolocalization of nuclear pore complex proteins shows that similar invaginations are seen in cells that are not overexpressing EGFP-actin. To determine whether there is a correlation between increased levels of invagination in the cell nuclei and the state of de-differentiation of the cell, we examined a variety of cell types, including live Xenopus embryonic cells. Cells that were highly de-differentiated, or cancerous, had an increased incidence of invagination, while cells that were differentiated had few nuclear invaginations. The nuclei of embryonic cells that were not yet differentiated underwent multiple shape changes throughout interphase, and demonstrated numerous transient invaginations of varying sizes and shapes. Although the function of these actin-filled invaginations remains speculative, their presence correlates with cells that have increased levels of nuclear activity.     

Cytoplasmic nonpolysomal messenger ribonucleoprotein containing actin messenger RNA in chicken embryonic muscles.

Cytoplasmic nonpolysomal mRNAs have been isolated in the form of 16-40S ribonucleoprotein particles from the postribosomal supernatant of 14-day-old chick embryonic muscles. An 8-20S RNA fraction isolated from these particles directs the synthesis of actin in a wheat germ embryo S-30 system, as judged by copurification of the products with chicken muscle actin by repeated cycles of G- to F-actin transformation; mobilities of the purified product on sodium dodecyl sulfate-polyacrylamide gels and urea gels; and analysis of the CNBr-cleaved peptides. The 16-40S particles have a buoyant density of 1.4 g/cm3 which corresponds to an RNA/protein ratio of 1:3. They do not contain detectable levels of ribosomal subunits, as judged by the absence of typical ribosomal proteins in the range of 15,000-30,000. They contain at least eight distinct polypeptide species in the molecular weight range of 44,000-100,000, including a prominent 44,000 species. The presence of these particles suggests that they may have a role in the regulation of translation in developing muscles.     

Poly(L-proline)-binding proteins from chick embryos are a profilin and a profilactin

Two poly(L-proline)-binding proteins (PBP-1 and PBP-2) were purified from chick embryos by using a poly(L-proline)-agarose column. PBP-1 was composed of two different polypeptides (molecular masses of 42 kDa and 15 kDa). The molar ratio of the two proteins in the complex was 1:1. The other poly(L-proline)-binding protein, PBP-2, was the 15-kDa protein itself. The 42-kDa protein was confirmed to be an actin from the amino acid composition, by immunochemical evidence and by its ability to self-polymerize. In addition, the 42 + 15-kDa protein complex (PBP-1) inhibited DNase I, just as a monomeric actin did. The amino acid composition of the 15-kDa protein was similar to that of mammalian profilin and it inhibited the salt-induced polymerization of rabbit skeletal muscle actin. Therefore, we conclude that the two poly(L-proline)-binding proteins from the chick embryo are a profilactin and a profilin in chick embryo. The ability of profilactin to bind poly(L-proline) must be due to profilin itself, because the profilin has a greater affinity for poly(L-proline) than does profilactin. Additionally, both the monomeric and filamentous actin from rabbit skeletal muscle have no affinity for poly(L-proline).