In Vitro Studies

Latex bead phagosomes can be purified by flotation in a sucrose gradient under different conditions, for example, different time-points. This allows the study ofthe influence of proteins and lipids present on the phagosomal membrane in actin binding or assembly (Figure 8.3). In addition, LBPs can be incubated with different

Latex Allergy Ppt

Figure 8.3 In vitro assays to monitor actin in latex bead phagosomes (LBPs). (a) Scheme showing the purification of LBPs from macrophages for biochemical analysis. Briefly, latex beads are internalized and then isolated using its property offlotation in a sucrose gradient. Then isolated LBPs are used for the assays described below. (b) Assays to assess the role of actin in phagosome biology. (1) Actin nucleation assay, with this approach LBPs are used in presence of ATP and the nucleation of actin filaments around the phagosome measured [35]. (2) Actin binding, actin is polymerized and then phagosomes incubated with this actin "network." Filamentous actin stabilized and labeled with rhodamine-

Figure 8.3 In vitro assays to monitor actin in latex bead phagosomes (LBPs). (a) Scheme showing the purification of LBPs from macrophages for biochemical analysis. Briefly, latex beads are internalized and then isolated using its property offlotation in a sucrose gradient. Then isolated LBPs are used for the assays described below. (b) Assays to assess the role of actin in phagosome biology. (1) Actin nucleation assay, with this approach LBPs are used in presence of ATP and the nucleation of actin filaments around the phagosome measured [35]. (2) Actin binding, actin is polymerized and then phagosomes incubated with this actin "network." Filamentous actin stabilized and labeled with rhodamine-

phalloidin is perfused into the chamber and then a solution containing purified phagosomes are also perfused into the chamber and incubated for 20 min. Binding of phagosomes is analyzed by fluorescence microscopy. The bound phagosomes are counted by eye under the different conditions [24]. (3) Phagosome aggregation on phagosomal membrane-assembled actin [39]. Phagosomes are mixed with macrophage cytosol and rhodamine-G-actin and incubated in the presence of an ATP-regenerating system at 37 °C in a sealed chamber. Individual phagosomes and clusters are then quantified at different time-points using confocal microscopy.

components, for example, cytosol, ATP, allowing the outcome to be evaluated in different processes (Figure 8.3).

8.2.1.2.1 LBP Actin-Binding Assay The LBP actin-binding assay monitors the binding of LBPs with fluorescent beads to rhodamine-F-actin bound to glass [24]. In contrast to actin assembly on phagosome membranes, this binding requires cytosolic factors. At least three factors involved in phagosome binding to F-actin have been detected:

(i) a factor with ATPase activity, which has been identified as myosin Va, (ii) a high molecular weight factor (about 600 kDa), but without ATPase activity, possibly the actin-binding protein filamin, (iii) an unknown inhibitory factor with low molecular weight (about 30 kDa), which blocks binding activities ofboth myosin Va and filamin. This suggests that inhibitory activity may be involved in the regulation of the association of the filamin and myosin Va, which is responsible for the stimulation of F-actin-phagosome interaction with the phagosome surface. Recently the Kuz-netsov group found that GTP and phosphatidylinositol bisphosphate (PIP2) are also involved in the regulation of phagosome binding to F-actin. In agreement with these data many GTPases and PIP2-regulated proteins have been implicated in the modulation of F-actin (see above). An intriguing observation in the comparison of the LBP actin-binding assay with the actin assembly assay is that the factors that stimulate binding tend to inhibit assembly and vice versa. For example, arachidonic acid (AA), PIP2, sphingomyelin (SM) and sphingosine-1-phosphate (S1P) stimulated actin nucleation but inhibited binding of phagosomes to F-actin in vitro in the absence of cytosolic factors and ATP. Furthermore, binding assays in the presence of functional antibodies against different actin-associated proteins revealed that proteins known to activate actin nucleation on LBP phagosomal membranes in macrophages (ezrin, N-WASP, see below) downregulate the F-actin-binding capability (Hoffmann and Kuznetsov, unpublished observations).

8.2.1.2.2 In Vitro Assembly of Actin by Phagosomes The LBP actin assembly assay monitors the polymerization of rhodamine-actin by LBP using light microscopy. Key players involved in this process are ezrin and/or its close homolog moesin from the ezrin radixin moesin (ERM) protein family [35]. Recent data by our group argue that ezrin on phagosomes can bind and activate N-WASP, which may in turn switch on the actin nucleating complex Arp2/3. In vitro experiments suggest that ezrin oligomers in solution stimulate the ability of N-WASP to activate actin assembly via the Arp2/3 complex (Marion and Griffiths, unpublished observations). The synthesis of the phosphoinositides phosphatidylinositol-4-phosphate (PIP) and PIP2 by LBP-bound phosphatidylinositol kinases are also essential for actin assembly and part of the role of PIP2 is to bind to the N-terminal ERMAD domain of ezrin [47]. Both the ERM proteins and N-WASP bind PIP2 and this interaction facilitates activation of both proteins. Activation of N-WASP exposes the VCA domain (verprolin, homology, cofilin homology and the acidic region), which in turn activates the Arp2/3 molecule to enhance the nucleation of actin [48]. As for all membrane systems, it is still not clear how the ERM protein, N-WASP, PIP2 and other molecules interact in the LBP actin assembly.

Surprisingly, actin assembly occurs on LBPs in the absence of GTP since this molecule has no effect on this process [35]. This argues that the initial nucleation of actin assembly is upstream of GTPases that are expected to play a key role in regulating the signaling processes on phagosomes. Indeed many studies have shown that isolated phagosomes from many species contain many GTPases, including the actin regulatory switches Rho, Rac, Cdc42 (inhibitors of which have no effect on LBP actin assembly (unpublished data)). In most studies on membrane-associated actin assembly the Rho family proteins are considered to be upstream of membrane-dependent actin polymerization. However, evidence supports the notion that the ERM proteins can be upstream of the Rho family GTPases in a complex signaling network [49]. For example, ezrin binds and activates Dbl, the GEF for Cdc42 [50]. Indeed, most of these membrane-bound regulators are found on phagosomes. In addition, most of these proteins have multiple interaction partners, which makes it increasingly difficult to understand how the different machineries operate at the systems level.

With respect to the assembly of actin by phagosomes we have identified a number of key players using the in vitro assay. Gelsolin was also found to stimulate this process [51] while an antibody against profilin inhibited it (Defacque and Griffiths, unpublished data). LBPs have adenylate cyclase that can synthesize cyclic AMP (cAMP) from ATP. cAMP inhibits actin assembly by phagosomes, at least in part by activating phagosomal PKA [52]. Protein kinase A (PKA) is known to phosphorylate serine 66 on ezrin in some cell types and serine 10 of its close homolog merlin [53]. In fact, PKA phosphorylation of ezrin is associated with inhibition of actin-dependent processes such as cell motility [54], although it has not been directly shown whether the phosphorylation of ezrin by the activated PKA is part of the mechanism whereby cAMP inhibits phagosome actin assembly.

The concentration of ATP used in the LBP actin assembly assay plays an important role in the final outcome. The standard assay works best at low ATP (0.2 mM), which allows around 10-30% of LBPs to assemble actin. In contrast, physiological ATP (5 mM) blocks assembly. Since LBPs make significant cAMP with 5 mM but not with 0.2 mM this provides one mechanism by which cAMP via PKA can inhibit the system at high ATP. The addition of 1 mM ATP to LBPs from J774 macrophages leads to the phosphorylation of a large number of phagosomal proteins by phagosome-associated kinases [55], a multitude of which are found in the different phagosome pro-teomes [56]. With respect to the actin assembly process it is also clear that an important role of the ATP is to be used by kinases, such as phosphatidylinositol (PI), sphingosine (Sph) and ceramide (Cer) to their phosphorylated counterparts PIP, PIP2, S1P and ceramide-1-phosphate (Cer-1P), respectively.

The identification of a key role for PIP and PIP2 opened the door for a detailed analysis of the lipids that interconnect with these phosphoinositides in the LBP actin assembly process, as well as sphingolipids and fatty acids [57]. When PIP or PIP2 are incorporated into the LBP membrane they stimulate actin assembly at low and high ATP. In contrast, other lipids such as phosphatidylcholine (PC), the polyunsaturated fatty acids eicosapentanoic acid (EPA) and docosahexanoic acid (DHA) inhibit the system in both conditions. AA is highly stimulatory under both ATP regimes. More intriguing is the effect of the sphingolipids, especially Sph and its downstream product S1P. Whereas Sph inhibits the system at high ATP and stimulates at low ATP, the opposite pattern is observed when S1P is added to LBPs. This lipid stimulates actin assembly at high ATP and inhibits it at low ATP [57]. The same pattern is seen with Cer and Cer-1P.

Our recent studies led us to an unexpected role of the phosphorylated lipids PIP, PIP2, S1P, Cer-1P and phosphatidic acid (PA), which are all stimulatory in the LBP actin assay at high ATP. The incorporation of these lipids into the phagosomal membrane in the presence of ATP and ADP led to the translocation of ADP, but not ATP into the LBP lumen, as shown using radioactive nucleotides and HPLC. In the lumen the ADP becomes converted to ATP by an adenylate kinase (AK). An inhibitor of AK (Ap2P5) prevented the luminal ADP to ATP conversion. Although the AK remains to be determined, one of the five known AKs, AK2, has been detected in two LBP proteome analyses (M. Desjardins, personal communication; [58]).

A number of experiments suggested that the luminal ATP was able to stimulate the actin assembly on the cytoplasmic side of the LBP. For example LBPs prepared using beads conjugated with apyrase (that degrades ATP and ADP) on their surface were deficient in actin assembly.

The identification of a well-characterized ATP-binding transmembrane receptor, P2X7R, in the J774 macrophage LBP proteome (M. Desjardins, personal communication) led us to focus on this receptor. P2X7R is a cationic channel that is able to bind via its cytoplasmic domain to a complex of 12 proteins, including actin, a-actinin, supervillin and a phosphatidylinositol 4-kinase (PI4K) and 10 other proteins [59]. Whereas LBP prepared from wild-type P2X7R-positive bone marrow macrophages could assemble actin and was stimulated by S1P or PIP at high ATP, the equivalent LBP from P2X7R-knockout mice macrophages failed to be stimulated by these lipids. This argues for a model in which PIP and S1P induce ATP accumulation in the LBP lumen. This ATP then activates the P2X7R that signals downstream of the phagosome actin assembly machinery. Since PI4K interacts with the P2X7R, this kinase could also be an integral player of the system (Kuehnel et al., 2009). Consistent with this idea, the synthesis of PIP and PIP2 is essential for actin assembly in LBPs [47].

These recent experiments define new mechanisms but open even more questions to address with respect to the actin assembly process on LBPs. These include: (i) the identification of the ADP transporter and the mechanism of regulation by phos-phorylated lipids, (ii) the identification of the AK and the mechanism that allows this protein to cross the phagosomal membrane and (iii) the molecular connection between the P2X7R and the PIP2-ezrin-N-WASP machinery that assembles actin. Each player in this process has its own complexity. For example, PIP2 can bind to a myriad ofactin-binding proteins, including many that are established to be bone fide phagosome constituents such as ezrin, N-WASP, gelsolin, profilin, a-actinin, cofilin and some capping proteins such as CapZ [60]. Many of these proteins have multiple binding partners. For example, as mentioned above, the ERM proteins have multiple interactions both upstream and downstream of Rho, Rac and Cdc42 [49].

It is important to note that these complex processes of actin assembly and actin binding to phagosomes must depend on proteins and lipids that are found in the phagosome proteomics and lipidomics. The most recent data for the J774 macrophage phagosome proteome using more sensitive MS analysis has identified up to 2000 different proteins in the system (M. Desjardins, personal communication). This is complicated by the dynamic changes in phagosomes, since as they mature in cells their protein pattern, and their ability to assemble actin, changes dramatically [9, 35, 61, 62]. On the other hand, an ongoing analysis of the lipids present in LBPs from J774 macrophages by our collaborators, Jos Brouwers and Bernd Helms (University of Utrecht), and in LBPs from Dictyostelium (Soldati and

Brugger, personal communication) makes it clear that the lipids are also complex. Moreover, just like the proteins, the phagosome lipid pattern also changes significantly during phagosome maturation (Brouwers and Helms, personal communication). In both analyses many hundreds of different lipid species are present. The ongoing proteomic and Western blotting analyses show that many of the enzymes that interconvert lipids are detected on phagosomes and our studies show that many of these enzymes are active on phagosomes under different conditions. So in one way the lipids are even more complex than proteins. While proteins can be posttranslationally phosphorylated by kinases, the lipids can be newly synthesized and degraded or interconverted into other lipids by enzymatic machinery in the phagosome itself.

From all these studies described above, it is now becoming clear that phagosomes are not passive vesicles that only acquire proteins and lipids via vesicular transport, or (for proteins) directly from the cytoplasm, and lose these components by recycling or dissociation. These organelles are also themselves capable of a plethora of biochemical activities, for example the enzymatic synthesis of different lipids. Even in a single cell, functionally different populations of phagosomes are usually observed. This led to the concept of phagosome individuality [63]. Recent innovative analyses towards this goal are now beginning to draw the maps of molecular interactions that are possible for the many hundreds of proteins that have been detected in phagosome proteomic studies [56, 61, 64-67]. A dynamic description, including a predictive model of the interactions of lipids linked to PIP2 has recently been initiated by our group. This includes a systems level analysis ofthese lipids and other molecules on phagosomal membranes (Kuehnel et al., 2008).

8.2.1.2.3 Phagosome Aggregation on Phagosomal Membrane-Assembled Actin Initial studies showed that a number of actin-interfering reagents, such as cytochalasin D and latrunculin A, inhibited in vitro fusion of LBPs with early endosomes (EEs) and late endocytic organelles (LEOs). However, two actin-binding proteins stimulate LBP fusion with EEs/LEOs (Jahraus et al. [34]). The first is a gelsolin fragment that can sever actin and bind to the free barbed ends of actin. The second positive effector of fusion is thymosin P4 (Tp4). To study in more detail the in vitro organization of LBPs, EEs and LEOs and F-actin, a confocal-based study using rhodamine-actin was developed to simultaneously visualize the interactions between the actin cytoskeleton assembled from cytosol and selectively labeled LBPs and/or EEs/LEOs (Figure 8.3). When postnuclear supernatants (PNS) containing fluorescent LBPs or fluorescent endosomes are added to the cytosol, a small number of these organelles were associated with distinct dots of labeled actin and these organelles were clustered with a concomitant formation of a visible actin network. Both LBPs and endosomes aligned along these actin fibers and formed large clusters at the junction points where many actin fibers crossed. This suggests that cellular membrane organelles can directly control an organized assembly of an actin network. This clustering of organelles along actin fibers is Tp4-dependent and under ATP depletion actin polymerizes but there is no clustering. Interestingly, phagosome clustering is actin-dependent but MT-independent [39].

Alternative models have been developed to explain how actin filaments nucleated from the surface of a membrane organelle could "attract" other bound organelles toward it. First, because the barbed ends of actin are close to the membrane surface, the polarity of the filaments is such that most myosins bound to the organelles would carry their cargo toward the nucleating organelle in the presence of ATP. In a second model, both fusion partners nucleate actin and the double-headed myosin II is proposed to crosslink and slide the filament bundles in opposite directions. This might also facilitate membrane organelle aggregation, leading to docking and fusion [39].

Was this article helpful?

0 0

Post a comment