Phosphatidylserine exposure promotes increased adhesion in Dictyostelium Copine A mutants

The phospholipid phosphatidylserine (PS) is a key signaling molecule and binding partner for many intracellular proteins. PS is normally found on the inner surface of the cell membrane, but PS can be flipped to the outer surface in a process called PS exposure. PS exposure is important in many cell functions, yet the mechanisms that control PS exposure have not been extensively studied. Copines (Cpn), found in most eukaryotic organisms, make up a family of calcium-dependent phospholipid binding proteins. In Dictyostelium, which has six copine genes, CpnA strongly binds to PS and translocates from the cytosol to the plasma membrane in response to a rise in calcium. Cells lacking the cpnA gene (cpnA-) have defects in adhesion, chemotaxis, membrane trafficking, and cytokinesis. In this study we used both flow cytometry and fluorescent microscopy to show that cpnA- cells have increased adhesion to beads and bacteria and that the increased adhesion was not due to changes in the actin cytoskeleton or cell surface proteins. We found that cpnA- cells bound higher amounts of Annexin V, a PS binding protein, than parental cells and showed that unlabeled Annexin V reduced the increased cell adhesion property of cpnA- cells. We also found that cpnA- cells were more sensitive to Polybia-MP1, which binds to external PS and induces cell lysis. Overall, this suggests that cpnA- cells have increased PS exposure and this property contributes to the increased cell adhesion of cpnA- cells. We conclude that CpnA has a role in the regulation of plasma membrane lipid composition and may act as a negative regulator of PS exposure.


Introduction
Biological membranes are composed of different types of phospholipids that are not equally distributed between the two leaflets of the lipid bilayer; the outer leaflet of the plasma membrane consists mainly of phosphatidylcholine and sphingomyelin and the inner leaflet consists mainly of phosphatidylethanolamine, phosphatidylserine, and phosphatidylinositol [1]. The phospholipid phosphatidylserine (PS) is a key signaling molecule and binding partner for

Cell strains and culture
The parental Dictyostelium axenic cell strain, NC4A2 [18], was grown on plastic Petri dishes in HL-5 media (0.75% proteose peptone #2, 0.75% proteose peptone #3, 0.5% yeast extract, 1% glucose, 2.5 mM Na 2 HPO 4 , and 8.8 mM KH 2 PO 4 , pH 6.5) at 18˚C. To prevent bacterial contamination, 60 U/mL of penicillin-streptomycin (Sigma-Aldrich, P4333) was added to the HL-5 media. The cpnA knockout cell line, cpnA -, was created previously by using homologous recombination to replace the cpnA gene with the blasticidin resistant gene (bsr) and grown in HL-5 containing 10 μg/mL blasticidin (InvivoGen, ant-bl-1) [14]. The bacterial strain, Klebsiella aerogenes, expressing GFP was obtained from the Dicty Stock Center and grown in Luria broth (LB) with ampicillin (100 μg/mL) in a shaking suspension culture at 250 rpm for 18 hours at 37˚C [19]. The cpnA cDNA was subcloned into the SacI site of the pTX-GFP plasmid for the expression of CpnA with a GFP tag at the N-terminus [12,20]. The plasmid was transformed into NC4A2 and cpnA -Dictyostelium cells by electroporation and cells were selected with G418 (50 mg/mL). Expression of GFP-tagged CpnA proteins were verified by western blot with an antibody to GFP [12].

Phagocytosis assays
Dictyostelium NC4A2 and cpnAcells were harvested by centrifugation (437 x g, 5 mins, 4˚C) and 2x10 6 cells/mL were incubated with either GFP-labeled K. aerogenes at 5x10 7 cells/mL or 1 μm yellow-green large FluoSphere carboxylate modified microspheres (Thermo-Fisher, F8823) at 3.64x10 10 beads/mL in 2.5 mL of HL-5 media. Cells were allowed to phagocytose for 30 minutes in a shaking suspension at 180 rpm in a 50 mL flask at room temperature. For bead phagocytosis, 100 μL of cells were transferred from the flask into a microcentrifuge tube at 5-minute timepoints and cells were washed free of unbound beads by centrifugation at 437 x g for 5 minutes at 4˚C. The cells were resuspended in ice-cold Sorensen's buffer (0.2 M NaH 2 PO 4 , 0.2 M Na 2 HPO 4 , pH 6.5), washed two times, fixed in 3.7% formaldehyde in Sorensen's buffer, and again washed two times with Sorensen's buffer. Cell fluorescence was measured with flow cytometry. To remove beads from the surface of cells before fixation, cells were washed in 5 mM sodium azide (Sigma-Aldrich, 26628-  in Sorensen's buffer twice and then in Sorensen's buffer alone. Mean cell fluorescence data from three trials were averaged. Significant differences at each timepoint between NC4A2 and cpnAcells with and without sodium azide washes were analyzed using a repeated measures ANOVA and post hoc Tukey comparisons. For fluorescence microscopy, NC4A2 and cpnAcells were allowed to phagocytose and washed with and without sodium azide as described above. Cells were plated on glass coverslips and allowed to settle for 15 minutes. Cells were then fixed in cold 3.7% formaldehyde (32% formaldehyde in aqueous solution, Electron Microscopy Sciences, 15714) diluted in methanol for 10 minutes at -20˚C. Fixed cells were washed three times with ice-cold Sorensen's Buffer with 5 minutes in between washes on coverslips. Coverslips were placed cell side down onto glass slides containing 50% glycerol. Cells were imaged with a Leica DMi 8 microscope. For bacteria phagocytosis, the number of bacteria associated with~900 individual Dictyostelium cells were counted and averaged per trial per sample. Data from the 30-minute timepoint, five trials of buffer alone and three trials of buffer with sodium azide, were averaged and analyzed for significant differences between NC4A2 and cpnAcells using a Student's ttest.

Bead adhesion assays
In 50 mL flasks containing 2.5 mL HL-5 media, NC4A2 and cpnAcells were incubated with 0.1 mg/mL Latrunculin A (LatA)(Cayman Chemical, NC0673768) in DMSO or 52.7 μL of DMSO for 30 minutes in a shaking suspension at 180 rpm. After 30 mins, 1 μm beads (3.64x1010 beads/mL) were added to the flasks. After 15 minutes, cells were centrifuged at 437 x g and washed twice with Sorensen's buffer containing sodium azide or Sorensen's Buffer alone. Cells were fixed in 3.7% formaldehyde in Sorensen's Buffer. After fixation and washing cells twice in Sorensen's buffer, flow cytometry was used to measure mean cell fluorescence. Mean cell fluorescence data for each cell type and condition were normalized to the average mean cell fluorescence within each trial. Normalized data from four trials were averaged and analyzed for significant differences using a Student's t-test. For fluorescence microscopy, NC4A2 and cpnAcells were allowed to adhere to beads for 30 minutes as described above and then placed on coverslips and fixed in 3.7% formaldehyde in methanol and imaged with a Leica DMi 8 microscope.
For proteinase K bead adhesion assays, NC4A2 and cpnAcells were incubated with LatA for 30 minutes and then 0 μg/mL, 100 μg/mL, and 500 μg/mL of proteinase K for 15 minutes in a shaking suspension. After 15 minutes, 1 μm beads (3.64x1010 beads/mL) were added to the flasks and allowed to adhere to the surface of the cells for 15 minutes at 180 rpm. Cell samples (100 μL) were washed three times with Sorensen's buffer by centrifugation at 437 x g for 5 minutes at 4˚C. Cells were fixed in 3.7% formaldehyde in Sorensen's buffer, washed twice in Sorensen's buffer, and analyzed by flow cytometry. Mean cell fluorescence data were normalized to the average mean cell fluorescence for all cell types and conditions within each trial. Normalized data from three trials of each cell type were averaged and analyzed for significant differences between parental and cpnAcells using an ANOVA and post hoc Tukey comparisons.
Cell samples treated with proteinase K were also analyzed by western blot. Cells were incubated with or without 500 μg/mL proteinase K for 30 minutes in a shaking suspension, washed three times with Sorensen's buffer, and counted using a hemocytometer for equivalent sample loading. Cells (3x10 6 cells/mL) were resuspended in 20 μL of 4X sample buffer with 2 mM Phenylmethylsulfonyl fluoride (PMSF) (VWR, 101079-016) and incubated at 95˚C for 2 minutes. For Western blotting, a 10% polyacrylamide gel with whole cell protein samples was transferred to a polyvinylidene fluoride (PVDF) membrane. The membrane was cut in half and incubated with blocking buffer (5% dry milk, 0.5% Tween-20 in PBS) for 30 minutes at room temperature. The top half of the membrane was incubated with rabbit polyclonal anti-SibA antibody (1:1000)(Geneva Antibody Facility) [21] and the bottom half was incubated with mouse anti-actin antibody (1:2000) (Santa Cruz, SC4778) in blocking buffer overnight at 4˚C. The membranes were washed in 0.5% Tween-20 in PBS and the top membrane was incubated with anti-rabbit HRP-conjugated antibody (1:15000) in blocking buffer and the bottom membrane was incubated with anti-mouse HRP-conjugated antibody(1:15000) for 2 hr at room temperature. The membranes were washed with 0.5% Tween in PBS three times and a chemiluminescence kit (Michigan Diagnostics, PWPD02-16) and BioRad imaging system was used to image the blots.

Annexin-V assays
NC4A2 and cpnAcells were centrifuged at 437 x g for 5 minutes at 4˚C and then resuspended to 1x10 5 cells/mL in 1X binding buffer (0.01 M HEPES, 0.14 M NaCl, 2.5 mM CaCl 2 ). For a positive control, 3 μM of calcium ionophore (Sigma-Aldrich, C7522) was added to 0.5 mL NC4A2 and cpnAcells and allowed to incubate at room temperature for 10 minutes. One drop of Annexin V-APC (Invitrogen, R37176) was added to 0.5 mL of cells treated with ionophore or buffer alone and allowed to incubate for 15 minutes. Live cells were analyzed via flow cytometry.
In a 25 mL flask, 2x10 6 cells/mL NC4A2 and cpnAcells were incubated with 0.1 mg/mL LatA for 30 minutes in a shaking suspension at 180 rpm. Unlabeled Annexin-V (0.1 μg/mL, BD Pharmagen, 556416) or BSA (0.1 μg/mL) was added to each flask with 1X binding buffer. After 15 minutes, 1 μm beads (3.64x1010 beads/mL) were added to the flasks for 15 minutes. Cell samples (100 μL) were removed from the flasks and centrifuged at 437 x g for 5 minutes at 4˚C. Samples were washed three times with 1X binding buffer and fixed in 3.7% formaldehyde in 1X binding buffer. Fixed cells were washed twice with 1X binding buffer and analyzed by flow cytometry. Mean cell fluorescence was normalized to the average fluorescence of all samples within each trial. Normalized data from three trials was analyzed for significant differences using an ANOVA and post hoc Tukey comparisons.

Polybia-MP1 cell death assay
In 25 mL flasks containing 1 mL of HL-5 media, 2x10 6 cells/mL of NC4A2 and cpnAcells, and the same cells expressing GFP-CpnA were incubated with different concentrations (0-5 μM) of Polybia-MP1 (BACHEM, 4099795) for 30 minutes in a shaking suspension at 180 rpm. After 30 minutes, three 100μL cell samples for each cell type and concentration were taken out of the flasks and cells were counted using a hemocytometer. Cell counts at each concentration of Polybia-MP1 (1-5 μM) were normalized to the cell counts for the 0 μM Polybia-MP1 sample for each cell type in each trial. Normalized data from three trials for each cell type were averaged and analyzed for significant differences using an ANOVA and post hoc Tukey comparisons. For Differential Interference Contrast microscopy, cells were plated on glass bottom dishes and allowed to adhere for 15 minutes in HL-5 media. After 15 minutes, the media was replaced with 4 μM Polybia-MP1 in HL-5 media and images were taken every 5 minutes for 15 minutes.

Flow cytometry
Cell samples were analyzed on a Beckman Coulter CytoFlex Flow Cytometer with the 488 nm laser and FITC detector (525/40) for beads, and with the 638 nm laser and APC detector (660/ 20) for Annexin V-APC. Forward and side scatter were used to gate for cells and beads. Gated cell events were analyzed for mean cell fluorescence and 10,000 gated events were recorded for each sample.

Statistical analysis
A repeated measures ANOVA with post hoc Tukey comparisons (R version 3.6.1) was used to analyze significant mean differences for cell populations sampled over time. An ANOVA and post hoc Tukey comparisons (R version 3.6.1) was used to analyze significant differences for cell populations with multiple treatments. A two-sample assuming unequal variances onetailed t-test (Excel version 16.43) was used to analyze significant mean differences between two cell populations and/or treatments. � indicates p-value<0.05, ns indicates no significant difference, and error bars represent standard error of the mean.

cpnAcells have increased adhesion to both beads and bacteria
To determine if cells lacking the cpnA gene exhibited any defects in phagocytosis, we performed phagocytosis assays with both the parental NC4A2 cells and a previously made cpnAcell line. Cells were incubated with 1 μm fluorescent beads over a 30-minute time period and cell samples were analyzed by flow cytometry at 5-minute intervals (Fig 1). cpnAcells had significantly more bead fluorescence compared to the parental cells at all timepoints ( Fig 1A, solid shapes). However, because cpnAcells are more adherent to surfaces [13], we investigated whether the increased fluorescence in the cpnAcells was due to increased adhesion of beads to the outside surface of the cells rather than increased phagocytosis. To distinguish adhered beads from phagocytosed beads, we washed the cells with buffer containing sodium azide, which releases particles bound to the cell surface [22]. Therefore, we repeated the phagocytosis

PLOS ONE
assays, but this time washed the cells twice with buffer containing 5 mM sodium azide before fixing for flow cytometry. We found that the sodium azide washes drastically reduced the overall mean cell fluorescence for both cell types (Fig 1A, hollow shapes). Although the mean cell fluorescence of cpnAcells was higher than NC4A2 cells after washes with buffer containing sodium azide, the differences at each time point between NC4A2 and cpnA-cells were no longer significant. This suggests that the increased mean cell fluorescence of cpnAcells was due to increased bead adhesion and that increased adhesion resulted in the observed, but not significant, increase in phagocytosed beads at the later timepoints (Fig 1A, hollow shapes). We also used florescence microscopy to examine cells that were incubated with beads for 15 minutes and then washed in buffer with or without sodium azide (Fig 1B). The images appeared to corroborate the flow cytometry data in that fewer beads were associated with cells when washed with sodium azide and most of the beads remaining appeared to be inside the cell and not attached on the outside surface. Overall, these data suggests that the increased bead fluorescence of the cpnAcells was due to the increased adhesion properties of the cpnAcells.
To determine if cpnAcells exhibited increased phagocytosis and/or adhesion to bacteria, we did phagocytosis assays using GFP-expressing Klebsiella aerogenes with parental and cpnAcells and imaged cells at the 30-minute timepoint (Fig 2). Cell samples were washed with buffer alone or buffer containing sodium azide and then fixed on coverslips. The cells were imaged with fluorescence microscopy and the number of bacteria per cell were counted (Fig 2). Similar to the results of the bead assay, cpnAcells were associated with significantly more GFP-bacteria compared to parental cells when not washed with buffer containing sodium azide. Yet, when cells were washed with buffer containing sodium azide to rid the cells of surface bound GFP-bacteria, there was no significant difference in the number of associated bacteria per cell. The number of associated bacteria per cell is low due to the GFP being readily quenched once the bacterium is phagocytosed.

Increased adhesion of cpnAcells is not dependent on actin filaments
Actin filament dynamics are necessary for the phagocytic engulfment of particles and are important in cell adhesion [23]. We previously showed that CpnA is able to interact with actin filaments in a calcium-dependent manner and may play a role in actin depolymerization at the surface of membranes [13,17]. To further understand CpnA's role in adhesion, we used the actin depolymerizing drug, Latrunculin A (LatA) to both inhibit phagocytosis [24] and to see if actin filaments play a role in the increased adhesion observed in cpnAcells. If F-actin is a major contributing factor to the increased adherence of beads observed in cpnAcells, then we would expect parental NC4A2 and cpnAcells treated with LatA to have similar amounts of bead adherence. Cells were incubated with LatA for 30 minutes and then incubated with 1 μm beads for 15 additional minutes in a shaking suspension. Cells were washed in buffer alone or containing sodium azide, fixed, and analyzed by flow cytometry (Fig 3A). Because the LatA treatment inhibits phagocytosis, we assume we are only measuring beads adhered to the cell surface. With the LatA treatment, cpnAcells had significantly more fluorescence compared to the parental cells. With both the LatA treatment and sodium azide washes, there were still some beads associated with cells, but there was not a significant difference between parental and cpnAcells (Fig 3A). Images of fixed cells corroborated the flow cytometry data and indicated that the beads were found mainly associated with the surface of cells when treated with LatA and that the sodium azide washes removed most, but not all beads (Fig 3B). If the increased adhesion observed in cpnAcells was solely due to an actin-based adhesion process, we would expect parental and cpnAcells to exhibit similar bead adherence in the absence of actin filaments. Given that cpnAcells exhibited significantly more bead adherence even with  LatA treatment, the increased adhesion properties of cpnAcells does not appear to be due to a difference in actin filament dynamics.

Increased adhesion of cpnAcells is not dependent on cell surface proteins
The increased cell adhesion to beads could be due to an increase or change in adhesion proteins at the cell surface of cpnAcells. To test this hypothesis, parental and cpnAcells were treated with proteinase K to degrade cell surface proteins before bead adherence was measured. If the increased adhesion was due to proteins on the cell surface, we would expect a decrease in bead adherence of cpnAcells to the level of parental NC4A2 cells when cells were treated with proteinase K. Parental and cpnAcells were treated with LatA first and then incubated with different concentrations of proteinase K for 15 minutes before adding 1 μm beads. The cells were allowed to adhere to beads for 15 minutes in a shaking suspension before samples were washed with buffer and fixed for flow cytometry. Treating the parental cells with 100 μg/mL and 500 μg/mL proteinase K concentrations resulted in a decrease in adhesion of beads ( Fig 4A). However, treating cpnAcells with 100 μg/mL of proteinase K did not result in a decrease in adhesion and treating cells with 500 μg/mL resulted in an increase in bead adhesion (Fig 4A). At each proteinase K concentration, the cpnAcells had significantly increased bead adherence compared to NC4A2 cells. These data indicate that proteins on the cell surface of cpnAcells do not contribute to the increased adhesion of beads and suggests that instead of changes in protein composition, changes in the lipid composition of the plasma membrane may be the cause of the increased adhesion.
Although surface proteins may be involved in bead binding, indicated by a decrease in beads adhesion observed with NC4A2 cells, the removal of surface proteins from the cpnAcells may lead to increased bead binding to surface lipids.
To confirm that proteinase K degraded cell surface proteins, whole cell samples were analyzed by a western blot using an antibody to the cell surface protein, SibA (Fig 4B). SibA is a 208 kDa integral membrane protein with a large extracellular domain that is recognized by the antibody used in the western blot. The first two lanes contain whole cell samples of cells that were not treated with the proteinase K, while the next two lanes contain whole cell samples of cells treated with proteinase K. Cells were counted again after Sorensen's buffer washes and the same number of cells was loaded into each lane. Proteinase K treated and untreated cells were imaged using phase contrast microscopy and appeared similar in cell size and shape. Cells treated with proteinase K had less SibA protein indicating that cell surface proteins were degraded (Fig 4B). As a loading control we also used an antibody to actin on the same blot to verify equivalent sample loading. Similar actin bands were observed when comparing the parental cells to cpnAcells; however, the proteinase K treated cells had slightly less actin. During cell lysis with sample buffer, it appeared some proteinase K was still active. Although we added proteinase inhibitor to the sample buffer, proteinase K is active at high temperatures and in SDS, so it was difficult to completely inhibit.

cpnAcells have more phosphatidylserine in the outer leaflet of the plasma membrane
Because proteins did not appear to be involved in the increased adherence of cpnAcells, we turned to the other main component of the plasma membrane: lipids. The cell membrane is shaking suspension and incubated with 1 μm beads for 15 minutes. Cells were washed in buffer with or without sodium azide (SA), fixed on coverslips, and imaged using DIC and fluorescence microscopy. Scale bar = 16 μm.
https://doi.org/10.1371/journal.pone.0250710.g003 composed of two leaflets, with the outer leaflet composed mainly of phosphatidylcholine and sphingomyelin and the inner leaflet composed of phosphatidylethanolamine, phosphatidylinositol, and phosphatidylserine [1]. We hypothesized that the increased adhesion may be due to a change in lipid composition at the cell surface. To investigate lipids on the cell surface we looked specifically at phosphatidylserine (PS). CpnA binds to PS in both a calcium-dependent and calcium-independent manner [8,12]. We hypothesized that changes to the cell membrane lipid composition could be caused by altered activity of either flippases or scamblases. Both scenarios could lead to cpnAcells having more PS in the outer leaflet of the plasma membrane. To test whether cpnAcells have increased PS in the outer leaflet of the plasma membrane, we used Annexin V-APC to label PS on the surface of NC4A2 and cpnAcells and analyzed cells with flow cytometry (Fig 5). Annexin-V is a calcium-dependent lipid binding protein that specifically binds anionic lipids with a strong affinity for PS. NC4A2 and cpnAcells were incubated in buffer with Annexin V-APC for 15 minutes or with buffer containing a calcium ionophore for 10 minutes, and then incubated with Annexin V-APC for 15 minutes. The cells were immediately analyzed via flow cytometry. We found that cpnAcells had significantly more Annexin V-APC fluorescence compared to the parental cells, indicating that cpnAcells had more PS on the outside of the cells (Fig 5A). This increase in PS exposure can also be observed by a small shift in fluorescence between NC4A2 and cpnAcells (Fig 5B). Stimulation of the cells with a calcium ionophore leads to a rapid elevation of intracellular calcium concentration and subsequent externalization of PS due to the activation of scramblase [25]. We found that cpnAcells had a similar amount of fluorescence as compared to the calcium ionophore treated parental cells (Fig 5A).

Annexin V binding to cpnAcells decreases adhesion
To examine the hypothesis that the increased PS in the outer leaflet of the plasma membrane in cpnAcells contributes to the observed increased adhesion, we tested whether unlabeled Annexin-V could block bead adhesion. Parental and cpnAcells were incubated with LatA in a shaking suspension for 30 minutes and then incubated with either 0.1 μg/mL of Annexin-V or BSA for an additional 15 minutes. We then added 1 μM beads for 15 minutes and fixed cells for flow cytometry. We found that the Annexin V was able to block bead adhesion to the cpnAcells with little effect on the parental cells (Fig 6A). We also used BSA as a control protein for nonspecific protein inhibition of bead adherence. BSA had little effect on bead adhesion in both cell types. Furthermore, NC4A2 and cpnAcells were treated as described above and fixed for fluorescence microscopy ( Fig 6B). The fluorescence images also showed that Annexin-V was able to block the increased bead adhesion of cpnAcells. These data suggest that the increased adhesion observed in cpnAcells is likely due to changes to the lipid composition of the plasma membrane and may be specifically due to the increased PS exposure on the plasma membrane.

cpnAcells are more sensitive to the lytic peptide, Polybia MP-1
Another way to determine if cpnAcells have altered plasma membrane lipid composition and increased PS exposure is to test the sensitivity of cells to the lytic peptide found in wasp venom, Polybia-MP1. This peptide binds to external PS residues on the membrane and creates pores in the membrane to induce cell lysis [26]. We placed NC4A2 and cpnAcells in suspension with different concentrations of Polybia-MP1 for 30 minutes and then counted the number of intact cells using a hemocytometer (Fig 7A). For the parental cells, increasing the Polybia-MP1 concentration did not induce cell lysis until 5 μM Polybia-MP1 concentration. However, there was significant cell death of the cpnAcells compared to the parental cells

PLOS ONE
starting at 2 μM Polybia-MP1 and very little cell viability at 4 μM and 5 μM. To see if the overexpression of CpnA could rescue the sensitivity to Polybia-MP1, we overexpressed a GFPtagged version of CpnA in both the parental and cpnAcell lines. Similar to the parental cell line, both of these cell lines showed very little cell death until 5 μM Polybia-MP1, indicating that expression of CpnA in the cpnAcell line is able to rescue this defect. We also imaged cells after Polybia-MP1 treatment using DIC microscopy. Cells were placed on glass bottom dishes with 4 μM Polybia-MP1 and imaged for 15 minutes (Fig 7B). At 5 minutes cpnAcells started to lyse, while relatively little lysis was observed in the three other cell lines. These data show that cpnAcells are more sensitive to Polybia-MP1, indicating that cpnAcells have more PS exposed on the cell surface than the parental cells.

Discussion
We began this study by investigating the role of CpnA in phagocytosis and found that more beads and bacteria were associated with cpnAcells than parental cells. Further investigations indicated that cpnAand parental cells had similar amounts of beads and bacteria phagocytosed and that increased adhesion to particles resulted in the increased bead and bacteria association observed with cpnAcells. To investigate how the lack of CpnA may contribute to this increased adhesion, we investigated different cellular components involved in cell adhesion and found that the increased adhesion of cpnAcells was not due to a change in actin filaments at the cell surface nor a change in cell surface proteins. However, we did find that cpnAcells had increased PS in the outer leaflet of the plasma membrane and the increased PS was linked to the increased adhesion.
This study provides the first evidence of a copine protein playing a role in the regulation of plasma membrane lipid composition and possibly PS exposure. In general, the copine family is made up of soluble, cytosolic proteins with calcium-dependent phospholipid binding activity [12]. Copines do not exhibit any homology to other proteins, like flippases and scramblases, that are able to transport PS from one side of a membrane to the other. Therefore, we propose that CpnA does not directly function in phospholipid transport, but has an indirect or regulatory role. Previously, we showed that copines in Dictyostelium translocate from the cytosol to the plasma membrane in response to a rise in calcium concentration [12]. Once at the plasma membrane, copines could interact with and regulate plasma membrane proteins that are involved in the transport or flipping of lipids.
Most of the processes in animal cells that require PS exposure involve some form of cell-tocell adhesion. Examples include phagocytic cells recognizing and engulfing apoptotic cells; red blood cell and platelet aggregation; sperm and egg fertilization, and myocyte fusion [5]. Our results also suggest a role for PS exposure in cell-to-cell adhesion in that cpnAcells have increased adhesion to bacterial cells. In addition, our previous studies showed that cpnAcells have development defects indicative of increased cell-to-cell adhesion [15,16]. Dictyostelium can live as single-celled amoebae. However, when placed in starvation conditions, cells undergo chemotaxis in response to secreted cAMP and aggregate into mounds [27]. The mound elongates to form a finger-shaped structure that falls over and migrates as a slug before culminating into a fruiting body consisting of a stalk, made up of cells that have undergone programmed cell death, with a mass of spores on top [27]. We found that cpnAcells formed larger than normal mounds and slugs did not culminate into fruiting bodies [14][15][16]. When cpnAcells were mixed with increasing percentages of parental cells, mixed cell populations were able to make progressively more normal slugs and fruiting bodies [15,16]. cpnAcells also formed small aggregates within the mixed population slugs [15]. Interestingly, when cpnA -cells were developed in buffer containing EGTA, they were able to complete the developmental cycle and culminate into small fruiting bodies [15].
In light of our new data, we hypothesize that the increased cell adhesion observed during development of cpnAcells may be due to altered cell surface lipid composition. The result that EGTA rescued the developmental culmination defect suggests a role for calcium in the increased adhesion of cpnAcells. The development of cpnAcells in buffer containing EGTA may decrease intracellular calcium levels, resulting in decreased scramblase activity. Alternatively, EGTA may reduce the calcium-dependent binding of other proteins that mediate adhesion to PS. One possibility for how the lack of CpnA causes increased adhesion is that CpnA has a role in the negative regulation of scramblase. Recent evidence indicates that there are two main protein families (XKr8 and TMEM16) involved in scramblase activity at the plasma membrane in animal cells [28]. Dictyostelium does not appear to have any XKr8 family members, but has a single member of the TMEM16 family [6]. Dictyostelium TMEM16 was shown to confer calcium-dependent scramblase activity when expressed in HEK293 cells [6]. CpnA could directly interact with this scramblase or indirectly regulate the activity of the scramblase by playing a role in the negative regulation of intracellular calcium levels.
In animal cells, PS exposed on the surface of cells acts as binding sites for proteins. For example, the PS exposed on activated platelets serves as assembly sites for tenase and prothrombinase complexes [29]. More specifically, the PS acts a binding site for negatively charged carboxyglutamate (Gla) residues at the N-termini of multiple coagulation factors via Ca 2+ ions [29]. A similar type of calcium ion bridge may be occurring between the PS exposed on cpnAcells and the carboxylate-modified beads used in our assays.
Altered membrane lipid composition and/or calcium concentrations in cpnAcells may also contribute to the increased postlysosome exocytosis phenotype we previously reported [17]. Other studies have reported a link between PS exposure and membrane trafficking. A study with PC12 cells showed that increasing PS levels increased calcium-triggered exocytosis [30], while a study with Jurkat cells showed that calcium activation of scramblase triggered both PS exposure and membrane expansion [31]. Alternatively, increased exocytosis may be responsible for the increased PS on the cell surface of cpnAcells.
Future studies will focus on identifying how CpnA is involved in the negative regulation of PS exposure and how PS exposure is involved in Dictyostelium development. During development, prestalk cells lose lipid asymmetry in that they expose PS in the outer leaflet of the plasma membrane [32] and our previous studies indicate CpnA plays a role in the differentiation of prestalk cells [15,16]. Therefore, PS exposure is most likely a tightly regulated process necessary for development and other fundamental cell processes. Cancer cells also have increased PS exposure and this is thought to be caused by low flippase activity and/or high scramblase activity [33,34]. Therefore, uncovering key regulators of PS exposure and the link to adhesion will be important to our understanding of tumorigenesis and cancer metastasis.