Copine C plays a role in adhesion and streaming in Dictyostelium

ABSTRACT

Copines are a family of calcium-dependent membrane-binding proteins. To study these proteins, anull mutant for cpnC was created in Dictyostelium, which has six copines genes (cpnA-cpnF). During development, cpnC⁻ cells were able to aggregate, but did not form streams. Once aggregated into mounds, they formed large ring structures. cpnC⁻ cells were less adherent to plastic substrates, but more adherent to other cells. These phenotypes correlated with changes in adhesion protein expression with decreased expression of SibA and increased expression of CsaA in developing cpnC⁻ cells. We also measured the expression of RegA, a cAMP phosphodiesterase, and found that cpnC⁻ cells have reduced RegA expression. The reduced RegA expression in cpnC⁻ cells is most likely responsible for the observed phenotypes.

KEYWORDS:

• Adhesion
• copine
• development
• Dictyostelium
• signaling
• streaming

Introduction

Copines are a family of highly conserved calcium-dependent membrane-binding proteins found in many eukaryotic organisms, ranging from single-celled organisms to multicellular plants and animals [1,2]. Copines are characterized by two C2 domains followed by an A domain that is similar to the VWA domain found in integrins. The C2 domains facilitate binding of calcium and acidic phospholipids [3,4], while the VWA domain facilitates protein binding [5]. Copines in mammalian cells and Dictyostelium are cytosolic and translocate to the plasma membrane in response to a rise in intracellular calcium concentration [3,6]. The domain structure of copines suggests two ways in which copines may function as calcium mediators in cells. Copines may bind to cytosolic proteins and bring them to membranes in response to a rise in intracellular calcium or copines may translocate to membranes in response to calcium to regulate membrane proteins. Indeed, copines may function in both ways; a myriad of both cytosolic and membrane proteins has been identified as targets of copine proteins [5,7,8].

There are nine copines in the human genome and several copines have been implicated in human diseases, including cancer [9]. Gene expression screens have shown that various copines are highly expressed in cancer cells and subsequent functional studies have indicated that copines promote cell proliferation, cell migration, and metastasis [10–14]. For example, in breast cancer cells, copine 3 translocated to the plasma membrane in response to increased calcium and interacted with a phosphorylated tyrosine on the ErbB2 receptor. Copine 3 also interacted with RACK1 and localized to focal adhesions in migrating cells, while copine 3 knockdown in these cells impaired ErbB2-dependent migration [15]. Another study showed that copine 3 is highly expressed in glioblastoma cells as compared to normal brain tissues and suppression of copine 3 expression significantly impaired the ability of the cells to proliferate and migrate into neighboring tissue due to inactivation of the focal adhesion kinase (FAK) signaling pathway [16].

We are studying copine function in Dictyostelium discoideum, a soil dwelling amoeba. Dictyostelium lives independently in the soil consuming bacteria through phagocytosis. In starvation conditions, the amoeba signal each other with the release of cAMP to aggregate and enter a phase of multicellular development. In this phase of development, cells will migrate toward aggregation centers in streams and form a mound where cells will differentiate. As development progresses, the mound will grow into a motile slug, and following photo- and thermo-tactic cues will migrate to a suitable environment to finish development. The multicellular development phase ends when the cells form a fruiting body, consisting of spores atop a rigid stalk [17].

The Dictyostelium genome has six copines genes, cpnA-cpnF [2]. Each of the copines in Dictyostelium has a different developmental expression pattern [18] and has distinct phospholipid binding properties, along with differences in the timing and the magnitude of membrane localization when cells are stimulated with cAMP [3]. These data indicate the six copines may have distinct and non-redundant functions. To investigate copine function in Dictyostelium, we are characterizing single copine gene knockout mutants. We have previously characterized the cpnA⁻ knockout cell line and found that cells lacking cpnA have defects in development and several defects associated with actin cytoskeleton dynamics, including cytokinesis, adhesion, and chemotaxis [18–20].

The current study describes the initial characterization of a cpnC knockout mutant cell line and focuses on cell adhesion and migration. Cell adhesion and migration mechanisms are important for both vegetative and developmental stages in Dictyostelium. Dictyostelium has been used extensively to study proteins involved in cell chemotaxis and migration [21–23]. While cell adhesion proteins have roles in migration, they also facilitate phagocytosis during vegetative growth and cell-cell adhesion during development. Multiple proteins have been identified that function in cell adhesion in Dictyostelium, some of which appear to be counterparts to mammalian focal adhesion proteins [24–27]. SibA (Similar to Integrin Beta) is a transmembrane protein that contains an extracellular VWA domain similar to the one found in the metazoan integrin beta subunit. SibA is involved in cell-substrate adhesion in vegetative cells and interacts with talin, an integrin and actin-binding protein that localizes to the site of cell-substratum contact [26]. Other types of adhesion proteins like CsaA (Contact Sites A) do not have a mammalian counterpart; CsaA facilitates cell-cell adhesion during multicellular development [24,28]. The investigation of the copine family’s role in cell adhesion and migration is important to determining the role of copines in cancer that relies on changes in cell adhesion and migration to metastasize.

Materials and methods

Dictyostelium strains and culture

The Dictyostelium discoideum strain used was NC4A2, which is an axenic strain derived from the wild-type NC4 strain [29]. NC4A2 is referred to as the ‘parental strain’ hereafter. Cells were grown on plastic Petri dishes or shaking in flasks at 18°C or room temperature in HL-5 media (0.75% proteose peptone, 0.75% thiotone E peptone, 0.5% Oxoid yeast extract, 1% glucose, 2.5 mM Na2HPO4, and 8.8 mM KH2PO4, pH 6.5) or VL6 media including glucose (Formedium, VL60102). Media was supplemented with penicillin-streptomycin at 60 U/mL. The cpnC⁻ knockout mutant strain was generated by inserting a bsr gene, conferring blasticidin S resistance (bsr), in place of the cpnC gene in the parental cell line through homologous recombination. cpnC⁻ mutant cells were grown in HL-5 or VL6 media including glucose supplemented with 10 µg/mL of blasticidin. Dictyostelium strains obtained from the Dicty Stock Center, sibA⁻ [26] and csaA⁻ [30] were grown in VL6 media including glucose supplemented with 10 µg/mL blasticidin and 7.5 mg/mL G418, respectively.

Creation of a cpnC⁻ mutant cell line

The 5’ and 3’ flanking sequences of the cpnC gene were amplified from purified genomic DNA of parental cells through polymerase chain reaction (PCR). The 5’ 986 bp flanking region was amplified using the forward primer: 5’-GGT ACC TGT GGT GAT TCT TGT GCA GT-3’ and the reverse primer 5’-AAG CTT TTG GAA ACC TCC TGG ACC TCT-3’. A KpnI restriction enzyme site was added to the 5’ flanking upstream primer and a HindIII restriction enzyme site was added to the 5’ flanking downstream primer. The 3’ 987 bp flanking region was amplified using the forward primer: 5’-GTG GTC CAA CTA ATT TTG CAT CAG-3’ and the reverse primer: 5’-TCG AAT TAC GTC CTG ATG GTG-3’. Amplified 5’ and 3’ flanking PCR fragments were ligated separately into the pCR2.1-TOPO plasmid according to the manufacturer’s protocol (Invitrogen 450641).

The cpnC 3’ flanking sequence was subcloned into the pLPBLP vector [30] on the 3’ side of the bsr cassette between the NotI and BamHI sites. Then, the cpnC 5’ flanking sequence was subcloned into the pLPBLP vector containing the cpnC 3’ flanking sequence on the 5’ side of the bsr cassette between the KpnI and HindIII sites. The pLPBLP plasmid containing both the 5’ and 3’ cpnC gene flanking sequences was isolated using the QIAprep spin miniprep kit (Qiagen 27,106), digested with KpnI for two hours at 37°C, and then purified using the QIAquick PCR purification kit (Qiagen 28,104). Linearized plasmid was transformed into parental cells using electroporation. Transformed cells were aliquoted into 96-well plates with HL-5 media containing penicillin-streptomycin (60 U/mL). After 24 hours, the HL-5 media was replaced with HL-5 media containing penicillin-streptomycin (60 U/mL) and blasticidin (10 μg/mL).

Genomic DNA was purified from the parental strain and a clonal population of blasticidin-resistant pLPBLP transformed cells. To confirm that the blasticidin-resistant cells were lacking the cpnC gene, PCR using the genomic DNA as template was performed with the following primers: cpnC upstream primer: 5’-ATG ATA CCA TCG TCA AAA C-3’ paired with cpnC downstream primer: 5’ CAA GAC CAA TAC CTT CAG C-3’ and cpnC upstream primer paired with the bsr reverse primer: 5’-AAT CGC AAT GGC TTC TGC AC-3’. The PCR amplicons were analyzed by gel electrophoresis on a 1% agarose gel, stained with 0.5 µg/mL ethidium bromide, and imaged with the Bio-Rad ChemiDoc image system.

Whole genome sequencing

Genomic DNA from parental and cpnC- cells was purified and treated with RNase. DNA samples were sent to the Research Technology Support Facility Genomics Core at Michigan State University for whole genome sequencing and analyzed to verify the cpnC gene was replaced with the bsr gene. Libraries were prepared using the Roche Kapa HyperPrep DNA Library Preparation Kit with Kapa Unique Dual Index (UDI) adapters following manufacturer’s recommendations. Completed libraries were quality checked and quantified using a combination of Qubit dsDNA HS and Agilent 4200 TapeStation HS DNA1000 assays. The libraries were pooled in equimolar amounts and the pool was quantified using the Invitrogen Collibri Quantification qPCR kit. The pool was loaded onto one lane of a NovaSeq S4 flow cell and sequencing was performed in a 2 × 150 bp paired end format using a NovaSeq 6000 v1.5 500 cycle reagent kit. Base calling was done by Illumina Real Time Analysis (RTA) v3.4.4 and output of RTA was demultiplexed and converted to FastQ format with Illumina Bcl2fastq v2.20.0.

The nf-core v 23.04.1/sarek v 3.3.0 pipeline was used to map reads and call variants [31–33]. Briefly, BWA-mem [34] v 0.7.17-r1188 was used to map 2 × 150 bp paired-end reads from each sample to a genome containing the Dictyostelium discoideum reference genome (Ensembl release 57) and the plasmid used to create the expected mutations (pPLBLP). Duplicates were marked using GATK4’s (v 4.4.0.0) markduplicates function [35]. Manta v 1.6.0 was used to call somatic variants in each mutant strain using the appropriate control strain as the ‘normal’ sample [36]. Samtools v 1.15 was used to filter cram files to remove duplicate reads, keep only paired reads with primary alignments with scores $\ge 2$ [37]. Samtools, Bedtools v 2.30.0, and custom R script v 4.1.0 was used to extract reads in which one member of the pair was mapped to the plasmid and the other is mapped to the reference genome (chimeric read pairs) [38,39]. Bedtools was also used to identify regions of the genome covered by at least 2 chimeric reads. BLAST v 2.10.0 was used to identify regions in the Dictyostelium discoideum reference genome homologous to regions in pPLBLP [40]. The Integrative Genomics Viewer (IGV v 2.15.2) was used to visualize genomic reads, chimeric reads, manta variants, and homologous regions between the reference genome and plasmid [41].

Growth assay

Parental and cpnC⁻ cell lines were harvested from Petri dishes and inoculated at 5 × 10⁴ cells/mL into 12.5 mL of HL-5 media or VL6 media including glucose (Formedium, VL60102) supplemented with 60 U/mL of penicillin-streptomycin in 125 mL flasks. Inoculated flasks were placed in a shaking incubator set to 125 rpm at 18°C. Every 24 hours for 7 days, three samples per cell line were removed from each flask and the cell density in each flask was estimated using a hemocytometer. The cell density for each of the three samples was averaged. A total of six growth assays were performed and the mean cell densities at each timepoint were averaged. A repeated measures ANOVA with post hoc Tukey comparisons was performed to determine if there were any significant differences in mean densities between parental and cpnC⁻ cells at each timepoint. Doubling times for both parental and cpnC⁻ cells were calculated using the formula [ln2/ln(C₁/C₂)* 48 hours] with cell counts at day 2 (C₁) and day 4 (C₂). The doubling times from six trials were averaged and t-test was used to analyze significant differences between parental and cpnC⁻ cells.

Nuclei count assay

Parental and cpnC⁻ cell lines were harvested from Petri dishes and inoculated at 1 × 10⁶ cells/mL into 12.5 mL of HL-5 media or VL6 media supplemented with 60 U/mL of penicillin-streptomycin in 125 mL flasks. The cells were grown in a shaking suspension for 48 hours at 18°C and 125 rpm. After 48 hours, 200 µL of cells at a concentration of 2 × 10⁶ cells/mL were placed on coverslips and allowed to adhere for 15 minutes. Parental and cpnC⁻ cells were fixed to the coverslips with 1% formaldehyde in methanol at − 20°C for 10 minutes. The coverslips were washed three times with phosphate buffered saline (PBS) (137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄) and then incubated with 200 µL of 0.1 μg/ml DAPI (4′,6′-diamidino-2-phenylindole) in PBS for 10 minutes. The coverslips were rinsed three times with PBS and mounted on slides. Fixed cells were imaged using epifluorescence on a Leica DMi8 with a 40× objective and DAPI filter. The number of DAPI stained nuclei present in > 300 randomly selected cells were counted for each cell line and the mean number of nuclei per cell was calculated. The mean number of nuclei from five trials were averaged and a t-test was used to analyze significant differences between parental and cpnC⁻ cells.

Development assay

Parental and cpnC⁻ cell lines were harvested from Petri dishes in 10 mL of Development Buffer (DB; 5 mM Na₂HPO₄, 5 mM KH₂PO₄, 1 mM CaCl₂, 2 mM MgCl₂, pH 6.5). Cells were washed three times by centrifugation at 437 × g for 5 minutes at 4°C and then resuspended to a final concentration of 2.5 × 10⁷ cells/ml in DB. Drops of cell suspension (20 µL each) were plated onto DB agar plates (DB, 15 grams agar/L) and incubated at 21°C. The lid of the Petri dish had a 1 cm diameter hole removed to prevent lid condensation. The developing cells were imaged every 10 minutes for 36 hours on a Nikon SMZ800N dissecting microscope.

Development on bacteria lawns

E. coli B/R was inoculated into 2 mL of HL-5 and incubated at 220 rpm at 37°C for 16 hours. Parental and cpnC⁻ cell lines were harvested from Petri dishes and counted using a hemocytometer and subjected to serial dilutions to obtain a final concentration of 500 cells/mL. E. coli (500 µL) and Dictyostelium cells (100 µL) were spread on SM/5 agar plates (0.2% proteose peptone 2, 0.2% yeast extract, 0.2% glucose, 0.2% MgSO₄·7 H₂O, 0.19% KH₂PO₄, 1% K₂HPO₄, 1.5% agar, pH 6.4). Plates were incubated at 18°C and examined for plaque formation every day over the course of one week. After three days, the plates were imaged with a Nikon SMZ800N dissecting microscope and individual plaques (~12 plaques per trial) were measured using ImageJ software. Mean plaque size data from three trials were averaged and a t-test was used to analyze significant differences between parental and cpnC⁻ cells.

Phagocytosis assays

Parental and cpnC⁻ cells were harvested from Petri dishes, counted using a hemocytometer, and collected by centrifugation at 437 × g for 5 mins at 4 ̊C. After centrifugation, 2 × 10⁶ cells/mL were incubated with 1 μm yellow-green large FluoSphere carboxylate-modified microspheres (ThermoFisher, F8823) at 3.64 × 10¹⁰ 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. Cells (100 μL) were transferred from the flask into a microcentrifuge tube at 5-minute timepoints and cells were washed by centrifugation at 437 × g for 5 minutes at 4 ̊C. The cells were resuspended in ice-cold Sorensen’s buffer (0.2 M NaH₂PO₄, 0.2 M Na₂HPO₄, pH 6.5), washed two times, fixed in 3.7% formaldehyde in Sorensen’s buffer, and washed two times with Sorensen’s buffer again. Mean cell fluorescence from three trials was measured with flow cytometry. To remove beads from the surface of cells before fixation, cells were washed in Sorensen’s buffer containing 5 mM sodium azide (Sigma-Aldrich 26,628-22-8) twice and then in Sorensen’s buffer alone. Mean cell fluorescence from four trials was measured using flow cytometry. Significant differences at each timepoint between parental and cpnC⁻ cells with and without sodium azide washes were analyzed using a repeated measures ANOVA and post hoc Tukey comparisons.

Bead adhesion assays

In 50 mL flasks containing 2.5 mL HL-5 media, 2 × 10⁶ cells/mL parental and cpnC⁻ cells 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 minutes, 1 μm beads (3.64×10¹⁰ beads/mL) were added to the flasks. After 15 minutes, cells were centrifuged at 437 × 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 fluorescence within each trial. Normalized data from three trials were averaged and analyzed for significant differences using a two-way ANOVA and post hoc Tukey comparisons. For fluorescence microscopy, parental and cpnC⁻ cells 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 using a 100× DIC objective and GFP filter.

For bead adhesion assays with proteinase K, parental and cpnC⁻ cells were incubated with LatA for 30 minutes and then either 0 μg/mL, 100 μg/mL, or 500 μg/mL of proteinase K for 15 minutes in a 180 rpm shaking suspension. After 15 minutes, 1 μm beads (3.64×1010 beads/mL) were added to the flasks and allowed to adhere to the surface of the cells for 15 minutes. Cell samples (100 μL) were washed three times with Sorensen’s buffer by centrifugation at 437 × 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 cpnC⁻ cells using a two-way ANOVA and post hoc Tukey comparisons. For fluorescence microscopy, parental and cpnC⁻ cells were incubated with LatA for 30 minutes and then 0 µg/mL or 500 µg/mL of proteinase K for 15 minutes in a shaking suspension. After 15 minutes, 1 µm FITC beads were added to cell samples. After an additional 15 minutes, cell samples were placed on coverslips and fixed in 3.7% formaldehyde in methanol and imaged with a Leica DMi8 microscope.

FITC-Lactadherin assay

Parental and cpnC⁻ cells were centrifuged at 437 × g for 5 minutes at 4 ̊C and then resuspended to 1 × 10⁶ cells/mL in 1X binding buffer (0.01 M HEPES, 0.14 M NaCl, 2.5 mM CaCl₂). For a positive control, 3 μM calcium ionophore A23187 (Sigma-Aldrich, C7522) was added to 250 µL parental and cpnC⁻ cells and allowed to incubate at room temperature for 10 minutes. Then, 7.5 µL of FITC-Lactadherin (Haematologic Technologies, LL0302) (83 µg/mL) was added to 250 µL of cells treated with ionophore or buffer alone and allowed to incubate for an additional 15 minutes. Live cells were analyzed via flow cytometry. Mean cell fluorescence data for each cell type and condition were normalized to the average fluorescence within each trial. Normalized data from four trials were averaged and analyzed for significant differences using a two-way ANOVA and post hoc Tukey comparisons.

Flow cytometry

Cell samples were analyzed on a Beckman Coulter CytoFlex Flow Cytometer with the 488 nm laser and FITC detector (525/40). 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.

Plate adhesion assay

Parental and cpnC⁻ cells were harvested from plates, counted using a hemocytometer, and centrifuged at 437 × g for 5 minutes. The cells were resuspended in media at 1 × 10⁶ cells/ml. On small Petri dishes, 2 ml of each cell type was allowed to settle and adhere for 20 minutes. The cells that had not adhered were removed and the media replaced. Images of the adhered cells were taken on a Nikon TE2000 microscope with a 20× phase-contrast objective at four marked spots on the Petri dishes. Cells in dishes were rotated for 15 minutes at 100 rpm. Detached cells were removed with the media and new media was added. New images at the 4 marked spots were captured. The procedure was repeated at 125 rpm and 150 rpm. The number of cells in each image were counted using the Cell Counter plugin in ImageJ and averaged. The percentage of detached cells after rotation was calculated by the subtraction of the average number of cells remaining after rotation from the average number of cells present before rotation divided by the number of cells present before rotation. Data from three trials were averaged and analyzed for significant differences between parental and cpnC⁻ cells by a repeated measures two-way ANOVA and Sidak’s multiple comparisons.

Streaming assay

Parental and cpnC⁻ cells were harvested from plates, counted using a hemocytometer, and centrifuged at 437 × g for 5 minutes. The cells were resuspended in DB at 1.5 × 10⁷ cells/ml. In 60 mm Petri dishes, 1 mL of cells was mixed with 1 mL of DB. The cells were incubated at room temperature. Images were taken every 2 minutes for 12 hours using a Nikon SMZ800N dissecting microscope.

Cell-cell adhesion assay

Parental and cpnC⁻ cells were harvested from plates, counted using a hemocytometer, and centrifuged at 437 × g for 5 minutes. The cells were resuspended in DB at 1.5 × 10⁷ cells/ml. In 60 mm Petri dishes 1 mL of cells was mixed with 1 mL of development buffer. The cells were incubated at room temperature until the parental cells began streaming, approximately 6–8 hours. Streaming cells were disrupted and transferred to a 15 mL conical vial and mixed by a vortex for 15 seconds. To determine the number of single and doublet cells, three samples of each cell type were collected and the number of single and doublet cells for each cell type were counted using a hemocytometer immediately after mixing. The cells were transferred to 25 mL flasks and incubated at 125 rpm shaking for 60 minutes. Cell samples were collected, and the number of single and doublet cells were counted using a hemocytometer at 30 and 60 minutes. The percentages of single and doublet cells were calculated by dividing the average number of single and doublet cells at each timepoint by the number of single and doublet cells immediately after disruption. Data from four trials were averaged and analyzed for significant differences between parental and cpnC⁻ cells using a repeated measures ANOVA and post hoc Tukey comparisons.

Western blot and densitometry

Parental and cpnC⁻ cells were harvested from plates, counted using a hemocytometer, and centrifuged at 437 × g for 5 minutes. The cells were resuspended in DB at 1.5 × 10⁷ cells/ml. In 60 mm Petri dishes, 1 mL of cell suspension was mixed with 1 mL of DB. The cells were incubated at room temperature. Cells were harvested from plates at 2, 4, 6, 8 and 10 hours. Cells were centrifuged at 500 × g for 5 minutes. The supernatant was removed and the cells were resuspended in 2× Laemmli buffer at a concentration of 2 × 10⁶ cells/20 µL and boiled at 95°C for 10 minutes. Samples (20 µL) from each timepoint for each cell line were loaded on a 10% Mini-PROTEAN® TGX Stain-Free™ Protein Gel (BioRad 4,568,033). Proteins were separated at 200 V for 50 minutes. Images of the Stain-Free gels were taken on Bio-Rad ChemiDoc image system. Proteins were transferred to a PVDF membrane (ThermoFischer 88,018) at 100 V for 60 minutes. The membrane was blocked in PBS with 1% Tween and 5% milk for 45 minutes at room temperature and then incubated with either anti-SibA (1:1000) polyclonal antibody (Geneva Antibody Facility), anti-csaA (1:100) monoclonal antibody (Developmental Studies Hybridoma Bank), or anti-RegA (1:4000) polyclonal antibody (gift from Robert Kay [42]) in PBS with 0.02% sodium azide for 2 hours at room temperature. The membrane was washed with 1% Tween in PBS three times for 10 minutes each and then incubated with HRP-conjugated secondary antibody (1:15,000) for 1 hour at room temperature. The membrane was washed with 1% Tween in PBS three times for 10 minutes each, then treated with FemtoGlow Plus: Western Blotting chemiluminescence substrate (Michigan Diagnostics) and imaged using Bio-Rad ChemiDoc image system. For densitometry analysis of bands, the SibA, CsaA, and RegA band densities were measured using ImageLab 6.1 software from BioRad. Each band density of the western blot image was measured and then normalized to the density of all proteins in the corresponding lane from the stain-free gel image. Normalized western blot band densities were then further normalized to the SibA band for parental cells at the 0 timepoint and the CsaA band for parental cells at the 10 hour timepoint. Normalized RegA western blot band densities were not normalized to the parental cells but were further normalized to the average adjusted total band volume intensities across all lanes. Blot and gel images from three different trials were analyzed and normalized band densities were averaged. Normalized mean band densities were analyzed by a two-way ANOVA and post-hoc Fisher LSD comparisons.

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. Two-way ANOVA with post hoc multiple comparisons (R version 3.6.1 and Prism version 9.3.0) was used to analyze significant differences for cell populations with multiple treatments. T-tests were used to analyze significant differences between two populations (Prism version 9.3.0). * indicates p-value <0.05 and error bars represent standard error of the mean.

Results

Generation of a cpnC null mutant cell line

To study the function of cpnC in Dictyostelium, we created a cpnC null mutant using homologous recombination. The parental NC4A2 Dictyostelium cells were transformed with a plasmid containing 5’ and 3’ cpnC gene flanking sequences on each side of a blasticidin resistance (bsr) gene cassette. Transformed cells were treated with blasticidin and then screened for the replacement of the cpnC gene with the bsr gene using PCR (Supplemental Figure S1A, B). In addition, whole genome sequencing was performed to verify the replacement of the cpnC gene with the bsr gene (Supplemental Figure S1C).

cpnC⁻ cells have normal growth rates and increased multinucleation

To characterize the phenotype of cpnC⁻ cells, we first assayed growth and development in both the parental and cpnC⁻ cells. Parental and cpnC⁻ cells were grown in suspension at ${18}^{\circ }\mathrm{C}$ for 7 days. Every 24 hours, cells were collected and counted to evaluate the growth rate. Six growth assays were performed and the average number of cells at each timepoint are shown in Figure 2. There was no significant difference between the growth rates of parental and cpnC⁻ cells (Figure 1A). The average doubling time for parental cells was 20.5 ± 4.6 hours and the doubling time for cpnC⁻ cells was 21.3 ± 3.7 hours. While cpnC⁻ cells grown in shaking suspension did not exhibit a defect in the rate of growth, these cells did have increased multinucleation, suggesting a cytokinesis defect. Cells were grown in shaking suspension for 48 hours and then fixed and the nuclei stained with DAPI. The nuclei were imaged with DIC and fluorescence microscopy and the number of nuclei per cell were counted. cpnC⁻ cells had significantly more nuclei per cell than parental cells (Figure 1B). Representative images of stained nuclei are shown in Figure 2C. Increased multinucleation could be due to a defect in cytokinesis, but could also be caused by dysregulation of the cell cycle.

Figure 2. cpnC⁻ mutants made large ring structures and small fruiting bodies during multicellular development. Parental and cpnC⁻ cells were allowed to development on agar plates. Cells were imaged every 10 minutes for 24 hours on a dissecting microscope. Representative images of specific developmental stages are shown with hours after plating (5,9,13,19, and 23 hrs) as indicated. Black arrows point to rings during mound stage at 9 hrs. White arrow points to multiple slugs formed from a single ring of cpnC⁻ cells at 13 hrs. Mature fruiting body images were captured with the addition of a darkfield adaptor at 48 hours post starvation. Scale bar = 500 µm. See accompanying time-lapse movies (S1 and S2).

Figure 1. cpnC⁻ cells had normal growth rates and increased multinucleation. A) Parental and cpnC⁻ cells were grown in suspension starting at a density of 5 × 10⁴ cells/mL. Density estimates were made using a hemocytometer once a day for 7 days. Mean cell densities were calculated from six trials and a repeated measures ANOVA and post hoc Tukey comparisons was used to analyze significant differences between parental and cpnC⁻ cells. B) parental and cpnC⁻ cells were grown in suspension for 48 hours, fixed, stained with DAPI, and imaged with fluorescence microscopy. The number of nuclei present in > 300 randomly selected cells were counted in each trial and the mean number of nuclei per cell was calculated. The mean number of nuclei from five trials were averaged and a t-test was used to analyze significant differences between parental and cpnC⁻ cells. *indicates p-value <0.05, error bars=standard error. C) Representative images of DAPI stained cells, scale bar = 5 µm.

cpnC⁻ cells form large ring structures during the mound stage of development

To assess the ability of cpnC⁻ cells to progress through the 24-hour multicellular development program, parental and cpnC⁻ cells were placed in starvation buffer to initiate development, plated on agar plates, and imaged every 10 minutes for 24 hours (Figure 2; for time-lapse videos, see supplemental movies S1 and S2). cpnC⁻ cells aggregated with similar timing as parental cells. At the mound stage, both types of cells moved in a circular pattern. However, instead of making small rings that produce small tight round mounds like the parental cells, cpnC⁻ cells made very large ringed structures (Figure 2; black arrows). The large rings of the cpnC⁻ cells then broke up to form multiple slugs from each ring (Figure 2; white arrow). These slugs went on to produce smaller fruiting bodies (Figure 2; 48 hours).

cpnC⁻ cells have decreased adhesion to beads and no phagocytosis defect

Previously, we found that cpnA⁻ cells exhibited increased adhesion to beads, bacteria, and substrate [19,20]. To determine if cpnC⁻ cells also have a defect in adhesion to beads and bacteria, we performed phagocytosis and plaque assays. Parental and cpnC⁻ cells were incubated with 1 μm FITC-labeled large FluoSphere carboxylate-modified microspheres for 30 minutes shaking in media. Cell samples were removed at 5 minute timepoints, washed and fixed. Mean cell fluorescence was measured using flow cytometry. cpnC⁻ cells had significantly less fluorescence than parental cells at all timepoints. However, when beads attached to the cell surface were removed with sodium azide washes, there was no significant difference in the fluorescence associated with parental and cpnC⁻ cells, indicating that cpnC⁻ cells are able to take up the beads at similar rates as parental cells, but they are significantly less adherent to beads (Figure 3A). We also assayed phagocytosis of bacteria. cpnC⁻ and parental cells were plated on agar with E. coli B/R. Clear plaques arise within the bacterial lawn as single Dictyostelium cells feed on the E. coli and divide. Images were taken of the plaques on day 3 and plaque sizes were measured. The plaques made from parental and cpnC⁻ cells were not significantly different in size (Figure 3B,C), indicating that cpnC⁻ cells did not exhibit a defect in the phagocytosis of bacteria or growth on bacterial plates.

Figure 3. cpnC⁻ cells exhibited decreased adhesion to beads and no phagocytosis defect. A) Parental (circles) and cpnC⁻ (squares) cells were incubated with 1 µm FITC-labeled beads. Cell samples were collected at different timepoints, washed in Sorensen’s buffer with (open symbols) or without (closed symbols) sodium azide (SA), fixed, and analyzed using flow cytometry. Data from 3 trials for cell samples without sodium azide and data from 4 trials with sodium azide (n = 10,000 cells per sample) were analyzed for significant differences using a repeated measures ANOVA and post hoc Tukey comparisons. *indicates a significant difference between parental and cpnC⁻ cells at all non-zero timepoints, p < 0.05. Error bars=standard error. B) parental and cpnC⁻ cells were plated with E. coli on agar. Plaques were imaged after three days with a dissecting microscope. Individual plaque areas were measured using ImageJ software. Mean plaque size data from three trials were averaged and a t-test was used to analyze significant differences between parental and cpnC⁻ cells. C) Representative images of plaques. Scale bar = 3 mm.

Decreased bead adhesion exhibited by cpnC⁻ cells is due to decreased cell-surface proteins

To separate bead phagocytosis from bead adhesion, we performed the bead assay again, but treated the parental and cpnC⁻ cells with Latrunculin A (LatA) first. LatA causes the depolymerization of actin filaments and the complete inhibition of phagocytosis. Therefore, the beads will adhere to the outside of the cell, but not be phagocytosed. After LatA treatment and incubation with beads, the parental cells had a greater mean fluorescence than the cpnC⁻ cells as measured by flow cytometry. There was no difference in fluorescence after the cells were washed with sodium azide (Figure 4A). In addition to flow cytometry, cells were also imaged using DIC and fluorescence microscopy (Figure 4B). To determine if the decreased adhesion properties of cpnC⁻ cells was due to differences in surface proteins, cells were treated with LatA, then Proteinase K, and then assayed for bead adhesion. There was a significant difference between the mean fluorescence of the parental and cpnC⁻ cells before the treatment with Proteinase K; however, treatment with 100 µg/mL or 500 µg/mL of Proteinase K resulted in no significant difference in mean cell fluorescence between the two cell types (Figure 4C), suggesting that a change in cell surface proteins may play a role in the decreased adhesion phenotype of cpnC⁻ cells. Cells were also imaged using DIC and fluorescence microscopy (Figure 4D).

Figure 4. Decreased adhesion of cpnC⁻ cells is due to decreased cell-surface proteins. (A) Parental (black bars) and cpnC⁻ (gray bars) cells were treated with LatA for 30 minutes in a shaking suspension and then incubated with 1 µm FITC beads for 15 minutes. Cells were washed in buffer with or without sodium azide (SA), fixed, and analyzed using flow cytometry. Mean cell fluorescence for each cell type and condition was normalized to the average mean cell fluorescence within each trial. Data from 3 trials (n = 10,000 cells per sample) were analyzed for significant differences between parental and cpnC⁻ cells using a two-way ANOVA and post hoc Tukey comparisons, *indicates p < 0.05. (B) Cells were treated the same as in (A) and then fixed on coverslips, and imaged using DIC and fluorescence microscopy (scale bar = 16 µm). (C) Parental and cpnC⁻ cells were incubated with LatA for 30 minutes in a shaking suspension then incubated with different concentrations of proteinase K for 15 minutes before adding beads for an additional 15 minutes. Cell samples were washed in buffer, fixed, and analyzed using flow cytometry. Mean cell fluorescence was normalized to the average fluorescence within each trial. Data from 3 trials (n = 10,000 cells per sample) were analyzed for significant differences between parental and cpnC⁻ cells using a two-way ANOVA and post hoc Tukey comparisons, *indicates p < 0.05. (D) Cells were treated the same as in (C) and then fixed on coverslips, and imaged using DIC and fluorescence microscopy (scale bar = 16 µm). (E) Parental and cpnC⁻ cells were incubated with FITC-Lactadherin for 15 minutes or incubated with calcium ionophore for 10 minutes, then FITC-Lactadherin. Cells samples were analyzed using flow cytometry and mean cell fluorescence was normalized to the average mean cell fluorescence for all cell types and conditions within each trial. Data from 4 trials (n = 10,000 cells per sample) were analyzed for significant differences between parental and cpnC⁻ cells using a two-way ANOVA and post hoc Tukey comparisons.

We previously found that treatment of cpnA⁻ cells with proteinase K did not reduce the increased adhesion observed in cpnA⁻ cells, suggesting that proteins were not involved [20]. We then discovered that cpnA⁻ cells exhibited increased phosphatidylserine (PS) exposure, which caused the increased adhesion. Therefore, we also tested whether cpnC⁻ cells had a change in PS exposure by treating cells with FITC-labeled lactadherin. Lactadherin is a C2-domain containing protein that binds to PS. There was no significant difference between the mean fluorescence of parental and cpnC⁻ cells treated with FITC-lactadherin, indicating that unlike cpnA⁻ cells, cpnC⁻ cells do not exhibit increased PS exposure. Cells were treated with calcium ionophore as a positive control to increase PS exposure. Both cell types exhibited an increase in PS exposure in response to increased intracellular calcium (Figure 4E).

cpnC⁻ cells exhibit decreased substrate adhesion

To evaluate the adhesion properties of cpnC⁻ cells to a solid substrate, parental and cpnC⁻ cells were allowed to adhere to plastic Petri dishes. Cells at four marked places on the dish were imaged and the number of cells in each image were counted and averaged. Plates were rotated at 100 rpm and the media was changed to remove the detached cells. This procedure was repeated after plate rotation at 125 and 150 rpm. Significantly more cpnC⁻ cells were detached from the plate than parental cells after each of the rotation speeds, indicating that cpnC⁻ cells are less adherent to the dish than parental cells (Figure 5).

Figure 5. cpnC⁻ cells exhibited decreased adhesion to plastic dishes. Parental (black bars) and cpnC⁻ (gray bars) cells were plated on small petri dishes. Cells were imaged with phase-contrast microscopy at four marked spots before and after rotation at 100, 125, and 150 rpm. The number of cells in each image were counted and averaged. The percentage of detached cells after rotation was calculated by the subtraction of the average number of cells remaining after rotation from the average number of cells present before rotation divided by the number of cells present before rotation. Data from three trials was averaged and analyzed for significant differences between parental and cpnC⁻ cells by a repeated measures two-way ANOVA and Sidak’s multiple comparisons, *indicates p < 0.05.

cpnC⁻ cells aggregated without streaming and exhibited increased cell-cell adhesion

To evaluate the ability of cpnC⁻ cells to adhere to each other, we first ask whether they could form streams during aggregation. When Dictyostelium cells are starved for development, they will typically attach to each other head to tail and form streams as they aggregate into mounds. Because we did not always capture streaming on the agar plate development, we set-up streaming assays with cells in dishes with starvation buffer. Cells were imaged every 2 minutes for 8 hours. Parental cells began streaming between 6 and 8 hours after plating, while the cpnC⁻ cells aggregated into small aggregates without forming any streams. These aggregates of cpnC⁻ cells formed earlier than the streams formed by parental cells (Figure 6A; see accompanying movies S3 and S4).

Figure 6. cpnC⁻ cells aggregated without streaming and exhibited increased cell-cell adhesion. A) Parental and cpnC⁻ cells were allowed to develop under starvation buffer. Images were taken every 2 minutes for 8 hours using a dissecting microscope. Representative images at 0, 2, 4, 6, and 8 hours are shown. Scale bar = 1000 µm. See accompanying time-lapse movies (S3 and S4). B) parental (circles) and cpnC- (squares) cells were allowed to develop under starvation buffer until the parental cells began streaming, approximately 6–8 hours. Cells were disrupted and the number of single cells and doublets were counted. Disrupted cells were incubated at 125 rpm shaking. Cell samples were counted for single cells and doublets at 30 and 60 minutes. The percentage of single cells and doublets was calculated by dividing the average number of singles/doublets at each timepoint by the number of singles/doublets immediately after disruption. Data from four trials were averaged and analyzed for significant differences between parental and cpnC⁻ cells using a repeated measures two-way ANOVA and post hoc Tukey multiple comparisons, *indicates p < 0.05 at the 30 and 60 minute timepoints.

Although cpnC⁻ cells did not form streams, they did form small mounds that appeared to move or float in the buffer, suggesting that cells were able to adhere to each other, but were less adherent to the dish. To measure cell-cell adhesion, parental and cpnC⁻ cells were prepared for streaming and then harvested after 6 hours. Adhered cells were disrupted with a vortex and the number of non-aggregated cells, single cells and doublets, were counted immediately. The cells were then transferred to a flask and shaken at 125 rpm. Cell samples were collected, and the number of single cells and doublets were counted at 30 and 60 minutes timepoints. The cpnC⁻ cells had significantly fewer single cells and doublets than parental cells at each timepoint, indicating the cpnC⁻ cells were more adherent to each other than parental cells at 6 hours of starvation (Figure 6B).

cpnC⁻ cells have reduced SibA expression and increased CsaA expression

To determine if the decreased substrate adhesion and increased cell adhesion observed in cpnC- cells were associated with changes in protein expression, we looked at SibA and CsaA expression in cpnC⁻ cells as compared to parental cells. Whole cell samples of cpnC⁻ and parental cells were prepared for western blot analysis at various timepoints during the streaming assay. Cell samples were analyzed by western blot using antibodies to SibA and CsaA. SibA is an integrin-like protein that plays a role in substrate adhesion. In the parental cells, SibA expression was highest in vegetative cells and decreased over the development timepoints with no detectable SibA protein at the 6 hour timepoint (Figure 7A, B). The expression of SibA in cpnC⁻ cells was significantly less than in the parental cells in vegetative cells (0 hour timepoint). The SibA antibody recognized several other proteins in addition to SibA, including one close to the same size. To determine which protein was SibA, we also did a western blot with sibA⁻ cells and found that the highest molecular weight band was SibA (Figure 8B, C; asterisk).

Figure 7. cpnC- cells have decreased SibA expression as vegetative cells and increased CsaA expression as starved cells. Parental and cpnC⁻ cells were allowed to develop under starvation buffer. Whole cell samples were made at 0, 2, 4, 6, 8, and 10 hours post starvation. Whole cell samples were analyzed by western blot with an antibody to SibA (A,B,C) and CsaA (D,E,F). The SibA and CsaA bands on blots from three trials were analyzed by densitometry. Mean densities were analyzed by a two-way ANOVA and post-hoc fisher LSD comparisons. A) SibA mean band densities for parental (black bars) and cpnC⁻ (gray bars) cells at each timepoint. Mean densities were normalized to the parental cell SibA band at the 0 timepoint. B) Representative image of a SibA western blot with parental (P) and cpnC⁻ (C-) whole cell samples. C) SibA western blot comparing parental, sibA⁻ and cpnC⁻ cells at the 0 timepoint. D) CsaA mean band densities for parental (black bars) and cpnC⁻ (gray bars) cells at each timepoint. Mean densities were normalized to the parental cell CsaA band at the 10 hour timepoint. E) Representative image of a CsaA western blot with parental (P) and cpnC⁻ (C-) whole cell samples. F) CsaA western blot comparing whole cell samples from parental, cpnC⁻, and csaA⁻ cells at the 6 hour timepoint.

Figure 8. cpnC⁻ cells exhibit decreased expression of RegA. Parental and cpnC⁻ cells were allowed to develop under starvation buffer. Whole cell samples were made at 0 and 6 hours post starvation. Whole cell samples were analyzed by western blot with an antibody to RegA. The RegA bands on blots from three trials were analyzed by densitometry. Mean densities were analyzed by a two-way ANOVA and post-hoc fisher LSD comparisons. A) RegA mean band densities for parental (black bars) and cpnC⁻ (gray bars) cells at each timepoint. Mean densities were normalized to average density within each trial. B) Representative image of a RegA western blot with parental (P) and cpnC⁻ (C-) whole cell samples.

CsaA is involved in cell-cell adhesion and is not expressed in vegetative cells. In the parental cells, CsaA expression was first detected at the 4 hour timepoint, greatly increased at the 6 hour timepoint, and remained high at the 8 hour and 10 hour timepoints. The expression of CsaA in cpnC⁻ cells was significantly increased compared to the parental cells (Figure 7D, E). In addition, the cpnC⁻ cell samples contained a higher molecular weight band that was recognized by the CsaA antibody. This higher band was not seen in any of the parental cell samples, nor in a csaA⁻ cell sample (Figure 7F). This suggests that CsaA may be differently modified in cpnC⁻ cells than parental cells.

cpnC⁻ cells have reduced RegA expression

We speculated that both the observed developmental and adhesion defects could be due to alterations in cAMP signaling in the cpnC⁻ cells. The ability to aggregate, but not form streams is a novel phenotype that has only been seen in a few other mutants. One of which is the regA mutant. RegA is a cAMP phosphodiesterase and is important in regulating cAMP signaling during Dictyostelium development [43]. mRNA levels of regA increase during early development peaking at around 5 hours [48,49]. We used western blotting to quantify the protein levels of RegA in parental and cpnC⁻ cells at 0 hours and 6 hours after being plated in starvation buffer to induce aggregation. We found that cpnC⁻ cells had less RegA than parental cells in both vegetative cells and cells 6 hours post starvation; statistical analysis of densitometry data indicated only a significant difference at the 6 hour timepoint. RegA protein levels increased ~ 7.5-fold in parental cells at 6 hours post starvation compared to vegetative cells, while RegA protein levels did not increase at 6 hours post starvation in cpnC⁻ cells (Figure 8A, B), indicating the RegA expression is not upregulated during development in cpnC⁻ cells.

Discussion

Copines make up a family of evolutionarily conserved calcium-dependent membrane-binding proteins. Copines are highly expressed in multiple human cancers and functional studies show copines promote proliferation and migration of cancer cells [9]. However, despite being studied for over two decades, no unified mechanistic function for copines has been identified. We are using the model organism, Dictyostelium discoideum, which has six copine genes to uncover this function. In this study, a cpnC null mutant was generated to identify defects in cellular processes that occur in the absence of cpnC.

We first evaluated the growth and development of the cpnC mutant. The absence of CpnC did not have an impact on the growth of Dictyostelium, either in suspension or on bacterial lawns. We did, however, discover that the cpnC mutant had a higher incidence of multinucleation compared to the parental strain. During our analysis of multicellular development, we found two major abnormalities in the development of cpnC⁻ cells: the complete lack of streaming during aggregation and large ring structures during the mound stage. Aggregation and streaming of starving Dictyostelium cells are directed by the periodic release of cAMP [51]. Timelapse imaging showed that while the parental cells were able to stream normally, the cpnC⁻ cells aggregated through clumping with smaller aggregates merging to form larger aggregates. In the mound stage, cells normally move in a circular pattern and often form small rings [51]. The parental cells formed tight small circles, while the cpnC⁻ cells formed very large rings that sometimes merged together.

Phagocytosis assays with fluorescent beads demonstrated that while the cpnC⁻ cells had no defect in the phagocytosis, they were less adherent to the beads than the parental cells. The difference in bead adhesion between the parental and cpnC⁻ cells was abolished when the cells were incubated with proteinase K, suggesting that cpnC⁻ cells may have altered cell surface adhesion proteins. In vegetative cells, the adhesion protein SibA facilitates substrate adhesion and during development the protein CsaA facilitates cell-cell adhesion. Western blot evaluation of adhesion protein expression showed that the cpnC⁻ cells expressed less SibA in vegetative cells and more CsaA in starved cells. The altered protein expression profiles correlated with the adhesion properties of cpnC⁻ cells in that cpnC⁻ cells were less adhesive to substrates as vegetative cells and more adhesive to other cells when starved.

The western blot analysis of CsaA in cpnC⁻ cells revealed an additional higher molecular weight band. This higher band could be a different protein that has a similar developmental expression pattern to CsaA. However, this band was not seen in the parental cell samples nor in the csaA⁻ cell sample. CsaA is a GPI-linked protein and the glycosylated version is ~ 80 kDa, while the non-glycosylated version is ~ 53 kDa [52]. On our western blots, we see the 80 kDa band and another band that is 40–70 kDa bigger, but only in the cpnC⁻ cell samples. This suggests that this larger band is not due to glycosylation, nor phosphorylation, which would most likely not cause such a big change in size. The larger protein could be a ubiquitinated version of CsaA; CsaA has eight predicted ubiquitin sites. The ubiquitinated version of CsaA could be more prevalent in the cpnC⁻ cells because of the increased expression of CsaA or could be due to a defect in the trafficking or processing of CsaA causing it to be targeted for ubiquitination.

The ability to aggregate, but not form streams is a novel phenotype that has only been seen in a few other mutants. These mutants include dhkD [53], rdeA [54], and regA [43]. All three proteins are part of a histidine kinase two-component signaling pathway; DhkD is a histidine kinase, RdeA is a phospho-transfer protein, and RegA is the response regulator. RegA is the only known response regulator for multiple histidine kinases in Dictyostelium and is an intracellular cAMP phosphodiesterase. The phosphorylation of RegA through this phospho-transfer system leads to a 20-fold increase in RegA activity [44]. RegA is also phosphorylated by the Map kinase, ErkB, in Dictyostelium. However, in this case, phosphorylation of RegA by ErkB is inhibitory [45]. ErkB phosphorylation targets RegA for ubiquitination by FbxA, a component of the SCF E3 ubiquitin ligase complex [46]. We found that cpnC⁻ cells have significantly decreased levels of RegA protein during aggregation. Decreased RegA activity would result in higher levels of cAMP, and the dysregulation of periodic cAMP release, which could explain the defects in streaming in cpnC⁻ cells. Increased cAMP levels could also explain the early downregulation of SibA in vegetative cells and the early upregulation of CsaA in starved cells as seen in cpnC⁻ cells.

Two of the major findings of this phenotypic study are that cpnC⁻ cells have defects in both development and adhesion. We speculate that both types of defects are due to alterations in cAMP signaling in the cpnC⁻ cells. We propose that CpnC is involved in cAMP signaling in Dictyostelium through the regulation of RegA. CpnC could be involved in the positive regulation of the transcription of regA. We previously found that GFP-tagged CpnC is located both in the cytoplasm and in the nucleus, suggesting that CpnC could be involved in transcriptional regulation [3]. Studies in human cells showed that copine I inhibits NF-kB mediated transcriptional activity by promoting the N-terminal cleavage of p65, one of the subunits of NF-kB [50]. Therefore, CpnC could regulate the activity of a transcription factor that is required for the upregulated expression of regA during development. Alternatively, CpnC could be involved in the regulation of the degradation of the RegA protein. Copines in mice have been reported to interact with kinases and ubiquitin ligases [5,8,47]. In future studies, we plan to investigate the role of CpnC in cAMP signaling during Dictyostelium development, specifically focusing on the regulation of RegA.

Data accessibility statement

All data is available from the Dryad repository without limitations: https://doi.org/10.5061/dryad.xsj3tx9kb.

Acknowledgments

We thank the Dictyostelium Stock Center for NC4A2, SibA⁻, and CsaA- Dictyostelium strains, the E.coli B/r stain, and the pLPBLP plasmid, the Geneva Antibody Facility for the SibA antibody, the Developmental Studies Hybridoma Bank for the CsaA antibody, and Robert Kay for the RegA antibody. We also thank Stephanie Hickey at the Research Technology Support Facility Bioinformatics Core at Michigan State University for the analysis of whole genomic DNA sequences and the Flow Cytometer and Imaging Facility at Central Michigan University.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19336918.2024.2315629

Additional information

Funding

This work was supported by the NIH (NIH R15GM078089-03 grant to C. Damer) and Central Michigan University (graduate student grant to R. Nichols).