Primary murine epidermal keratinocytes were prepared from 1 to 3 day old neonatal ICR CD-1 outbred mice as described in [11]. Treatment of mice conformed to policies in the Guide for the Care and Use of Laboratory Animals and monitored by the Institutional Animal Care and Use Committee (IACUC) of Augusta University. Harvested keratinocytes were plated at a density of 25,000 cells/cm² and incubated overnight at 37°C with 5% carbon dioxide in Plating Medium composed of calcium-free minimum essential medium alpha (MEMα) supplemented with 2% dialyzed fetal bovine serum, 25µM CaCl₂, 5ng/mL epidermal growth factor, 2mM glutamine, ITS+, 100U/mL penicillin, 100µg/mL streptomycin and 0.25µg/mL fungizone as in [12]. After approximately 24 hours, Plating Medium was replaced with either a laboratory-prepared serum-free keratinocyte medium (SFKM) or commercially purchased Keratinocyte-serum free medium (K-SFM) (Gibco, Gaithersburg, MD). Initial experiments used SFKM containing 25µM CaCl₂, 90µg/mL bovine pituitary extract, ITS+, 5ng/mL epidermal growth factor, 2mM glutamine, 0.05% BSA, 100U/mL penicillin, 100µg/mL streptomycin and 0.25µg/mL fungizone as described by Griner et al. [12]. Our laboratory subsequently switched to commercial K-SFM supplementing pre-prepared K-SFM with 50μM CaCl₂, 2.5μg recombinant human EGF and 25mg bovine pituitary extract per the supplier’s recommendations [13]. Medium was replaced every 1-2 days.

Liposomes were prepared from egg-derived PG (Avanti Polar Lipids, Alabaster, AL). Briefly, PG in organic solvent was distributed into amber glass vials as 1mg aliquots, the solvent evaporated with nitrogen gas and the lipid stored under nitrogen at -20°C until use. For experiments, 0.5 mL serum-free medium was added to the amber vial to hydrate the lipid film followed by bath sonication using a Branson Sonifier with a microprobe and a cup horn.

Cultured primary mouse keratinocytes, epidermis or skin were collected in 0.5-1mL of Trizol and RNA was extracted according to the manufacturer’s protocol. The epidermis was isolated following overnight incubation of neonatal mouse skin in 0.25% trypsin in Hank’s buffered saline solution at 4°C to allow the enzyme to permeate along the dermal-epidermal junction of the skin. After a brief incubation at 37°C to promote trypsin proteolysis, the epidermis was manually separated from the dermis using forceps. Total skin was extracted immediately after harvest. Mouse brain tissue was also collected and immediately RNA-extracted. First-strand cDNA synthesis was performed using Thermoscript RT-PCR System and oligo(dT) nucleotides (Sigma-Aldrich, St. Louis, MO) according to the manufacturer’s protocol. RT-PCR was performed using JumpStart RedTaq reaction mix (Sigma-Aldrich) and the primers for PKCβII and GAPDH (mPKCβ forward: 5’-GCTGACAAGGGCCCAGCCTC-3’; reverse: 5’-GTGTGGTTCCGTGCCGCAGAG-3’ and mGAPDH forward: 5’-GCGGCACGTCAGATCCA-3’; reverse: 5’ CATGGCCTTCCGTGTTCCCTA-3’), also according to the manufacturer’s protocol. The reaction parameters consisted of heat activation at 94ºC for 2 minutes followed by 35 cycles of denaturation at 94ºC for 15 seconds, annealing at 50ºC for 30 seconds and elongation at 72ºC for 30 seconds. The amplified product was resolved on a 1% TAE agarose gel. Quantitative RT-PCR was performed using Taqman probes (ThermoFisher Scientific, Waltham, MA) and analyzed by the delta-delta Ct method as described previously [13].

Near-confluent cultures of keratinocytes were incubated in SFKM (25μM CaCl₂) or K-SFM (50μM CaCl₂) alone or with medium containing the desired treatment, elevated calcium (125μM CaCl₂) or 100μg/mL PG for 24 hours. Cells were then harvested in lysis buffer, with 30μL/cm² of heated buffer (containing 0.1875M Tris-HCl, pH 8.5, 3% SDS and 1.5mM EDTA) added to each well. Protein concentrations were determined using a BioRad protein assay with BSA as the standard. After protein determination, 3X sample buffer (containing 30% glycerol, 15% β-mercaptoethanol and 1% bromophenol blue) was added to each sample to constitute Laemmli buffer. Total protein was also extracted from mouse brain, homogenized epidermis and freshly isolated keratinocytes after shearing using an 18-gauge needle. Samples were stored at -20°C until analysis at which time protein samples were heated to near boiling and equal amounts were loaded onto 8% SDS polyacrylamide gels, separated by electrophoresis and transferred to Immobilon-FL transfer membranes (Millipore, Billerica, MA). After washing and blocking, the membranes were incubated overnight with primary antibody [recognizing PKCβII (Abcam, Cambridge, MA), pPKCβII (pSer⁶⁶⁰, Epitomics, Burlingame, CA), 1:10,000; K10 (Covance, Denver, PA), 1:15,000; and actin (Sigma-Aldrich or Santa Cruz, Santa Cruz, CA), 1:15,000] followed by secondary AlexaFluor florescent antibodies (Invitrogen, Carlsbad, CA or Licor, Lincoln, NE, 1:10,000), all diluted in Odyssey blocking buffer containing Tween-20 (LiCor). Immunoreactive bands corresponding to the proteins of interest were visualized via an Odyssey®SA infrared imaging system from Li-Cor and quantified with the internal software according to the manufacturer’s instructions. The data are reported as means ± SEM after normalization to actin levels.

DNA proliferation assays were performed on primary mouse keratinocytes overexpressing PKCβII or empty vector incubated in SFKM containing 25μM or 125µM calcium. Briefly, cells were transfected as previously described and exposed to experimental treatments for 24 hours. The cells were then incubated with 1μCi/mL [³H]thymidine (Moravek Biochemicals, Brea, CA) for 1 hour at 37°C. Reactions were terminated and macromolecules precipitated with cold trichloroacetic acid and the cells solubilized in 0.3M sodium hydroxide. [³H] Thymidine incorporation was measured in an aliquot using Ecolite scintillant (MP Biomedicals, Santa Ana, CA) and a Beckman Coulter LS 6500 multi-purpose scintillation counter (Brea, CA).

For immunocytochemistry PKCβII-specific and phosphoPKCβII-specific antibodies were generously provided by Dr. Denise Cooper (University of South Florida, Tampa, FL) [14]. The antibody recognizing PKCβII was raised against residues 655 to 671 (the C-terminus specific to PKCβII), and the phospho-specific antibody recognizes PKCβII phosphorylated on serine 660 of the C-terminus (residues 657-673) [14]. Primary murine keratinocytes were plated on glass BD BioCoat fibronectin-coated slides and at near-confluence were incubated in SFKM (25µM CaCl₂) or SFKM containing 100µg/ml PG at 37°C. After the desired incubation period, cells were washed with PBS and fixed in 4% paraformaldehyde. After permeabilization in 0.2% Triton X-100, the slides were blocked in buffer containing 10% goat serum and 1% BSA in PBS and incubated in buffer containing either phospho-PKCβII (1:500) or keratin-10 (Abcam, 1:250) overnight. The slides were then incubated in Cy3-conjugated secondary goat anti-rabbit IgG antibody (1:150) in 10% goat serum at room temperature and mounted with ProLong Antifade with DAPI (Invitrogen). Staining was visualized by multiphoton microscopy with a Zeiss LSM 510 confocal laser scanning microscope with a Meta System equipped with a Coherent Mira 900 tunable Ti:Sapphire laser for multi-photon excitation at 488nm, 543nm and 760nm wavelengths (Carl Zeiss Microscopy, Germany).

Primary mouse keratinocytes were transfected with wild-type PKCβII plasmid in a pcDNA3 vector backbone (or the vector plasmid) via AMAXA nucleofection (Lonza, Cologne, Germany), using an Amaxa Nucleofector Kit for primary endothelial cells as in [15] according to the manufacturer’s instructions. The authors reported 40-60% transfection efficiency with this method [15]. The PKCβII plasmid was a generous gift from Dr. Lan Ko, (Augusta University). Transfected cells were then incubated in RPMI medium containing 10% fetal bovine serum and antibiotic/antimycotic (100U/mL penicillin, 100µg/mL streptomycin and 0.25µg/mL fungizone) for 20 minutes, plated in Plating Medium (described above) and allowed to attach overnight. After 24 hours the plating medium was replaced with K-SFM (containing 50μM CaCl₂).

All experiments were performed independently a minimum of three times. Values were analyzed for statistical significance by analysis of variance or repeated measures analysis of variance with a Student-Newmann-Keuls or Dunn’s post-hoc test using Prism (GraphPad Software, San Diego, CA). All quantitative data were expressed in the form of bar graphs, with the bars representing mean ± standard error of the mean (SEM).

PKCβII is known to bind to and be activated by PG to trigger cell cycle progression in human leukemia cells [8]. However, although multiple PKC isozymes have been identified in keratinocytes, there has been some debate regarding the presence of PKCβ in these cells [16-19], with an initial report failing to detect PKCβ in mouse keratinocytes using northern analysis [16]. Two subsequent studies found PKCβ in human skin [17], mouse keratinocytes [18] and mouse skin [19]. To resolve this issue, we first sought to determine if PKCβ could be detected by semi-quantitative RT-PCR using mouse brain as a positive control. PKCβ was found to be transcribed in primary mouse keratinocytes (Fig. 1A). Although the mRNA was much less abundant than in brain (Fig. 1B), quantitative RT-PCR using Taqman assays indicated that the cycle threshold was within a detectable range (approximately 30 cycles). The difference in the amount of cDNA amplified and separated also likely explains the slight apparent differences in molecular weight of the PKCβ band observed in brain and keratinocytes with semi-quantitative RT-PCR (Fig. 1A), since greatly different amounts of nucleic acid can separate slightly differently by electrophoresis. We next sought to determine whether the PKCβII protein was expressed using a PKCβII-specific antibody obtained from Dr. Denise Cooper [14]. This antibody was raised against residues 655 to 671 (the C-terminus specific to PKCβII). Using this antibody it was shown that keratinocytes also express PKCβII protein (Fig. 1C). To ensure that expression of the enzyme was not an artifact of culture, we also demonstrated PKCβII protein expression in freshly isolated keratinocytes and epidermis (Fig. 1D), as well as mRNA expression in total skin (Fig. 1B). Therefore, we hypothesized that PKCβII might be an effector enzyme for PG in keratinocytes.

Fig. (1). PKCβII protein is expressed in primary mouse epidermal keratinocytes, freshly isolated epidermal keratinocytes and the epidermis. (A) RNA was isolated from primary mouse keratinocytes (1°MK) cultured to near confluence before incubation for 24 hours in medium containing basal (25µM) or moderately elevated (125µM) extracellular Ca2+concentrations as indicated and PKCβ expression monitored by semi-quantitative RT-PCR. (B) RNA was isolated from primary mouse keratinocytes (1°MK) incubated in medium containing a basal (50µM) extracellular Ca2+concentration, freshly isolated keratinocytes (before plating), isolated epidermis, total skin and brain as indicated and PKCβ expression monitored by quantitative RT-PCR using primer-probe sets from Applied Biosystems and a StepOne system as described in Methods. Results are expressed as the fold change in normalized cycle threshold relative to primary cultures of primary mouse keratinocytes (1°MK) cultured for 4 days, analyzed using the ΔΔCt method with GAPDH as the normalization control. Note that the brain expresses a significantly greater amount of PKCβ than do keratinocytes or skin tissue; therefore, in the inset the values obtained only from 1°MK cultured for the indicated number of days, freshly isolated keratinocytes (KC), isolated epidermis or total skin are plotted using a different scale. (C) Protein lysates were prepared from primary mouse keratinocytes (1°MK) cultured in medium containing basal (25µM) or moderately elevated (125µM) extracellular Ca2+concentrations. (D) Protein lysates prepared from primary mouse keratinocytes (1°MK) cultured in medium containing basal (25µM) or 125µM Ca2+, freshly isolated keratinocytes (fresh MK) or a homogenate of epidermis were analyzed by western blotting. Western analysis was performed using the PKCβII antibody obtained from Dr. Denise Cooper (University of South Florida, Tampa, FL). Commercially available antibodies yielded similar results. Brain was used as a positive control.

Fig. (2). PG stimulates phosphoPKCβII redistribution in keratinocytes. Keratinocytes grown on fibronectin-coated slides were stimulated for 60 minutes with medium containing a basal (25µM) calcium concentration (A) without (Con) or (B) with 100µg/mL egg PG (PG), provided in the form of liposomes. Cells were then stained with an antibody recognizing phosphorylated PKCβII, an Alexa468-conjugated secondary antibody and the nuclear stain, DAPI, and visualized using a multiphoton Zeiss microscope. The primary antibody was omitted to serve as a negative control and showed no staining for PKCβII although DAPI was visualized (data not shown).

We have shown that PG can inhibit proliferation and promote differentiation of keratinocytes, and based on the literature demonstrating that PKCβII is a PG-activated enzyme, we hypothesized that stimulation of PKCβII activity by PG may be the mechanism by which the lipid signal exerts its effects. Phosphorylation and translocation to cell membranes are considered hallmarks of PKC activation [20]. We utilized immunocytochemical techniques to visualize the cellular localization of phosphorylated/activated PKCβII upon treatment with PG, again using an antibody provided by Dr. Cooper recognizing PKCβII phosphorylated on serine 660 of the C-terminus (residues 657-673). Primary mouse keratinocytes plated on collagen-coated slides were subjected to immunohistochemical analysis. Under basal conditions, autophosphorylated PKCβII (phosphoPKCβII) was found diffusely throughout the entire cell with increased staining around the perinuclear area (Fig. 2, left panel). A 1h treatment with PG (100μg/mL egg-derived PG in the form of liposomes) increased staining in the perinuclear area (Fig. 2A, right panel) compared to the control (Fig. 2B, left panel). In addition, PG induced localization of PKCβII in the plasma membrane (Fig. 2B, right panel, arrows). Since PKCβII requires phospholipids for its activity, translocation to the membrane is thought to mark activation of the enzyme [21, 22]. These results suggest that, as in leukemia cells [7-9], PG activates PKCβII. Treatment of keratinocytes with PG followed by western analysis with the antibody recognizing phosphoPKCβII also showed a trend towards enhanced autophosphorylation (activation) of PKCβII (to a value of 1.32 ± 0.13-fold over the control of 1.0; n=6), but the increase did not quite achieve statistical significance (p=0.054).

We have previously observed an ability of moderately elevated extracellular calcium concentrations to increase PG levels in keratinocytes [3], suggesting the possibility that PG-activated PKCβII may play a role in calcium-induced differentiation. To explore this possibility we over-expressed PKCβII and first assessed the effect of calcium on the autophosphorylation/activation of this enzyme. Primary mouse keratinocytes were transfected with either an empty vector or PKCβII plasmid vector, as described in the Materials and Methods section, and then cultured in the presence or absence of an elevated extracellular calcium concentrations. Total and autophosphorylated PKCβII levels were increased in mouse keratinocytes transfected with PKCβII plasma vector, and autophosphorylation of this overexpressed PKCβII was stimulated in response to elevated extracellular calcium (125μM) (Figs. 3A-C).

To determine whether this over-expression of PKCβII affected the ability of elevated calcium levels to inhibit keratinocyte proliferation, we treated vector- and PKCβII-transfected cells with medium containing basal calcium levels or a moderately elevated calcium concentration. Proliferation was assessed by measuring the incorporation of [³H]thymidine into the DNA of dividing cells. In cells expressing basal, physiological levels of PKCβII, that is, transfected with empty vector plasmid, stimulation with elevated extracellular calcium resulted in calcium-induced inhibition of proliferation (Fig. 3D), as described elsewhere [23]. A similar inhibition was observed in the PKCβII-transfected cells, although this effect did not achieve statistical significance, suggesting that this enzyme likely does not mediate the anti-proliferative effect of an elevated calcium concentration.

Expression levels of keratin-10, a marker of keratinocyte differentiation, were also examined in cells over-expressing PKCβII. There was no change in the keratin-10 expression of keratinocytes over-expressing PKCβII and cultured under basal conditions. PKCβII over-expression in combination with an elevated extracellular calcium concentration, however, resulted in a substantial up-regulation in keratin-10 expression, with p<0.01 versus all other conditions (Fig. 4). These results suggest that PKCβII alone is not sufficient to induce an increase in keratin-10 expression, but instead works in concert with calcium to promote early differentiation, but not growth arrest, in primary mouse keratinocytes.

Our previous findings suggested that PG stimulates keratinocyte differentiation and inhibits proliferation [5, 6]; therefore, we next examined the morphological effect of PG on keratinocytes over-expressing PKCβII. Mouse keratinocytes were again transfected with either PKCβII or empty vector and were then cultured on collagen-coated slides in medium containing basal calcium with or without 100μg/ml egg PG. The cells were then fixed, permeabilized and stained with an antibody specific for keratin-10 (Fig. 5). Interestingly, treatment with PG led to morphological changes consistent with later differentiation in cells transfected with PKCβII (Fig. 5A-D), as well as the formation of keratin-10 filaments, changes that were not seen in cells transfected with empty vector and stimulated with PG (Fig. 5D). Thus, PG in PKCβII-overexpressing cells induced enlargement and flattening reminiscent of the alterations observed with later keratinocyte differentiation; these changes also were not seen in keratinocytes overexpressing PKCβII in the absence of PG (Fig. 5A). These data suggest that keratinocyte differentiation in response to elevated calcium concentrations may be mediated through the activation of PKCβII induced by increased production of PG. This hypothesis is consistent with the observed ability of PKCβII overexpression to increase keratin-10 levels in the presence of a moderately elevated calcium concentration and of PG to promote differentiative changes in PKCβII-overexpressing keratinocytes.

Fig. (3). Overexpressed PKCβII is autophosphorylated/activated in response to an elevation of extracellular calcium concentration in mouse keratinocytes but has no effect on calcium-inhibited proliferation. Mouse keratinocytes were nucleofected with wild-type PKCβII (βII) or empty vector (EV) and then cultured in the presence of basal (25µM) or a moderately elevated extracellular calcium level (Ca; 125µM) for 24h. Cells were harvested and lysates were resolved on 8% SDS gels, transferred to PVDF membranes and probed with antibodies recognizing total PKCβII, autophosphorylated PKCβII and actin. (A) A representative experiment is illustrated. (B) Total PKCβII and (C) pPKCβII levels were quantified, normalized to actin and expressed relative to the PKCβII-transfected cells under basal conditions. Data represent the means ± SEM from at least 3 separate experiments. For total PKCβII, *p<0.05 vs EV or EV+Ca and for autophosphorylated PKCβII, #p<0.01 vs EV, *p<0.01 vs EV+Ca, ^p<0.001 vs EV, +p<0.05 vs βII. (D)Keratinocytes nucleofected with wild-type PKCβII (βII) or empty vector (EV) were cultured in the presence of basal (25µM) or a moderately elevated extracellular calcium level (Ca; 125µM) for 24h. [3H]Thymidine was added to the medium for 1 hour and DNA synthesis measured as described in Materials and Methods. [3H]Thymidine incorporation into DNA is expressed as the percentage of the control value and shown as the mean ± SEM (n=4; *p<0.01 vs EV or βII).

Fig. (4). Overexpressed PKCβII enhances calcium-induced keratin-10 protein expression (differentiation). Keratinocytes nucleofected with wild-type PKCβII (βII) or empty vector (EV) were cultured in the presence of basal (25µM) or a moderately elevated extracellular calcium level (Ca; 125µM) for 24h. Following cell harvest total lysates were resolved on 8% SDS gels, transferred to PVDF membranes and probed with antibodies recognizing keratin-10 (K10) and actin. A representative experiment is shown. Keratin-10 levels (normalized to actin and expressed relative to the PKCβII-transfected cells under basal conditions) were quantified and expressed as the mean ± SEM (n=4; *p<0.01 vs EV, βII or EV+Ca).

Fig. (5). PKCβII overexpression increases the PG-induced formation of keratin-10-containing intermediate filaments. Mouse keratinocytes were nucleofected with wild-type (A and C) PKCβII (βII) or (B and D) empty vector (EV) and then cultured on coated glass slides in the (C and D) presence or (A and B) absence of 100µg/mL PG in 25µM calcium-containing medium for 24h. Cells were fixed, permeabilized, probed with an antibody recognizing keratin-10 and visualized with an Alexa468-conjugated secondary antibody. Immunofluorescence was examined by confocal microscopy.