Ecto-nucleoside triphosphate diphosphohydrolase 2 modulates local ATP-induced calcium signaling in human HaCaT keratinocytes

Abstract

Keratinocytes are the major building blocks of the human epidermis. In many physiological and pathophysiological conditions, keratinocytes release adenosine triphosphate (ATP) as an autocrine/paracrine mediator that regulates cell proliferation, differentiation, and migration. ATP receptors have been identified in various epidermal cell types; therefore, extracellular ATP homeostasis likely determines its long-term, trophic effects on skin health. We investigated the possibility that human keratinocytes express surface-located enzymes that modulate ATP concentration, as well as the corresponding receptor activation, in the pericellular microenvironment. We observed that the human keratinocyte cell line HaCaT released ATP and hydrolyzed extracellular ATP. Interestingly, ATP hydrolysis resulted in adenosine diphosphate (ADP) accumulation in the extracellular space. Pharmacological inhibition by ARL 67156 or gene silencing of the endogenous ecto-nucleoside triphosphate diphosphohydrolase (NTPDase) isoform 2 resulted in a 25% reduction in both ATP hydrolysis and ADP formation. Using intracellular calcium as a reporter, we found that although NTPDase2 hydrolyzed ATP and generated sustainable ADP levels, only ATP contributed to increased intracellular calcium via P2Y2 receptor activation. Furthermore, knocking down NTPDase2 potentiated the nanomolar ATP-induced intracellular calcium increase, suggesting that NTPDase2 globally attenuates nucleotide concentration in the pericellular microenvironment as well as locally shields receptors in the vicinity from being activated by extracellular ATP. Our findings reveal an important role of human keratinocyte NTPDase2 in modulating nucleotide signaling in the extracellular milieu of human epidermis.

Figure 1. HaCaT cells release ATP and hydrolyze extracellular ATP.

(A) Time course of basal ATP levels in the extracellular buffer after a media change. Data represent values from one representative experiment performed in triplicate (mean±SD). (B) ATP release from HaCaT cells into the extracellular space with (serum-free) and without (low-serum) the washing step to remove residual fetal calf serum (FCS). Bars denote ATP concentrations in the buffer after 1 hr in the presence of cells (mean±SEM, n = 16–17, Student's t test, *** p<0.001). (C) Hydrolysis of extracellular ATP (100 nM) in the presence of HaCaT cells. Apyrase (the Desiree isoform derived from red potatoes, 10⁻³ units/ml) serves as a positive control of ATP hydrolysis. Bars denote the concentration of remaining ATP compared with cell-free input (gray bar; mean±SEM, n = 5, one-way ANOVA with Dunnett's post-analysis, comparing all columns to cell-free input, *** p<0.001).

Figure 2. HaCaT cells express NTPDase2 that hydrolyzes extracellular ATP and accumulates ADP.

(A) The relative mRNA levels of four surface-located NTPDases in HaCaT cells as determined using real-time PCR. Bars denote mRNA copies per 10⁴ TATA-binding protein (TBP) mRNA (mean±SEM, n = 2–21). (B) Hydrolysis of 100 µM ATP (light gray bar) when NTPDase expression is suppressed using siRNAs (white bars) or 100 µM ARL 67156, an NTPDases inhibitor (dark gray bar). ATPase activity is defined as the rate of ATP hydrolysis (amount of ATP hydrolyzed per hr) relative to the rate of control [mean±SEM, n = 2–21, one-way ANOVA with Dunnett's post-analysis, comparing all columns to control siRNA treatment (black bar), * p<0.05; *** p<0.001; ns = not significant]. (C) ATP hydrolysis kinetics in the presence of control or NTPDase2 siRNA-treated cells. Curves denote Michaelis-Menten kinetics normalized to the V_max of control cells (mean±SEM, n = 2). (D) Extracellular ATP hydrolysis and ADP accumulation when NTPDase2 was suppressed using siRNA and/or ARL 67156. Bars denote ATP (left Y-axis) and ADP (right Y-axis) concentrations compared to the respective controls (mean±SEM, n = 5, one-way ANOVA followed by Bonferroni's multiple comparison test, * p<0.05; ** p<0.01; *** p<0.001).

Figure 3. ATP evokes [Ca²⁺]_i increase in HaCaT cells via P2Y2 receptor activation.

(A) Visualization of extracellular ATP and intracellular calcium when one HaCaT cell was activated by mechanical force. Images represent montages of ATP bioluminescence (upper panel) and intracellular calcium fluorescence (lower panel) after touching a cell with a glass needle. White dotted lines outline the position of glass needle (a.u., arbitrary unit). (B) The kinetics (left) and amplitude (right) of the [Ca²⁺]_i increase when HaCaT cells were pre-incubated in Ca²⁺-free HBSS (red line), 1 µM thapsigargin (blue line), 10 µM U73122 (gray line), or untreated (black line), and stimulated with 5 µM ATP. Dotted lines indicate SD. Bars denote the amplitudes of the increases in the left panel (mean±SEM, n = 8, one-way ANOVA followed by Dunnett's multiple comparison test, *** p<0.001, a.u., arbitrary unit). (C) The relative expression levels of eight P2Y receptor mRNAs in HaCaT cells, as evaluated using real-time PCR. Bars denote the mRNA levels expressed as copies per 10³ TBP mRNAs. (D) ATP- and ADP-evoked [Ca²⁺]_i increases in HaCaT cells in which P2Y1 and P2Y2 were knocked down using siRNA or inhibited using the P2Y1-selective antagonist MRS 2179 (10 µM). Sigmoid curves denote the dose-response relationships of peak [Ca²⁺]_i increases to various concentrations of ATP and ADP. Values are normalized to the maximal response of siControl (mean±SEM, n = 2 to 5).

Figure 4. Knocking down P2Y2 expression attenuates the ATP-mediated [Ca²⁺]_i increase.

(A) Fluorescence images showing how software-assisted cell demarcation and population identification is performed according to DNA and cell tracer dye staining. Cyan lines define control cells, and red lines outline P2Y2-deficient cells. (B) Time series of fluorescence imaging of Alexa Fluor-647 ATP released from microinjection needle. Concentric circles define the radial distances from needle tip in 20 µm steps (a.u., arbitrary unit). (C) Time series of fluorescence imaging of calcium wave spreading in a co-culture monolayer containing an equal ratio of control and P2Y2-deficient HaCaT cells. Four representative cell doublets with various distances from the needle are outlined in cyan and red. White dotted lines outline the position of glass needle. Insets: enlarged pseudo-color images of control and P2Y2-deficient cell doublets depicted in ΔF/F₀. Dotted lines depict the cell outline (a.u., arbitrary unit). (D) Traces showing Ca²⁺ change (fold change from baseline, ΔF/F₀) in cell doublets from panel C. The black arrow indicates the initiation of ATP point release, and the black line defines the duration of ATP point release. (E) Frequency distribution and log Gaussian fit comparing the peak Ca²⁺ change between control (gray) and P2Y2-deficient (red) cells. (832 vs. 953 cells pooled from 12 calcium waves, Mann-Whitney test was used to determine statistical significance). (F) Bar graph comparing the average of median peak [Ca²⁺]_i response between control and P2Y2-deficient cells within the same range relative to the needle tip (mean±SEM, n = 12 calcium waves, Student's t test, * p<0.05; ** p<0.01; *** p<0.001; ns = not significant).

Figure 5. Knocking down NTPDase2 locally potentiates ATP-induced [Ca²⁺]_i increase.

(A) Fluorescence images showing how software-assisted cell demarcation and population identification is performed according to DNA and cell tracer dye staining. Cyan lines define control cells, and red lines outline P2Y2-deficient cells. (B) Time series of fluorescence imaging of calcium wave spreading in a co-culture monolayer containing an equal ratio of control and NTPDase2-deficient HaCaT cells. White dotted lines indicate the position of the glass needle. Insets: enlarged pseudo-color images of control and NTPDase2-deficient cell pairs depicted in ΔF/F₀. Dotted lines define the cell outline (a.u., arbitrary unit). (C) Traces showing Ca²⁺ change (fold change from baseline, ΔF/F₀) in individual control (gray) and NTPDase2-deficient (red) cells from panel B. The black arrow indicates the initiation of ATP point release, and the black line defines the duration of ATP point release (a.u., arbitrary unit). (D) Frequency distribution and log Gaussian fit comparing the peak Ca²⁺ change between control (gray) and NTPDase2-deficient (red) cells. (414 vs. 422 cells pooled from 5 calcium waves, Mann-Whitney test was used to determine statistical significance). (E) Bar graph comparing the average median peak [Ca²⁺]_i response between control and NTPDase2-deficient cells within the same range relative to the needle tip (mean±SEM, n = 5 calcium waves, Student's t test, * p<0.05; ** p<0.01; ns = not significant).