Introduction

The enzyme calcineurin (Cn) plays a key role in the immune response. In T cells, it is responsible for the dephosphorylation of the NFAT (nuclear factor of activated T cells), and its subsequent nuclear translocation ultimately results in T-cell activation (Winslow et al., 2003). In organ transplantation patients, Cn is the main target for immunosuppression by the Cn inhibitors cyclosporine A (CsA) and tacrolimus (Trl).

Besides its role in T cells, Cn plays a role in various tissues and also in the skin (Reynolds and Al Daraji, 2002; Aramburu et al., 2004). Al Daraji et al. (2002) have shown that in keratinocytes, the nuclear translocation of NFAT is regulated by Cn and influenced by Cn inhibitors. Furthermore, a role for Cn in the differentiation of keratinocytes has been described (Santini et al., 2001). Cn activity has been demonstrated before in cultured lung fibroblasts with an assay using ³²P-radiolabeled substrate and the activity could be inhibited by CsA (Chen et al., 2003). A role for Cn in melanocytes is suggested, as CsA was shown to decrease pigmentation in cultured human melanocytes, and Cn inhibitors are used topically for treatment of vitiligo (Lee and Kang, 2003; Taieb, 2005). The Cn inhibitors CsA, Trl, and pimecrolimus (Prl) are also used for the treatment of other dermatological diseases, such as psoriasis and atopic eczema (Griffiths, 2001; Reynolds and Al Daraji, 2002).

For a long time, the treatment of organ transplant patients with Cn inhibitors has been known to result in an increased incidence of non-melanoma skin cancer (Hartevelt et al., 1990), and nowadays the occurrence of various other types of post-transplant malignancies are becoming an important cause of mortality (Buell et al., 2005).

The studies mentioned above indicate a role for Cn in skin. However, so far no direct measurement of Cn activity in skin tissue or skin cells has been described. Recently, we developed a new sensitive assay for spectrophotometric measurement of Cn in leukocytes (Sellar et al., 2006). The increased use and topical application of Cn inhibitors in dermatological practice and the strong elevation of non-melanoma skin cancer in transplant patients treated with Cn inhibitors was the rationale to use this assay for our study on Cn activity in the skin and the cells, which are the main constituents of the dermis and epidermis.

Results

In this study, the Cn assay was used to study Cn activity in skin tissue and cultured cells derived from the skin. Our first measurements using a total skin tissue homogenate prepared in lysis buffer showed that Cn could be measured in the sample with the activity of 1.90.1 nmol per min per mg protein (n=4; 95% confidence interval (CI) 1.82–1.98 nmol min^{- 1} mg^{- 1}). After 14 days of culture (in Fei medium), the keratinocytes derived from the epidermis of the same skin showed a Cn activity of 11.10.9 nmol min^{- 1}mg^{- 1}. After the separation of the dermis and the epidermis of the skin from four other donors, an 8.0 higher Cn activity was measured in the epidermis (1.70.4 and 14.03.2 nmol min^{- 1} mg^{- 1} in dermis and epidermis, respectively) (Figure 1).

Figure 1.

Measurement of calcineurin activity in dermis and epidermis. Dermis and epidermis were separated by dispase treatment, homogenized, and lysed, and calcineurin activity was measured. Material was used from four different donors.

Full figure and legend (6K)

In cultures of fibroblasts derived from the dermis and keratinocytes from the epidermis of the same donor, a 40% lower Cn activity was found in the keratinocytes. Inhibitions of Cn activity with 124 nM Trl were 74 and 76% in the fibroblasts and keratinocytes, respectively (see Table 1). For another donor, a 55% lower activity was found for the melanocytes compared to the corresponding keratinocytes. Inhibition with 124 nM Trl was 89 and 75% in the keratinocyte and melanocyte cultures, respectively (Table 1). With 831 nM CsA, this was 25 and 49% , respectively. Mean Cn activity was higher in fibroblasts (13.67.0 nmol per min per mg protein) than it was in keratinocytes (7.23.0 nmol per min per mg protein) (P=0.018) (Figure 2). A considerable spread was observed for the Cn activities in the fibroblast and keratinocyte cultures (95% CI 9.5–17.7 and 4.9–9.5 nmol per min per mg protein, respectively). Activities in melanocytes were the lowest (2.70.9 nmol min^{- 1} mg^{- 1}). Low Cn activities were also found for two melanoma cell cultures (1.71.0 and 1.40.5 nmol min^{- 1} mg^{- 1} for An (n=8) and M14 (n=5), respectively).

Figure 2.

Calcineurin activity measurements in different fibroblast, keratinocyte, and melanocyte cultures. Cn activity measurements were performed in lysates of fibroblasts (Fb, n=14), keratinocytes (Kc, n=9), and melanocytes (Mc, n=3). Significant differences in Cn activity were observed between fibroblasts and keratinocytes (^*P=0.018) and between keratinocytes and melanocytes (^**P=0.030). All keratinocyte cultures were maintained in Fei medium.

Full figure and legend (8K)

Table 1 - Activity and inhibition of Cn in cell cultures from two different donors.

Full table

Figure 3 shows the inhibition curves for the inhibition of Cn after a 24-hour incubation with CsA in two different fibroblast cultures (F0502, F0503). Maximal inhibition of 63% was found in the culture F0503, and for both cultures similar IC₅₀ (the half maximal inhibitory concentration) values were found (Table 2). Inhibition curves for Trl and Prl were measured using the same two cultures. As can be seen in Table 2, maximum inhibition of Cn was stronger for Trl than for CsA and Prl, and IC₅₀ values calculated for Trl are lower than those for CsA and Prl. All three Cn inhibitors were more effective in the culture F0503 than in F0502.

Figure 3.

Effect of cyclosporine A on Cn activity in two different fibroblast cultures. Inhibition curves of Cn activity were measured for CsA in culture medium for two different fibroblast cultures (F0502, F0503). The cultures were incubated for 24 hours with CsA at 0–8,313 nM in duplicate. CsA was used from Sigma Chem Co.

Full figure and legend (11K)

Table 2 - Inhibition of activity with CsA, Trl, and Prl.

Full table

As can be concluded from these experiments and our earlier work with CsA (Sellar et al., 2006), maximal inhibitions are achieved at concentrations of 831 nM CsA, 124 nM Trl, and 123 nM Prl. Numerous inhibition experiments were subsequently performed using the inhibitors at these higher concentrations. In Figure 4, it can be seen that maximal reduction of Cn activity is found for Trl in both fibroblast and keratinocyte cultures. Somewhat milder inhibition is obtained with both CsA and Prl (CsA, 36 and 33% ; Trl, 57 and 55% ; and Prl, 43 and 35% inhibition in fibroblast and keratinocytes, respectively). As described above, inhibition of Cn activity in melanocytes with 831 nM CsA and 124 nM Trl was only measured once. Both melanoma cell cultures showed relatively low Cn activity compared to that of the melanocytes. After 24 hours treatment of the melanoma cells, reductions in Cn activity were similar for CsA and Trl (both n=6) (71 and 69% , respectively).

Figure 4.

Inhibition of Cn activity by cyclosporine A, tacrolimus, and pimecrolimus. Fibroblast (Fb,) and keratinocyte (Kc, ) cultures were incubated for 24 hours with CsA (831 or 4,156 nM), Trl (124 or 622 nM), and Prl (123, 617, 1,235, or 6,173 nM). Inhibition of Cn is shown as the mean percentage inhibition. Error bars represent SEM for fibroblasts treated with CsA (n=5), Trl (n=11), Prl (n=11), and for keratinocytes (n=5, 9, and 4, respectively). A total of seven different fibroblast and seven different keratinocyte cultures were used for the experiments. Inhibition by Trl was stronger than by CsA and Prl (^*P=0.0028, ^**P=0.0114). CsA was used from Novartis. Keratinocytes were maintained in either Fei medium or defined keratinocyte medium (Spiekstra et al., 2005) (see Materials and Methods).

Full figure and legend (38K)

Discussion

Using a Cn enzyme assay, we were able to investigate Cn activity in skin. In four different skin samples, the highest activity was found in the epidermal part containing keratinocytes (Figure 1). Total skin homogenates containing both dermis and epidermis showed a Cn activity similar to that found in dermis alone, which may be explained by the much larger contribution of the dermal part of the skin. In a large number of cell cultures of fibroblasts, keratinocyte and melanocyte activities of Cn were measured. On average, the highest activity was found in the fibroblast cultures derived from dermis, and keratinocytes from the epidermis showed about half the activity of that in fibroblasts (Figure 2). The activities found in epidermis and fibroblasts were in the same order as those found for peripheral blood mononuclear cells isolated from blood. Mean activities in melanocytes were the lowest (<20% of fibroblasts) and were similar to that in the two melanoma cell cultures. The melanoma cells were cultured in the same medium as the fibroblasts, suggesting that the low Cn activities are cell-specific (for the melanocytic cells) and not dependent on the culture conditions. The order of activities of Cn in fibroblasts, keratinocytes, and melanocytes was also observed in the experiments where cells were grown from the same donors (Table 1). In monoculture conditions, the dermal fibroblasts showed the highest Cn activities, whereas this was not the case for dermis compared to epidermis. This may be explained by the high content of connective tissue (containing macromolecules such as collagen, elastin, and fibronectin) and relatively low amounts of fibroblasts in dermis compared to the epidermis that contains high numbers of closely connected layers of keratinocytes. Therefore, in fibroblast monocultures, the specific Cn activity (per mg protein) may be expected to be higher than in the in vivo total dermis. The meaning of this relatively high Cn activity in fibroblasts remains to be further investigated. Cn activity has been demonstrated before in fibroblasts derived from lung tissue and was suggested to play a role in the synthesis of collagen (Chen et al., 2003). In keratinocytes, Cn activity has not been measured before, although it has been recognized that the Cn/NFAT pathway plays an important role in keratinocyte differentiation and growth regulation (Santini et al., 2001; Mammucari et al., 2005; Sakaguchi et al., 2005). In this respect, variations in Cn activity and the inhibitory capacity of the Cn inhibitors in different individuals or patients may be of great value to predict the efficacy of treatment. For Cn, a considerable spread in activities was found in both keratinocyte and fibroblast cultures. Also, for the inhibition curves of Cn activities, some difference in IC₅₀ values and maximal inhibitions were found for two fibroblast cultures (Table 2). Strongest inhibition was found for Trl in these cultures (with lowest IC₅₀ values and lowest remaining activities). In another larger set of experiments, variation in Cn inhibition was also observed in a group of different fibroblast and keratinocyte cultures (Figure 4). 95% CIs for the inhibition of Cn activity in both fibroblasts and keratinocytes combined were 28.5–41.1% for CsA, 47.2–65.7% for Trl, and 33.6–47.8% for Prl. The inhibition of Cn in these experiments was relatively low, especially for CsA as compared to the maximal Cn inhibitions calculated for CsA from the inhibition curves (Table 2). This might be a consequence of the use of two different preparations of CsA (Sigma Chem Co., Bornem, Belgium and Novartis, Basel, Switzerland, respectively) used in the experiments, although no difference in CsA concentrations for either source was found. In contrast, the inhibition with CsA in Table 2 could be dependent on the fibroblast cultures used and their status of differentiation (during culture). In this respect, we did not observe clear differences in Cn activities between differentiated keratinocytes cultured with different media (as in Materials and Methods) or with a keratinocyte-defined, serum-free medium (KD-SFM; Invitrogen Ltd, Paisley, UK), resulting in much less differentiated cultures (data not shown). However, further studies will be required to clarify regulation of Cn activity during the various growth phases in culture.

Inhibition by the Cn inhibitors as measured in the different cell types seems to be relevant for clinical practice. For example, the concentrations of CsA in epidermis after oral administration of the drug has been reported to vary between 0.8 and 2.5 M (Fisher et al., 1988; Taieb, 2005). This is about 1,000 higher than the IC₅₀ values for CsA that we measured (Table 2). The IC₅₀ values obtained for CsA, Trl, and Prl in fibroblasts are relatively low compared to our earlier data obtained after incubations with fresh blood (Sellar et al., 2006). Differences in efficiency of Cn inhibition have been described before for incubation in whole blood and in culture media (Batiuk et al., 1996). Also incubation times with the inhibitors for 1 hour, as in our previous experiments (Sellar et al., 2006), and 24 hours in this study will be of influence. Nevertheless, differences in efficiency of the inhibitors can be measured confirming strong potency of Trl as a Cn inhibitor. Using more functional assays, Trl has also been shown to be more potent than Prl (Grassberger et al., 1999). The strong Cn inhibition by Trl is in agreement with the lower doses that are used for treatment of transplant patients in comparison to CsA and the lower concentrations used topically (0.1 and 0.03% ) in creams, compared to the 1% used for Prl.

Our study shows that keratinocytes and fibroblasts may be direct targets for Cn inhibition, and functional consequences need to be further clarified. Interindividual variation in Cn activities and monitoring responses to Cn inhibition may be helpful to improve treatment of skin diseases such as psoriasis, atopic eczema, and vitiligo.

Materials and Methods

Skin preparation and cell cultures

Foreskin was obtained from surgery after circumcision and maintained in PBS. Subcutaneous fat was removed with scissors. Total skin tissue was homogenized in a dismembrator after freezing in liquid nitrogen. The same procedure was used for dermis and epidermis after separation by overnight treatment with dispase (grade II; Roche Diagnostics, Mannheim, Germany) at 4 °C. Part of the samples was used for culture of dermal fibroblasts and epidermal keratinocytes and melanocytes.

Melanocytes were routinely cultured in Ham's F10 medium (Invitrogen Ltd) with 2% fetal calf serum (Hyclone, Logan, UT), 1% Ultroser-G (Pall Biosepra SA, Cergy, France), 2 ng ml^{- 1} basic fibroblast growth factor (PeproTech, Rocky Hill, NJ), 2 ng ml^{- 1} endothelin-1 (Sigma Chem Co.), 16 nM TPA (12-O-tetradecanoylphorbol-13-acetate; Sigma), and 0.1 mM IBMX (3-isobutyl-1-methylxanthine) (Sigma Chem Co.) (Fei medium). Presence of TPA causes preferential attachment of melanocytes. Melanocytes were selectively detached with trypsin/EDTA (1:8; Invitrogen Ltd) and are thus separated from any remaining keratinocytes after the first passage. Contaminating keratinocytes or fibroblasts at later passage of melanocytes were removed by geneticin treatment (Halaban and Alfano, 1984; Smit et al., 1995). Epidermal keratinocytes were also grown in the melanocyte medium without TPA (Fei medium) during the first three passages only. In other experiments (as indicated), a defined keratinocyte medium (DMEM/Ham's F12 medium (3:1), containing Ultroser-G, insulin, isoproterenol, hydrocortisone, and penicillin/streptomycin) was used according to concentrations and suppliers mentioned, as described earlier (Spiekstra et al., 2005).

Fibroblasts and melanoma cells (UCLA-SO-M14 and MM-AN) were cultured in DMEM (Invitrogen Ltd) with 2.5% fetal calf serum (Hyclone) and penicillin/streptomycin (Invitrogen Ltd).

All surgery materials were obtained with informed consent. The Declaration of Helsinki Principles was followed and experimental procedures were approved by the Leiden University Medical Center and the VU University Medical Center.

Calcineurin activity measurements

Homogenized tissues (as above) and harvested cell pellets were lysed in the standard lysis buffer from the Biomol Green Cellular Calcineurin Assay Kit supplemented with 50 mg l^{- 1} phenylmethanesulfonylfluoride, 50 mg l^{- 1} soybean trypsin inhibitor, 5 mg l^{- 1} leupeptin, 5 mg l^{- 1} aprotinin, and 5.0 mM ascorbic acid. Lysis was performed by three freeze-thaw cycles (liquid N₂/30 °C) followed by removal of the cell debris by centrifugation (10,000 g, 10 minutes 4 °C) (Sellar et al., 2006). Briefly, the Cn protein phosphatase 2B activity was measured by inhibiting other protein phosphatases with excess okadaic acid. Cn activity was defined as the calcium-dependent protein phosphatase 2B by measuring phosphate production in the assay in the presence and absence of EGTA (ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid) (Sellar et al., 2006). For maximal performance of this Cn assay, the protein concentration of the samples was standardized to 0.1–0.2 mg ml^{- 1}.

Protein concentration was determined using the Pierce Coomassie Plus Total Protein Assay (PerBio Science, Aalst, Belgium) based on the method of Bradford (Bradford, 1976).

Inhibition measurements were performed with CsA (was used either from Novartis (SandImmune) or from Sigma Chem Co. as indicated for different experiments), Trl (Sigma Chem Co.), and Prl (Novartis). Concentration ranges used were 0–8,313 nM CsA, 0–622 nM Trl, and 0–6,173 nM Prl. These concentrations were partly based on earlier experience, and concentrations of CsA and Trl were controlled by routine immunoassay methods (Sellar et al., 2006).

Statistics

All data are presented as meanSD unless otherwise indicated. Inhibition curves and IC₅₀ values were fitted and analyzed by sigmoidal dose–response curves using Graphpad Prism software (San Diego, CA). For the purpose of curve fitting, control (0) values were set to - 2 (0.01 nM) (no difference was found for settings of - 3 and - 4). Student's t-test was used for comparison of data to test significance of difference, which was defined by P-value <0.05.

Conflict of Interest

The authors state no conflict of interest.

References

1. Al Daraji WI, Grant KR, Ryan K, Saxton A, Reynolds NJ (2002) Localization of calcineurin/NFAT in human skin and psoriasis and inhibition of calcineurin/NFAT activation in human keratinocytes by cyclosporin A. J Invest Dermatol 118:779–788 | Article | PubMed | ISI | ChemPort |
2. Aramburu J, Heitman J, Crabtree GR (2004) Calcineurin: a central controller of signalling in eukaryotes. EMBO Rep 5:343–348 | Article | PubMed | ISI | ChemPort |
3. Batiuk TD, Pazderka F, Enns J, De Castro L, Halloran PF (1996) Cyclosporine inhibition of leukocyte calcineurin is much less in whole blood than in culture medium. Transplantation 61:158–161 | Article | PubMed | ChemPort |
4. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 | Article | PubMed | ISI | ChemPort |
5. Buell JF, Gross TG, Woodle ES (2005) Malignancy after transplantation. Transplantation 80:S254–S264 | Article | PubMed | ISI |
6. Chen Y, Zhao M, Fu M, Yao W, Tang C (2003) The role of calcineurin in the lung fibroblasts proliferation and collagen synthesis induced by basic fibroblast growth factor. Chin Med J (Engl) 116:857–862 | PubMed | ChemPort |
7. Fisher GJ, Duell EA, Nickoloff BJ, Annesley TM, Kowalke JK, Ellis CN et al. (1988) Levels of cyclosporin in epidermis of treated psoriasis patients differentially inhibit growth of keratinocytes cultured in serum free versus serum containing media. J Invest Dermatol 91:142–146 | Article | PubMed | ISI | ChemPort |
8. Grassberger M, Baumruker T, Enz A, Hiestand P, Hultsch T, Kalthoff F et al. (1999) A novel anti-inflammatory drug SDZ ASM 981, for the treatment of skin diseases: In Vitro pharmacology. Br J Dermatol 141:264–273 | Article | PubMed | ISI | ChemPort |
9. Griffiths CE (2001) Ascomycin: an advance in the management of atopic dermatitis. Br J Dermatol 144:679–681 | Article | PubMed | ChemPort |
10. Halaban R, Alfano FD (1984) Selective elimination of fibroblasts from cultures of normal human melanocytes. In Vitro 20:447–450 | Article | PubMed | ISI | ChemPort |
11. Hartevelt MM, Bavinck JN, Kootte AM, Vermeer BJ, Vandenbroucke JP (1990) Incidence of skin cancer after renal transplantation in The Netherlands. Transplantation 49:506–509 | Article | PubMed | ISI | ChemPort |
12. Lee JY, Kang WH (2003) Effect of cyclosporin A on melanogenesis in cultured human melanocytes. Pigment Cell Res 16:504–508 | Article | PubMed | ChemPort |
13. Mammucari C, Tommasi di Vignano A, Sharov AA, Neilson J, Havrda MC, Roop DR et al. (2005) Integration of Notch 1 and calcineurin/NFAT signaling pathways in keratinocyte growth and differentiation control. Dev Cell 8:665–676 | Article | PubMed | ISI | ChemPort |
14. Reynolds NJ, Al Daraji WI (2002) Calcineurin inhibitors and sirolimus: mechanisms of action and applications in dermatology. Clin Exp Dermatol 27:555–561 | Article | PubMed | ISI | ChemPort |
15. Sakaguchi M, Sonegawa H, Nukui T, Sakaguchi Y, Miyazaki M, Namba M et al. (2005) Bifurcated converging pathways for high Ca2+- and TGFbeta-induced inhibition of growth of normal human keratinocytes. Proc Natl Acad Sci USA 102:13921–13926 | Article | PubMed | ChemPort |
16. Santini MP, Talora C, Seki T, Bolgan L, Dotto GP (2001) Cross talk among calcineurin, Sp1/Sp3, and NFAT in control of p21(WAF1/CIP1) expression in keratinocyte differentiation. Proc Natl Acad Sci USA 98:9575–9580 | Article | PubMed | ChemPort |
17. Sellar KJ, van Rossum HH, Romijn FP, Smit NP, de Fijter JW, Van Pelt J (2006) Spectrophotometric assay for calcineurin activity in leukocytes isolated from human blood. Anal Biochem 358:104–110 | Article | PubMed | ChemPort |
18. Smit NP, Westerhof W, Menko WJ, Verbeek NM, Pavel S (1995) Stimulation of cultured melanocytes in medium containing a serum substitute: Ultroser-G. Pigment Cell Res 8:19–27 | Article | PubMed | ChemPort |
19. Spiekstra SW, Toebak MJ, Sampat-Sardjoepersad S, van Beek PJ, Boorsma DM, Stoof TJ et al. (2005) Induction of cytokine (interleukin-1alpha and tumor necrosis factor-alpha) and chemokine (CCL20, CCL27, and CXCL8) alarm signals after allergen and irritant exposure. Exp Dermatol 14:109–116 | Article | PubMed | ISI | ChemPort |
20. Taieb A (2005) Tacrolimus and the pigmentary system. Dermatology 210:177–178 | Article | PubMed | ChemPort |
21. Winslow MM, Neilson JR, Crabtree GR (2003) Calcium signalling in lymphocytes. Curr Opin Immunol 15:299–307 | Article | PubMed | ISI | ChemPort |

Acknowledgments

We thank W. Temmink and S. van Iterson for technical support with cell cultures. We gratefully acknowledge Dr R. Schneider from Novartis Pharma AG, Basel, Switzerland, for providing us with pimecrolimus for the experiments.