Matrix Biology 22 (2003) 491–500
Multidrug resistance protein-6 (MRP6) in human dermal fibroblasts. Comparison between cells from normal subjects and from Pseudoxanthoma elasticum patients F. Boraldi, D. Quaglino, M.A. Croce, M.I. Garcia Fernandez, R. Tiozzo, D. Gheduzzi, B. Bacchelli, I. Pasquali Ronchetti* Department of Biomedical Sciences, University of Modena and Reggio Emilia, via Campi, 287, 41100 Modena, Italy Received 28 January 2003; received in revised form 8 September 2003; accepted 9 September 2003
Abstract Multidrug resistance protein-6 (MRP6) is a membrane transporter whose deficiency leads to the connective tissue disorder Pseudoxanthoma elasticum (PXE). In vitro dermal fibroblasts from normal and PXE subjects, homozygous for the R1141X mutation, were compared for their ability to accumulate and to release fluorescent calcein, in the absence and in the presence of inhibitors and competitors of the MDR–multidrug resistance protein (MRP) systems, such as 3-(3-(2-(7-choro-2 quinolinyl) ethenyl)phenyl ((3-dimethyl amino-3-oxo-propyl)thio) methyl) propanoic acid (MK571), verapamil (VPL), vinblastine (VBL), chlorambucil (CHB), benzbromarone (BNZ) and indomethacin (IDM). In the absence of chemicals, calcein accumulation was significantly higher and the release significantly slower in PXE cells compared to controls. VBL and CHB reduced calcein release in both cell strains, without affecting the differences between PXE and control fibroblasts. VPL, BNZ and IDM consistently delayed calcein release from both control and PXE cells; moreover, they abolished the differences between normal and MRP6deficient fibroblasts observed in the absence of chemicals. These findings suggest that VPL, BNZ and IDM interfere with MRP6dependent calcein extrusion in in vitro human normal fibroblasts. Interestingly, MK571 almost completely abolished calcein release from PXE cells, whereas it induced a strong but less complete inhibition in control fibroblasts, suggesting that MRP6 is not inhibited by MK571. Data show that MRP6 is active in human fibroblasts, and that its sensitivity to inhibitors and competitors of MDR–MRPs’ membrane transporters is different from that of other translocators, namely, MRP1. It could be suggested that MRP1 and MRP6 transport different physiological substances and that MRP6 deficiency cannot be overcome by other membrane transporters, at least in fibroblasts. These data further support the hypothesis that MRP6 deficiency may be relevant for fibroblast metabolism and responsible for the metabolic alterations of these cells at the basis of connective tissue clinical manifestations of PXE. 䊚 2003 Elsevier B.V.yInternational Society of Matrix Biology. All rights reserved.
Keywords: ABCC6; Fibroblast; Membrane transporter; MRP6; Pseudoxanthoma elasticum
Abbreviations: ABCC, ATP-binding cassette C; BNZ, benzbromarone; Calcein-AM, calcein acetoxy-methyl-ester; CHB, chlorambucil; DMEM, Dulbecco modified eagle medium; DMSO, dimethyl-sulphoxide; FBS, fetal bovine serum; IDM, indomethacin; MK571, 3-(3-(2-(7choro-2 quinolinyl) ethenyl)phenyl ((3-dimethyl amino-3-oxo-propyl)thio) methyl) propanoic acid; MRP, multidrug resistance protein; NEMGS, N-ethyl-maleimide-S-glutathione; PBS, phosphate buffered saline; PXE, Pseudoxanthoma elasticum (on-line Mendelian inheritance in man 264800, MIM 177850); VBL, vinblastine; VPL, verapamil *Corresponding author. Tel.: q39-59-2055418; fax: q39-59-2055426. E-mail address:
[email protected] (I. Pasquali Ronchetti). 0945-053X/03/$30.00 䊚 2003 Elsevier B.V.yInternational Society of Matrix Biology. All rights reserved. doi:10.1016/j.matbio.2003.09.001
492
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
1. Introduction Membrane transporters of the multidrug resistance protein (MRP) family have been found in several types of human cells (Borst et al., 1999). They are involved in cell efflux of a series of cellular products as well as of chemicals that passively enter the cells and are extruded with consumption of ATP (Borst et al., 1999). These transporters have been studied in cancer cells where they are, at least partially, responsible for tumor cell chemoresistance (Hipfner et al., 1999). This study has been undertaken with the aim of investigating whether multidrug resistance protein-6 (MRP6) is active in human dermal fibroblasts. Since MRP6 deficiency has been found to be responsible for the genetic disorder Pseudoxanthoma elasticum (PXE), the efficiency of MRP6 in normal cells has been compared with that in cells from PXE patients. PXE is a genetic disorder characterized by systemic alterations of connective tissues, involving mainly the skin and the cardiovascular system (Neldner and Struck, 2002). The most relevant clinical manifestations are recurrent hemorrhages of retinal and gastrointestinal vessels due to progressive mineralization and fragmentation of elastic fibers. However, alterations of collagen fibrils and of several matrix constituents indicate that the gene defect in PXE leads to rather complex and widespread connective tissue alterations (BaccaraniContri et al., 1996; Pasquali-Ronchetti et al., 1981, 1986; Neldner and Struck, 2002; Gheduzzi et al., in press). The PXE gene (ABCC6) encodes for the protein MRP6 which shows the transmembrane domains, the Walker motifs and the ATP-binding elements typical of the ABC membrane transporters (Bergen et al., 2000; Le Saux et al., 2000; Ringpfeil et al., 2000). This is a large family of proteins that have been extensively studied as implicated in the extrusion from cells of endogenous metabolites and of a series of exogenous drugs such as those used in tumor chemotherapy (Borst et al., 1999; Klein et al., 1999). In particular, the MRP6 protein, whose physiological function is still unknown, has high homology with the MRP1 transporter, whose gene (ABCC1) is located next to ABCC6, and which is deeply implicated in cancer multidrug resistance (Hipfner et al., 1999). Over-expression of ABCC6 has been observed in tumor cells only in association with chemically stimulated over-expression of ABCC1, suggesting that ABCC6 should not have an important role in drug resistance of tumor cells (Kool et al., 1999). Moreover, ABCC6 expression cannot be modified by chemicals commonly used to over-express multidrug resistanceassociated proteins in liver (Madon et al., 2000; Ogawa et al., 2000), further indicating that ABCC6 may play a peculiar role. Even though ABCC6 mRNA has been showed to be mostly expressed in liver and in kidney suggesting that this transporter may be involved in the
extrusion of endogenous metabolites by hepatic and renal cells (Kool et al., 1999), interestingly, clinical manifestations and laboratory parameters do not indicate any liver or kidney altered function that might be responsible for PXE (Neldner and Struck, 2002). By contrast, PXE clinical and ultrastructural alterations affect soft connective tissues of the whole body (Gheduzzi et al., in press) and seem to indicate that complex metabolic alterations of mesenchymal cells, and of fibroblasts in particular, may be responsible for the clinical manifestations in this genetic disorder, as supported by reports describing that PXE dermal fibroblasts exhibit, either in vivo or in vitro, structure, behavior and synthetic capabilities for matrix constituents deeply modified compared to controls (Baccarani-Contri et al., 1996; Bacchelli et al., 1999; Pasquali-Ronchetti et al., 1981, 1986; Passi et al., 1996; Quaglino et al., 2000; Tiozzo-Costa et al., 1988). The aims of the present study were to investigate: (1) if MRP6 is active on in vitro cultured dermal fibroblasts; (2) the sensitivity of MRP6 to chemicals that are already known to interfere with the activity of some MDRy MRPs’ membrane transporters. We compared the efficiency of the MDRyMRP system of normal fibroblasts with that of fibroblasts from PXE patients homozygous for the nucleotide change 3421 C™T in exon 24 of the ABCC6 gene resulting in the creation of a premature stop codon at R1141X (Le Saux et al., 2000, 2001). The efficiency of membrane transporters was assessed by evaluating the ability of these cells in vitro to accumulate and to extrude calcein, that is already known to be transported by the multidrug resistance system and in particular by MRP1 and Pgp. Data show that the membrane transporter MRP6 is active in normal human dermal fibroblasts as normal cells extrude calcein much faster than MRP6-deficient PXE cells. Moreover, at least in fibroblasts, the sensitivity of MRP6 to some inhibitors andyor competitors of transmembrane transporters is different from that of MRP1 and Pgp. Therefore, MRP6 deficiency in fibroblasts and, probably, in similar mesenchymal cells could be likely responsible for connective tissue alterations typical of PXE. 2. Results 2.1. Calcein accumulation and release The accumulation of fluorescent calcein by both control and PXE fibroblasts was assayed at different calcein acetoxy-methyl-ester (calcein-AM) concentrations with very similar results. In all cases, the rate of calcein accumulation was significantly higher in PXE than in control fibroblasts (Fig. 1A). By using 0.1 mM calcein-AM, the rate of accumulation was 0.037 miny1 in PXE (rs0.76) and 0.015 miny1 in controls (rs
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
Fig. 1. Accumulation of fluorescent calcein in control and PXE fibroblasts evaluated by flow cytometry (A, B) and confocal microscopy (C–F). In both cell types, the plateau was reached after approximately 20-min incubation in the presence of 0.1 mM calcein-AM (A), but the rate of fluorescent calcein accumulation (B) was higher in PXE cells than in controls (*P-0.05). Calcein accumulation was barely detectable by confocal microscopy in control (C) and PXE (E) fibroblasts after 2-min incubation with 0.1 mM calcein-AM (C, E). Calcein appeared to accumulate into globules around the nucleus after 30-min incubation (D, F). PXE cells (F) were more fluorescent compared to normal fibroblasts (D). Data are expressed as rate of fluorescent calcein accumulation with time (A), and constant rate of accumulation (B). Confocal microscopy images were obtained in the middle of cells by optical sectioning (C–F).
0.79) (Fig. 1B) (P-0.05). The higher rate of fluorescent calcein accumulation in PXE fibroblasts compared to controls was abolished by pre-treatment of cells with digitonin (data not shown). Visualization of calcein accumulation with time is shown in Fig. 1C–F. After 2 min-incubation in the presence of 0.1 mM calcein-AM (Fig. 1C and E), fluorescent calcein started to accumulate into discrete globules around the nucleus in both control (C) and PXE (E) fibroblasts. After 30 minincubation (Fig. 1D and F), both control (D) and PXE (F) cells accumulated large amount of fluorescent calcein. By flow cytometry, after 30-min incubation with 0.1 mM calcein-AM, the amount of fluorescent calcein inside PXE fibroblasts was approximately 30% higher than that in controls, being the mean fluorescence
493
73.9"4.2 and 107.55"12.0 in 1=106 control and PXE cells, respectively (P-0.05). The higher calcein accumulation by PXE fibroblasts was not due to an higher number of lysosomes in these cells compared to controls. As shown in Fig. 2, control (A) and PXE (B) fibroblasts exhibited a similar content of lysosomes. The structural evidence was confirmed by spectrophotometry, since the amount of neutral red accumulated by lysosomes in control and PXE fibroblasts was not significantly different when evaluated per number of cells (Fig. 2, lower panels). Calcein release is shown in Fig. 3. Both control and PXE fibroblasts were incubated for 30 min in the presence of 0.1 mM calcein-AM, carefully washed and the release measured in the absence of calcein. The rate of calcein release with time was significantly lower in PXE compared to control fibroblasts (Fig. 3B) (P0.05). The phenomenon was evident at all concentrations of calcein-AM used (data not shown). Confocal microscopy confirmed that the release of fluorescent calcein was progressive with time from both normal (Fig. 3C and D) and PXE cells (Fig. 3E and F), being the fluorescent calcein retention always higher in PXE than in control fibroblasts after 1 h (Fig. 3C and E) and 4 h (Fig. 3D and F) release. Moreover, confocal images showed that calcein was retained into discrete cytoplasmic globules. The rate of calcein release was independent from the growing conditions of cells. Fig. 4 shows that calcein release was almost identical in fibroblasts
Fig. 2. Lysosomes were visualized by neutral red. Both control (A) and PXE (B) fibroblasts exhibited similar amount and distribution of lysosomes. When referred to the cell number, the amount of neutral red accumulated into control and PXE fibroblasts was not statistically different (lower panels: optical density of neutral red accumulated into 1=106 cells).
494
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
(CHB) (Fig. 5C) induced a 30–40% delay of the rate of calcein release from both control and PXE cells, without affecting the higher retention of calcein by PXE fibroblasts compared to controls. By contrast, verapamil (VPL) (Fig. 5D) and benzbromarone (BNZ) (Fig. 5E) significantly delayed calcein release from both control and PXE cells and abolished the higher retention of calcein by MRP6-deficient cell lines observed in the absence of competitors (Fig. 5A). In the presence of indomethacin (IDM) (Fig. 5F) and of MK571 (Fig. 5G), calcein release was represented by a regression line. IDM (Fig. 5F) induced a significant delay of calcein release from both control and PXE cells and almost abolished the differences between normal and MRP6-deficient cell lines. In the presence of MK571 calcein release from control fibroblasts was dramatically reduced whereas it was almost completely abolished from PXE cells (Fig. 5G). The x-intercept of the curve of calcein release was calculated in 23 and 91 h for control and PXE fibroblasts, respectively. 3. Discussion
Fig. 3. Calcein efflux from control and PXE fibroblasts, preloaded with 0.1 mM calcein-AM for 30 min, was evaluated by flow cytometry (A, B) and by confocal microscopy (C–F). The rate of calcein efflux was more than 20% higher in control compared to PXE cells (P-0.05) (A–B). By confocal microscopy, calcein release was indirectly evaluated by the amount of fluorescent calcein retained within fibroblasts (C–F). PXE cells (E–F) were always more fluorescent than controls (C–D) after 60 min (C, E) and 240 min (D, F) of calcein release. Data are expressed as percentage decrease of fluorescence with time (A) and calcein efflux constant rate (hy1 ) (B). Confocal images were taken in the middle of cells by optical sectioning (C–F).
The present study investigated if MRP6 is active in in vitro cultured dermal fibroblasts, by using primary cell lines from normal subjects and from PXE patients, bearing a specific mutation in the ABCC6 gene which encodes for the membrane transporter MRP6 (Le Saux et al., 2000, 2001). Only fibroblasts from patients homozygous for the R1141X mutation were used in the present study. This mutation is a nucleotide change 3421 C™T, which is located in exon 24 of the ABCC6 gene and results in a premature stop codon in the ABCC6
in proliferation and at confluence, and that, in both growing conditions, PXE fibroblasts exhibited a significantly slower calcein release (P-0.01). 2.2. Effect of competitors and inhibitors Fig. 5 illustrates calcein efflux from proliferating normal and PXE fibroblasts which were incubated for 30 min in the presence of 0.1 mM calcein-AM and of different competitorsyinhibitors of the MDRyMRP transporters. After calcein loading, cells were carefully washed and the release was measured either in the absence (Fig. 5A) or in the presence of the different drugs (Fig. 5B–G). Vinblastine (VBL) (Fig. 5B) and chlorambucil
Fig. 4. The rate of calcein release from fibroblasts was similar in confluent and proliferating cells. In both conditions, calcein efflux was always higher in controls than in PXE cells (P-0.05). Cells were incubated with 0.1 mM calcein-AM for 30 min, carefully washed and the efflux monitored by flow cytometry. Values are given as constant rate expressed as hy1.
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
495
Fig. 5. Calcein release was evaluated by flow cytometry in the absence (A) and in the presence of 5 nM VBL (B), 100 mM CHB (C), 100 mM VPL (D), 100 mM BNZ (E), 100 mM IDM (F) and 50 mM MK571 (G) in control and PXE fibroblasts. The enclosed table reports the mean values "S.D., obtained in three experiments with different cell lines, and show the rate of calcein release evaluated as K (hy1 ) or as K (%F=hy1) in each experimental condition. VBL (B) and CHB (C) caused a delayed release of calcein, without affecting the differences between control and PXE cells in the absence of competitors. By contrast, VPL (D), BNZ (E) and IDM (F) strongly delayed calcein release and abolished differences between control and PXE cells. MK571 (G) induced a complete and a partial inhibition of calcein release from PXE and control cells, respectively (P-0.001). Data in panel A–G are expressed as percentage decrease of fluorescence with time. *P-0.05; **P)0.01 PXE vs. normal fibroblasts within the same experimental conditions.
mRNA. Clinical manifestations in these patients are severe, therefore the truncated protein is probably not functional or is scarcely produced andyor rapidly destroyed (Le Saux et al., 2000, 2001). We compared the efficiency of normal and PXE fibroblasts in extruding fluorescent calcein, a chemical that is widely used to monitor the activity of membrane transporters of the ATP-binding cassette family (Feller et al., 1995; Hollo, 1994). Calcein-AM is a highly lipophilic and non fluorescent
compound, which rapidly penetrates the plasma membrane and is converted to fluorescent calcein by intracellular esterases (Hollo, 1994). Calcein is extruded from cells through the MRP andyor MDR systems (Barnouin et al., 1998; Homolya et al., 1996). The efficiency of calcein release can be evaluated by confocal microscopy and by flow cytometry by measuring the time-dependent residual fluorescence of pre-loaded cells (Barnouin et al., 1998; Dhar et al., 1998; Feller et al., 1995; Hollo, 1994; Homolya et al., 1996).
496
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
PXE fibroblasts showed a significantly higher accumulation and slower release of fluorescent calcein compared to normal cells. The higher rate of calcein accumulation by PXE fibroblasts compared to controls might be explained with a more rapid entrance and esterase conversion of calcein-AM to fluorescent calcein, or to an immediate major retention of the fluorescent dye due to MRP6 deficiency in PXE fibroblasts. The lysosome content was identical in control and in PXE cells. Therefore, lysosome esterases should not be responsible for the higher content of fluorescent calcein in PXE compared to normal fibroblasts. Moreover, digitonin treatment abolished the differences in calcein accumulation and release between control and PXE cells, showing that the major calcein retention by PXE fibroblasts was due to the less efficient calcein efflux from these cells caused by MRP6 deficiency. Therefore, MRP6 deficiency, due to the R1141X mutation, induces fibroblasts to accumulate calcein in higher amount and to release it in a time 25% longer than normal cells. These data indicate that MRP6 is active in human skin fibroblasts and is involved in calcein release from these cells. Moreover, MRP6 is active in all phases of the fibroblast’s life since differences in calcein efflux between normal and PXE cells were independent from the cell duplication phase. Therefore, calcein accumulation and release may be used to monitor the efficiency of the MRP6 transporter in human skin fibroblasts in vitro. VBL and CHB are chemicals that are used for their toxic effects on tumor cells (Begleiter et al., 1996; Vertessy et al., 1997). They may be considered competitors of calcein for the same membrane transporters, as cell sensitivity to these toxic substances depends mostly on the cell efficiency to extrude them through the MDR and MRP systems (Flanagan et al., 2002; Hamilton et al., 2001). To put some light on the MRP6 efficiency, we investigated whether MRP6 is involved in the extrusion of these chemicals. VBL and CHB partially inhibited calcein efflux in both control and PXE cells, without modifying the differences between the two cell strains observed in the absence of these drugs. PXE fibroblasts always exhibited a higher accumulation and a slower release of calcein compared to controls, indicating that VBL and CHB do not interfere with MRP6 activity in normal dermal fibroblasts. MK571 is a quinoline-based chemical described to be rather specific for the MRP system (Pulaski et al., 1996) and, in particular, to inhibit fluorescent calcein efflux through the MRP1 transporter (Essodaigui et al., 1998). In our experimental conditions, the theoretical discharge time in the presence of MK571 was 91 and 23 h in PXE and normal fibroblasts, respectively. This means that, in the presence of MK571, calcein release was practically abolished in PXE cells, whereas it was
delayed but still active in normal fibroblasts. Therefore, data confirm that MK571 is an inhibitor of MRP1 (Essodaigui et al., 1998), as it blocked almost completely calcein efflux from MRP6-deficient fibroblasts. Moreover, its less dramatic effect on normal fibroblasts, expressing MRP6, would indicate that MK571 is not an inhibitor for MRP6. In our experimental conditions, VPL, IDM and BNZ induced a relevant inhibition of calcein efflux from control and MRP6-deficient cells. In addition, it is worth mentioning that the slope of calcein efflux, in the presence of 100 mM IDM, approached that of a straight line, indicating the high efficiency of this chemical in interfering with the activity of fibroblast’s membrane transporters. Moreover, these three drugs almost abolished the differences in calcein efflux between normal and MRP6-deficient cell lines, suggesting that they interfere with and block the MRP6 activity in normal fibroblasts. These data are in agreement with those recently obtained on reconstituted vesicles expressing human MRP6, where IDM and BNZ have been shown to rather specifically inhibit the release of N-ethylmaleimide-S-glutathione (NEM-GS) (Ilias et al., 2002). The effect of VPL is less clear, as it is a competitor of Pgp and MRP1, and it is known to stimulate the ATPase activity (Crivellato et al., 2002). Therefore, its inhibitory effect on calcein release also from normal fibroblasts could be due to a direct effect on MRP6 andyor to an indirect effect more related to ATP-depletion of cells. In conclusion, the different rates of calcein accumulation and release by PXE dermal fibroblasts compared to controls indicate that MRP6 is active in human skin fibroblasts and that MRP6 deficiency may cause an impaired transmembrane transport of physiologically important substrates andyor metabolites. Whatever is the precise cell membrane localization of MRP6 in fibroblasts, and whatever isyare its physiological substrateys, retention of thisythese substances within dermal fibroblasts is very likely responsible for the metabolic alterations at the basis of skin clinical manifestations in PXE (Maccari et al., 2003; Passi et al., 1996; Quaglino et al., 2000; Tiozzo-Costa et al., 1988). However, the structural alterations typical of PXE have been found in almost all connective tissues of the human body (Gheduzzi et al., in press). This may indicate that MRP6 is active and physiologically important for fibroblast metabolism in all connective tissues. The high expression of MRP6 in liver and kidney (Kool et al., 1999; Scheffer et al., 2002) seems to suggest that MRP6 has a functional role in these organs. However, since PXE patients do not manifest any liver or kidney disorder, other transporters might substitute for MRP6 in these organs in MRP6 deficiency. As far as fibroblasts are concerned, the different sensitivity of normal and PXE cells to inhibitorsycompetitors of the MDRyMRP systems seems to indicate that MRP6 behaves different
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
from, at least, MRP1 andyor Pgp. This may suggest that MRP6 has peculiar functions in fibroblasts, where its deficiency could not be compensated by other transporters. This would explain why MRP6 deficiency induces peculiar alterations of fibroblast metabolism, leading to the generalized connective tissue clinical manifestations observed in PXE. 4. Experimental procedures 4.1. Chemicals Calcein-AM was purchased from Molecular Probes, Eugene, OR. MK-571 was from Alexis Biochemicals, San Diego, CA; CHB was from ICN Biomedicals, Costa Mesa, CA; dimethyl-sulphoxide (DMSO), VPL, VBL, IDM, BNZ, digitonin, trypsin, Triton X100, paraformaldehyde, Tween 20, neutral red were from Sigma, St. Louis, MO. All chemicals were of analytical grade. 4.2. Subjects and cells After informed signed consent, skin biopsies were taken under local anesthesia from the neck or from the axilla of eight PXE patients (five women and three men; mean age 35"15 year) and from the axilla of six normal subjects (four women and two men; mean age 40"15 year) who underwent surgery and did not show any sign of connective tissue alterations. Control and PXE individuals were Italians. Patients were homozygous for the mutation R1141X (see further for details) showing no detectable mRNA for MRP6 (Le Saux et al., 2000). Patient’s clinical manifestations were severe. All of them had pronounced skin papules on the neck, axillae and groin, angioid streaks andyor retinal hemorrhages; one had gastro-intestinal hemorrhages. All of them exhibited Von Kossa positive alterations of the elastic fibers in the medium dermis. Moreover, ultrastructural alterations typical of PXE were shown by electron microscopy on the same skin biopsies used to isolate fibroblasts. Fibroblast cultures were established and maintained according to Quaglino et al. (2000). Briefly, skin biopsies were fragmented and cells were grown as monolayer in 75 cm2 flasks (Nunc, Roskilde, Denmark) in Dulbecco modified eagle medium (DMEM) containing 10% fetal bovine serum (FBS) and used between the third and the eighth passage in vitro. Control and PXE cells were used at the same passage within each experiment. Synchronized cells were obtained as described by Ashihara and Baserga (1979) and the G-0 phase was checked by flow cytometry. All experiments were performed by using synchronized cells.
497
4.3. R1141X mutation identification Genomic DNA was isolated from blood leukocytes of PXE patients and purified with a QIAamp blood kit (Qiagen), according to standard procedures. DNA was first screened for this mutation, by using a rapid technique based on the amplification of exon 24 followed by digestion with the restriction endonuclease BseAI. PCR products were synthesized by amplification of the region of exon 24 spanning the mutation site, by using a mismatch antisense-primer, containing an artificial mutation. This strategy allows the production of a PCR product of 199 bp containing a point mutation that introduces a restriction site for BseAI. In the case of the wild type allele, BseAI digestion generates two fragments of 177 and 22 bp, respectively. In the mutant allele, the C-T substitution, eliminates the BseAI site and the PCR fragment maintains its original size (199 bp). 4.4. Flow cytometry Fibroblasts were plated in 25 cm2 flasks at a density of 2.5=105 cells in 5 ml DMEM with 10% FBS. Different calcein-AM concentrations of 0.01, 0.05, 0.1, 0.25 and 0.5 mM were assayed in order to find the best concentration allowing to perform measurements without any saturation effect. The concentration of 0.1 mM was found to give the best signal in all experiments. Cells were analyzed in an Epics XL flow cytometer with excitation and emission wavelengths set at 488 and 525 nm. Samples were gated on forward scatter vs. side scatter to exclude cell debris and aggregates. Experiments were repeated two or three times in duplicate by using cell strains from six different controls and from eight PXE patients. Data were mediated and significance assessed by unpaired t-test. 4.4.1. Calcein accumulation After 48 h from seeding cells were removed from the tissue culture flasks with 0.25% trypsin in phosphate buffered saline (PBS) for 10 min at 37 8C. Trypsin was blocked by addition of 5 ml DMEM plus 10% FBS. Cells were centrifuged for 5 min at 1000 rpm, washed with PBS and centrifuged again. Pellets were re-suspended in 500 ml of PBS plus 0.1 mM calcein-AM and continuously analyzed for 25 min by flow cytometry. In some experiments, cells were re-suspended in 500 ml of PBS plus 0.1 mM calcein-AM and 10 mg mly1 digitonin and continuously analyzed for 10 min. Fluorescent calcein (Fcalcein) accumulation had a time course that fits well by a single exponential. Rate constants for calcein accumulation (k) were estimated from fits of Eq. (1) to the data: FcalceinsTopUŽ1yeyKt.
(1)
498
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
where Top is the maximum amount (F a.u.) of accumulation, K is the rate constant that describes the process and t is time. Analyses were performed with a PRISM 3.0 GRAPHPAD software (San Diego, CA). 4.4.2. Calcein release Fibroblasts, grown in DMEM plus 10% FBS for 48 h, were incubated for 30 min at 37 8C with 0.1 mM calcein-AM in the presence or in the absence of: (a) 50 mM, 500 mM, 1 mM CHB in DMSO; (b) 5, 100, 500 nM VBL in H2O; (c) 50, 100, 500 mM MK571; (d) 100 mM IDM; (e) 100, 300 mM BNZ; (f) 100 mM VPL. After incubation, cells were detached from tissue culture flasks with 0.25% trypsin for 10 min at 37 8C. Trypsin was blocked by addition of DMEM and 10% FBS. Cells were centrifuged for 5 min at 1000 rpm, washed with PBS, and centrifuged again. The pellet was suspended in 500 ml of PBS and the fluorescence was measured by flow cytometry. The fluorescence of these cells represented time 0. After incubation with calceinAM in the presence or in the absence of competitorsy inhibitors, as described above, parallel sets of cells were washed in PBS and incubated for 30, 60, 240 and 480 min in calcein-free and FBS-free DMEM at 37 8C, in the presence or in the absence of inhibitorsycompetitors. At each time point, the medium was removed and the calcein efflux was stopped by adding ice-cold PBS. Cells were immediately detached from tissue culture flasks and treated for flow cytometry as specified above. Vitality of cells in the presence of the different chemicals and at the various times was checked by staining with propidium iodide and by cell counting. Experiments were done in triplicate and repeated with different cell lines from four controls and four PXE patients. Fcalcein decays with a time course that fits by a single exponential. Rate constants for calcein efflux (k) were estimated from fits of Eq. (2) to the data: FcalceinsTop=eyKt
(2)
where Top is the maximum amount (F a.u.) of accumulation and K is the rate constant that describes the process. For MK571 and IDM, the efflux rates were computed by linear regression analysis. Analyses were performed with a PRISM 3.0 GRAPHPAD software (San Diego, CA). 4.5. Confocal microscopy Fibroblasts, plated in 2 wells ‘chamber slide’ at a density of 3.0=104 cells, were grown for 24 h in 2 ml of DMEM with 10% FBS. After incubation with calceinAM (see further for details) cells were observed with a
confocal laser scanning microscope (TCS 4D) based on a Leica IRBE inverted microscope and equipped with an Ar–Kr laser. An oil immersion Planapo objective at 63= was employed. Experiments were done in duplicate with different cell lines from four control subjects and four PXE patients. 4.5.1. Calcein accumulation After 24 h culture, calcein-AM was added to the culture medium at a final concentration of 0.1 mM for 2 and 30 min. After incubation, cells were washed with PBS, fixed in 3% paraformaldehyde in PBS for 15 min, washed two times in PBS and observed by confocal microscopy. In a series of experiments, after 24 h culture, cells were simultaneously treated with 50-mM MK-571 and 0.1 mM calcein-AM for 30 min at 37 8C, washed with PBS, fixed with 3% paraformaldehyde, washed in PBS, and observed by confocal microscopy. 4.5.2. Calcein release After 24 h culture, calcein-AM was added to the culture medium as above and left for 30 min at 37 8C and 5% CO2. Cells were then washed with PBS, and placed for 15, 30, 60, 240 and 480 min in calcein-free and FBS-free DMEM at 37 8C in 5% CO2. At each time point, the medium was removed and the efflux was stopped by adding ice-cold PBS. Cells were then fixed with 3% paraformaldehyde in PBS, washed in PBS and observed by confocal microscopy. 4.6. Lysosome staining Lysosomes were stained by the neutral red method that is based on the accumulation of the stain by lysosomes of living cells. Fibroblasts from five control subjects and five PXE patients were cultured for 12 h in multi-chamber culture slides in DMEM and 10% FBS and then in DMEM without FBS for 12 h. A 0.4% neutral red solution was prepared in DMEM without FBS and centrifuged at 1500=g for 10 min, in order to remove insoluble crystals. Fresh solutions were prepared for each assay. Cells were rinsed with PBS and incubated with DMEM plus neutral red for 4 h at 37 8C. After incubation, cells were either rinsed with 1 ml of ‘washy fix solution’ (1.3 ml of 37% formaldehyde and 10 ml of a 10% solution of anhydrous CaCl2 in 89 ml of distilled water) for 2–3 min, and observed with a Zeiss Axiophot light microscope. In a series of dishes the washyfix solution was replaced with 1 ml of a solution made of 50 ml ethanol, 1 ml glacial acetic acid and 49 ml of distilled water, incubated for 20 min and the absorbance read in a spectrophotometer at 540 nm. Measurements were done in triplicate.
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
4.7. Statistical analysis Data derived from the different cell strains and obtained in the same experimental conditions were expressed as the mean "S.D. and were compared by the unpaired t-test; differences between groups were considered significant at P-0.05. Acknowledgments This work was supported by Italian MIUR, project no. MM06178355, and by EC, project QLRT-200100332; authors are grateful to PXE International, to PXE-Italia, to patients and families for blood and tissue samples, and to Prof. C.D. Boyd and Dr O. Le Saux for helpful discussion. References Ashihara, T., Baserga, R., 1979. Cell synchronization. Methods Enzymol. 58, 248–262. Baccarani-Contri, M., Boraldi, F., Taparelli, F., De Paepe, A., Pasquali Ronchetti, I., 1996. Matrix proteins with high affinity for calcium ions are associated with mineralization within the elastic fibers of Pseudoxanthoma elasticum dermis. Am. J. Pathol. 148, 569–577. Bacchelli, B., Quaglino, D., Gheduzzi, D., et al., 1999. Identification of heterogygote carriers in families with a recessive form of Pseudoxanthoma elasticum (PXE). Mod. Pathol. 12, 1112–1123. Barnouin, K., Leier, I., Jedlitschky, G., et al., 1998. Multidrug resistance protein-mediated transport of chlorambucil and melphalan conjugated to glutathione. Br. J. Cancer 77, 201–209. Begleiter, A., Mowat, M., Israels, L.G., Johnston, J.B., 1996. Chlorambucil in chronic lymphocytic leukemia: mechanism of action. Leuk. Lymphoma 23, 187–201. Bergen, A.A.B., Plomp, A.S., Schuurman, E.J., et al., 2000. Mutations in ABCC6 cause pseudoxanthoma elasticum. Nat. Genet. 25, 228–231. Borst, P., Evers, R., Kool, M., Wijnholds, J., 1999. The multidrug resistance protein family. Biochim. Biophys. Acta 1461, 347–357. Crivellato, E., Candussio, L., Rosati, A.M., Bartoli-Klugmann, F., Maliardi, F., Decorti, G., 2002. The fluorescent probe Bodipy-FLverapamil is a substrate for both P-glycoprotein and multidrug resistance-related protein (MRP)-1. J. Histochem. Cytochem. 50, 731–734. Dhar, S., Nygren, P., Liminga, G., et al., 1998. Relationships between cytotoxic drug response patterns and activity of drug efflux transporters mediating multidrug resistance. Eur. J. Pharmacol. 346, 315–322. Essodaigui, M., Broxterman, H.J., Garnier-Suillerot, A., 1998. Kinetic analysis of calcein and calcein-acetoxymethylester efflux mediated by the multidrug resistance protein and P-glycoprotein. Biochemistry 37, 2243–2250. Feller, N., Broxterman, H., Wahrer, D., Pinedo, H., 1995. ATPdependent efflux of calcein by the multidrug resistance protein (MRP): no inhibition by intracellular glutathione depletion. FEBS Lett. 386, 385–388. Flanagan, S.D., Cummins, C.L., Susanto, M., Liu, X., Takahashi, L., Benet, L.Z., 2002. Comparison of furosemide and vinblastine secretion from cell lines overexpressing multidrug resistance protein (P-glycoprotein) and multidrug resistance-associated proteins (MRP1 and MRP2). Pharmacology 64, 126–134.
499
Gheduzzi, D., Sammarco, R., Quaglino, D., et al., Extracutaneous ultrastructural alterations in pseudoxanthoma elasticum. Ultrastruc. Pathol., in press. Hamilton, K.O., Topp, E., Makagiansar, I., Siahaan, T., Yazdanian, M., Audus, K.L., 2001. Multidrug-associated-protein 1 functional activity in colu-3 cells. J. Pharmacol. Exp. Ther. 298, 1199–1205. Hipfner, D.R., Deeley, R.G., Cole, S.P.C., 1999. Structural, mechanistic and clinical aspects of MRP1. Biochim. Biophys. Acta 1461, 359–376. Hollo, Z., 1994. Calcein accumulation as a fluorimetric functional assay of the multidrug transporter. Biochim. Biophys. Acta 384, 1191–1198. Homolya, L., Hollo, M., Muller, M., Mechetner, E.B., Sarkadi, B., 1996. A new method for quantitative assessment of P-glycoproteinrelated multidrug resistance in tumor cells. Br. J. Cancer 73, 849–855. Ilias, A., Urban, Z., Seidi, T.L., et al., 2002. Loss of the ATPdepending transport activity in Pseudoxanthoma elasticum-associated mutants of human ABCC6 (MRP6). J. Biol. Chem. 277, 16860–16867. Klein, I., Sarkadi, B., Varadi, A., 1999. An inventory of the human ABC proteins. Biochim. Biophys. Acta 1461, 237–262. Kool, M., van der Linden, M., de Haas, M., Baas, F., Borst, P., 1999. Expression of human MRP6, a homologue of the multidrug resistance protein gene MRP1, in tissues and cancer cells. Cancer Res. 59, 175–182. Le Saux, O., Urban, Z., Tschuch, C., et al., 2000. Mutations in a gene encoding an ABC transporter cause pseudoxanthoma elasticum. Nat. Genet. 25, 223–227. Le Saux, O., Beck, K., Sachsinger, C., et al., 2001. A spectrum of ABCC6 mutations is responsible for Pseudoxanthoma elasticum. Am. J. Hum. Gen. 69, 749–764. Maccari, F., Gheduzzi, D., Volpi, N., 2003. Anomalous structure of urinary glycosaminoglycans in patients with pseudoxanthoma elasticum. Clin. Chem. 49, 380–388. Madon, J., Hagenbuch, B., Landmann, L., Meier, P.J., Stieger, B., 2000. Transport function and hepatocellular localization of MRP6 in rat liver. Mol. Pharmacol. 57, 634–641. Neldner, K., Struck, B., 2002. Pseudoxanthoma elasticum. In: Royce, P.M., Steinmann, B. (Eds.), Connective Tissue and Its Heritable Disorders. Molecular, Genetic and Medical Aspects. Wiley-Liss & Sons, Inc. Publ, pp. 561–583. Ogawa, K., Suzuki, H., Hirohashi, T., et al., 2000. Characterization of inducible nature of MRP3 in rat liver. Am. J. Physiol. Gastrointest. Liver Physiol. 278, G438–446. Pasquali-Ronchetti, I., Volpin, D., Baccarani-Contri, M., Castellani, I., Peserico, A., 1981. Pseudoxanthoma elasticum: biochemical and ultrastructural study. Dermatologica 163, 307–326. Pasquali-Ronchetti, I., Baccarani-Contri, M., Pincelli, C., Bertazzoni, M.G., 1986. Effect of selective enzymatic digestions on skin biopsies from Pseudoxanthoma elasticum patients. Arch. Dermatol. Res. 278, 386–392. Passi, A., Albertini, R., Baccarani-Contri, M., et al., 1996. Proteoglycan alterations in skin fibroblast cultures from patients affected with Pseudoxanthoma elasticum. Cell Biochem. Funct. 14, 111–120. Pulaski, L., Jedlitschky, G., Leier, I., Buchholz, U., Keppler, D., 1996. Identification of the multidrug resistance protein (MRP) as the glutathione-S-conjugate export pump of erythrocytes. Eur. J. Biochem. 24, 644–648. Quaglino, D., Boraldi, F., Barbieri, D., Croce, A., Tiozzo, R., PasqualiRonchetti, I., 2000. Abnormal phenotype of in vitro dermal fibroblasts from patients with pseudoxanthoma elasticum (PXE). Biochim. Biophys. Acta 1501, 51–62. Ringpfeil, F., Lebwohl, M.G., Christiano, A.M., Uitto, J., 2000. Pseudoxanthoma elasticum: mutations in the MRP6 gene encoding a transmembrane ATP-binding cassette (ABC) transporter. PNAS 97, 6001–6006.
500
F. Boraldi et al. / Matrix Biology 22 (2003) 491–500
Scheffer, G.L., Hu, X., Pijnenborg, A.C.L.M., Wijnholds, J., Bergen, A.A.B., Scheper, R.J., 2002. MRP6 (ABCC6) detection in normal human tissues and tumors. Lab. Invest. 82, 515–518. Tiozzo-Costa, R., Baccarani-Contri, M., Cingi, M.R., et al., 1988. Pseudoxanthoma elasticum (PXE): ultrastructural and biochemical
study on proteoglycan and proteoglycan-associated material produced by skin fibroblasts in vitro. Collagen Rel. Res. 8, 49–64. Vertessy, B.G., Kovacs, J., Low, P., et al., 1997. Characterization of microtubule–phosphofructokinase complex: specific effects of MgATP and vinblastine. Biochemistry 36, 2051–2062.
