SEARCH

SEARCH BY CITATION

Keywords:

  • cell culture;
  • keratinocytes;
  • melanocytes;
  • melanosome transfer;
  • pigmentation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Abstract:  Three major difficulties must be overcome to establish a quantitative method for melanosomal transfer analysis: (i) establishing a three-dimensional co-culture reassuring direct melanocyte to keratinocyte transfer, (ii) separation of melanocytes and keratinocytes following co-culture and (iii) melanosome quantification in each cell population. Melanocytes and keratinocytes are cultured on the opposite sides of the porous membrane of hanging cell inserts (1 μm pores, 2 × 106 pores/cm2). Cell separation is performed after 3 days of co-culture by simple trypsinisation. Melanosome quantification in separated cell populations was accomplished by an ELISA-like method using gp-100 as the antigen. Melanocytes and keratinocytes come into ‘direct’ contact through the pores, and melanosomal transfer is accomplished without cell passage through the membrane. Cell separation by simple trypsinisation results in pure melanocyte and keratinocyte populations. Melanosome quantification by the ELISA-like method proved to be sensitive and specific to distinguish the known inhibitors and inducers of melanosomal transfer.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Melanocytes synthesise melanin pigments within melanosomes. Melanosomes undergo intramelanocytic trafficking that implies the interaction of the organelle with microtubules through the plus-end- and minus-end-directed cellular motors kinesin II and dynein, respectively (1). Anterograde movement of melanosomes to melanocytic dendrites is primarily mediated by kinesin II (2), but both motors tightly cooperate to facilitate melanosomal movement to dendritic tips (3–5). Microtubules and actin cytoskeleton thus play an important role for the whole process of melanosomal transfer from melanocytes to keratinocytes (5). Despite many efforts to observe the actual transfer of melanosomes between melanocytes and keratinocytes, either in skin or in artificial epidermal models, there is still no consensus on how melanosomal transfer occurs thereafter (6,7) and the mechanism of this process is still considered as a ‘black box’ (7). The most generally accepted hypothesis, however, is the phagocytosis of melanocyte dendrite tips by keratinocytes (8,9). According to this hypothesis, once melanosomes are trafficked into dendrite tips, melanocytes would take the role of inactive bystanders until the phagocytosis-mediated transfer is accomplished by keratinocytes (9–12). More recently, the model suggesting a role for filopodia in melanosomal transfer was further strengthened by using a method that directly allows visualisation of transferred melanosomes (13,14). On the other hand, pharmacological modulation of melanosomal transfer has recently gained much attention as a new strategy for modifying skin pigmentation for therapeutic purposes (15), raising the requirement of a screening method to identify transfer inhibitors and inducers.

Herein, we present a new method for three-dimensional co-culture of melanocytes and keratinocytes that permits direct melanocyte to keratinocyte melanosomal transfer followed by simple and reliable cell separation and simple spectrophotometric quantification of melanosomes in both cell types. This is a rapid, non-expensive and simple assay that can be performed in almost any laboratory with standard equipments. We believe that the present model could serve as a screening assay to identify melanosomal transfer modifiers.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Unless otherwise specified, all chemicals were purchased from Sigma (Buchs, Switzerland). Cultures are incubated at 37°C in an atmosphere with % CO2. See the complete description in Data S1.

Cell cultures and pretreatments

Murine B16 melanoma cells were cultured in DMEM/FCS 10% supplemented with 1 nm cholera toxin, 100 μm 3-isobutyl-1-methylxanthine (IBMX), 100 U/ml penicillin G and 100 μg/ml streptomycin. The medium was changed every day during 7 days. Melanocytes were then harvested, and 5 × 105 melanocytes were cultured during five consecutive days. The treatment with melanosome transfer modifiers was done during this 5-day period. The medium was changed every day.

Murine BDVII keratinocytes were cultured in DMEM devoid of cholera toxin and IBMX for 5 days as described previously (16). The treatment with melanosome transfer modifiers was done during this 5-day period.

The three-dimensional co-culture

Millicell-96 Cell Culture Insert Plate (Millipore/Milian, Geneva, Switzerland) was placed ‘up-side down’ in a sterile container. The keratinocyte medium was added into the container up to the level of insert membranes. Seven hundred μl of a BDVII keratinocyte suspension (5 × 105 cells) was added on each membrane, and then the container was covered with a sterile air-permeable cover and incubated overnight. Cell inserts were taken from the container, washed with PBS (with Ca2+ and Mg2+), hanged in 6-well insert plates and filled with keratinocyte medium (Figure S1).

A suspension containing 7 × 105/ml B16 melanoma cells was prepared in DMEM without cholera toxin or IBMX. Three hundred μl of the suspension (2 × 105 melanoma cells per membrane) was then put inside each hanging insert. Some inserts never co-cultured with melanocytes (‘naïve’ keratinocytes) served as negative controls in melanosome quantification. Co-cultures were incubated during 3 days. The medium was changed every day without any treatment during the co-culture condition.

Cell separation

After 3 days of co-culture, the medium was removed, the hanging inserts were washed with PBS (without Ca2+ and Mg2+), and then both cell types were detached and separated using trypsin-EDTA. Melanoma cells were harvested first, and then the keratinocytes. Both cell types were then counted (=‘Ktot’ and ‘Mtot’ in transfer index formula).

Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Melanoma cells and keratinocytes were washed with PBS and transferred separately into conical 1.5 ml Eppendorf tubes. All the procedure was performed at room temperature. Centrifugations were accomplished at 340 g.

  • 1
     Melanoma cells or keratinocytes were fixed in 3% paraformaldehyde in PBS (with Ca2+ and Mg2+) during 10 min.
  • 2
     Samples were centrifuged, and paraformaldehyde was removed.
  • 3
     One millilitre 0.05% saponin in PBS was added to each tube, and the cell pellet was pipetted for 5 s. Cells were incubated at room temperature during 15 min.
  • 4
     Saponin solution was removed after centrifugation, and 1 ml of a solution containing 3% bovine serum albumin (BSA) and 0.05% saponin in PBS was added to each tube and incubated for 1 h.
  • 5
     HMB45 antibody (mouse anti-human HMB45 IgG1, Enzo Life Sciences, Lausen, Switzerland) was diluted 1:10 in saponin/BSA solution, and then 500 μl of the dilute was added to the cell pellet and incubated for 45 min. Samples were then centrifuged and washed twice with the saponin/BSA solution.
  • 6
     The second antibody (goat anti-mouse IgG (H + L) HRP-conjugated; BioRad, Reinach, Switzerland) was diluted 1:1000 in the saponin/BSA solution, and then 1 ml of the dilute was added to each cell pellet and incubated for 30 min.
  • 7
     Tubes were centrifuged, and the supernatant was removed. Cells were washed four times (twice with saponin/BSA, once with 0.05% saponin and once with PBS).
  • 8
     Cells were counted after the last wash. From 2 × 105 to 5 × 105 keratinocytes were transferred into new Eppendorf tubes (=‘K’ in transfer index formula).
  • 9
     8 × 104 melanoma cells were put in each of the three tubes per condition, using new tubes (=‘M’ in transfer index formula).
  • 10
     Three hundred micro litre of the ‘substrate’ solution [TMB (3,3′,5,5′-tetramethylbenzidine dihydrochloride)/H2O2; MBL/LabForce, Nunningen, Switzerland] was then added to each tube, and samples were incubated for 45 min in a light-protected place. The colour of the supernatants turned blue at this stage.
  • 11
     Three hundred micro litre of the ‘stop’ solution (1 m sulphuric acid; MBL/Lab Force) was added to all samples at the end of the incubation period. Supernatants turned yellow at this stage. Samples were incubated for 5 min and then centrifuged, and the absorbance of the supernatant was determined at 450 nm.
  • 12
     Controls were performed by omitting the immunolabelling with the first antibody (step 4) to verify the specificity of the signal. Naïve keratinocytes served as negative control.

Confocal microscopy

Melanoma cells and keratinocytes were grown on each side of a Millicell insert membranes as described earlier. Membranes were then recovered and fixed in paraformaldehyde 1% for 30 min. Membranes were permeabilised with saponin 0.05% and then sequentially labelled with goat anti-human Melan-A (Santa Cruz, Nunningen, Switzerland) coupled to Alexa Fluor® 647 chicken anti-goat IgG (Invitrogen, Basel, Switzerland), with mouse anti-human HMB-45 (Enzo Life Sciences) coupled to Alexa Fluor® 546 donkey anti-mouse IgG (Invitrogen), with rabbit anti-mouse keratin-5 (Biocompare/VWR, Dietikon, Switzerland) coupled to Alexa Fluor® 488 donkey anti-rabbit IgG (Invitrogen) as well as for nuclei with DAPI (4,6′-diamidino-2-phenylindole). Membranes were finally processed for confocal microscopy analysis with a Zeiss LSM 510 Meta confocal microscope (Carl Zeiss, Göttingen, Germany). Images were acquired at a magnification of 63× and analysed with Adobe Photoshop software (Adobe Systems, San Jose, CA, USA).

Flow cytometry

Flow cytometry analyses were performed on melanoma cells and keratinocytes after 3 days of co-culture on membranes. Cells were trypsinised at the end of co-culture and were immunolabelled using HMB45 as first antibody (diluted 1:2 in saponin/BSA) and goat anti-mouse IgG-allophycocyanin conjugated (Invitrogen), diluted 1:20 in saponin/BSA, as second antibody.

Tubes were centrifuged and the supernatant was removed. Cells were washed four times (twice in saponin/BSA, once in 0.05% saponin and once with PBS) to remove the second antibody.

Cells were resuspended in 1 ml PBS before the flow cytometry analysis (FACSCalibur, BD Biosciences, Franklin Lakes, NJ, USA).

Electron microscopy

Hanging cell inserts were washed in PBS then immerged in 3% phosphate-buffered glutaraldehyde pH 7.4 for 2 h at 4°C. Inserts were then postfixed in 1% osmium tetroxide in colloidin buffer (pH 7.4) for 2 h at 4°C. Membranes were cautiously detached from the hanging inserts by a scalpel blade, dehydrated in acetone and embedded in Araldit®. The sections were collected on Singleslot 2 × 1 mm grids coated with Formvar films. Samples were stained with 7% uranyl acetate in methanol and poststained with Reynold’s lead hydroxide and examined by electron microscopy (Philips CM10, Philips, Amsterdam, The Netherlands).

Statistics

For replicate experiments with numerical values, an analysis of variance was performed using Tukey comparison test. Statistically significant results were indicated (***P < 0.001).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Co-cultures

Electron microscopic examinations showed melanoma cells dendrites penetrating through the pores of the membrane (Fig. 1a) and confirmed the direct melanocyte–keratinocyte contact through membrane pores already 24 h after the co-culture (Fig. 1b), as well as the presence of melanosomes in co-cultured keratinocytes (Fig. 1c). Melanosomes could be observed in keratinocytes after 48 h of co-culture, while no melanosomes could be identified in naïve keratinocytes (i.e. keratinocytes that were never co-cultured with melanoma cells).

image

Figure 1.  Electron microscopy of melanocyte–keratinocyte contacts. (a) Melanocytes with dendrites that penetrate through the pores of the membrane. (b) Melanocyte dendrites in contact with keratinocyte pseudopodia within the pores. (c) Melanosomes transferred to keratinocytes on the other side of the membrane. Magnification: 2200×.

Download figure to PowerPoint

Flow cytometry

The two cell populations were separated from the membrane by trypsin-EDTA treatment and then analysed by flow cytometry using HMB45. The different controls performed with or without the first antibody on the co-cultured keratinocytes, as well as naïve keratinocytes, ruled out the non-specific antibody reactions as the source of the signal. The signal-to-noise ratio was 7.10 for co-cultured melanoma cells, 2.04 for co-cultured keratinocytes and 1.05 for naïve keratinocytes (not shown).

Confocal microscopy analysis

Immunolabelling of co-cultured cells on membranes by melanocyte-, keratinocyte- and melanosome-specific markers confirmed the transfer of melanosomes to keratinocytes and excluded cell passage through membrane pores (Fig. 2). Segregated populations of keratinocytes were observed in one side of the membrane (Fig. 2a–f), whereas pure melanocytes were observed on the other side (Fig. 2g–l). Interestingly, points of contact between melanoma cells and keratinocytes were visualised at the level of the interconnecting pores of the membranes (Fig. 2m–r), suggesting that both cell types were allowed to get in close contact across the membranes. More importantly, transfer of melanosomes, as assessed by HMB-45 immunostaining, was also observed on the keratinocytes layer (Fig. 2c,f). Co-cultures were not damaged during trypsinisation despite the dense incorporation of their processes into membrane pores. This was shown by trypan blue examination of the trypsinised cells showing always <10% damaged cells, being comparable to results from ordinary cultures.

image

Figure 2.  Confocal microscopy analysis of the co-culture (magnification 63×). B16 melanocytes and BDVII keratinocytes were co-cultured on each side of a Millicell insert observed under bright field (e, k, q) and stained for keratin 5 (a, g, m; green), Melan-A (b, h, n; blue), HMB45 (c, i, o; red) and nuclei (d, j, p; blue). f, l and r pictures show the superposition of a–c, g–i and m–o pictures, respectively. Pictures a–f (3× zoom compared to the others) represent the keratinocyte side, g–l the melanocyte side and m–r the porous membrane with melanocyte dendrites and keratinocyte pseudopodes.

Download figure to PowerPoint

CELISA – melanosome transfer index

The amount of transferred melanosomes into keratinocytes depends on the keratinocyte/melanoma cell ratio on the membrane. Considering that before co-culture no melanosome is present in keratinocytes, the melanosomes found in keratinocytes were transferred from melanoma cells to keratinocytes during the co-culture. We thus define the melanosome transfer index as the proportion of melanosomes found in keratinocytes compared to the total number of melanosomes in the co-culture and calculate it according to the following formula:

  • image

where AK = A(450) for keratinocytes, Ktot is the number of keratinocytes on the membrane at the moment of analysis, K is the number of keratinocytes analysed by CELISA, AM = A(450) for melanoma cells, Mtot is the number of melanocytes on the membrane at the moment of analysis, and M is the number of melanocytes analysed by CELISA. The CELISA assay, which labels the melanosomes in a cell population, is proportional to the cell number for a given population (correlation coefficient R2 = 0.9984). The melanosome transfer index increases with the duration of the co-culture up to 72 h, which is the maximum duration for such a co-culture, indicating a transfer of melanosomes from melanoma cells to keratinocytes over the time (Fig. 3a). IBMX, a melanogenic compound that increases the intracellular cAMP concentration (17), increased the melanosome transfer index, an effect prevented by nicotinamide and soybean trypsin inhibitor, two agents that decrease melanosomal transfer in vitro (18,19) (Fig. 3b). Ebselen, a small cell-permeable glutathione peroxidase mimic, when applied at 40 μm to melanoma cells before the co-culture, decreased the melanosome transfer index by 50% (Fig. 3c). The pretreatment of BDVII keratinocytes with ebselen did not affect melanosomal transfer.

image

Figure 3.  Melanosome transfer index. (a) Melanosome transfer index as a function of co-culture time. B16 melanocytes and BDVII keratinocytes were co-cultured on each side of Millicell inserts for various periods of time and analysed using the CELISA, and then the melanosome transfer index was determined. (b) Effect of pro- and anti-melanogenic compounds on the melanosome transfer index. B16 melanocytes were pretreated with 100 μm 3-isobutyl-1-methylxanthine (IBMX) for 5 days (except for negative control), BDVII keratinocytes were pretreated for 5 days with 10 μm nicotinamide (NA), 0.1% soybean trypsin inhibitor (STI) or vehicle, then both cell populations were co-cultured on each side of Millicell inserts for 3 days, and the melanosome transfer index was determined following CELISA assay. ***Value significantly higher than untreated cells (p < 0.001); ###Values significantly lower than cells treated with IBMX only. (c) Effect of ebselen on the melanosome transfer index. B16 melanocytes or BDVII keratinocytes were pretreated with 40 μm ebselen for 5 days (except for negative control). Cells were then co-cultured on each side of Millicell inserts for 3 days, and the melanosome transfer index was determined following CELISA assay.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Three major difficulties must be overcome to establish a reliable, simple and specific assay for melanosome transfer: first, establishing a co-culture of melanocytes and keratinocytes in which melanosomal transfer via cell to cell contact can be achieved; second, separation of melanocytes from keratinocytes after such a co-culture; and third, the quantification of melanosome content in each cell population (14,20).

Starting from the hypothesis that the uptake of melanosomes is mediated by phagocytosis, some assays have chosen to quantify the phagocytotic activity of keratinocytes, including their uptake of fluorescent latex beads (21) and/or purified melanosomes (22) as an index of melanosomal transfer. Such methods ignore the role of melanocyte to keratinocyte contact and the possible implication of melanocytes in the transfer process. More advanced methods have used two-dimensional co-cultures of melanocytes and keratinocytes together with the fluorescent labelling of melanocytes and tracing the transfer of the melanocyte fluorescent particles into keratinocytes (23). However, such systems do not distinguish between a direct transfer by cell to cell contact and melanosome uptake by phagocytosis of melanocyte debris in the medium (14).

In the present method, a three-dimensional co-culture is established. This co-culture method permits a direct melanocyte to keratinocyte contact and precludes a significant medium exchange between the chambers. To analyse the medium exchange between the chambers, we filled the higher chamber with a 1/1000 dilution of trypan blue (0.04%) in the medium and determined its concentration in the lower chamber medium after up to 6 h of incubation, by spectrophotometric analysis (24 h co-cultures, 100% cell confluence). After 6 h, a 15-fold lower trypan blue concentration was detected in the lower chamber compared to the higher chamber (data not shown). Considering the huge size of melanocyte debris/melanosomes compared to the trypan blue molecule, the free passage of melanocyte debris in the melanocyte medium into the lower chamber medium appears to be of low probability. In our model, all melanosomes found in keratinocytes were transferred by melanoma cells, as documented by flow cytometry analyses. Simple trypsinisation was shown to be sufficient to obtain pure melanocyte and keratinocyte populations. Immunocytochemistry as well as flow cytometry analyses showed that HMB45-labelled melanosomes were transferred to keratinocytes after co-culture in our model. To evaluate whether HMB45 labelling is valuable in transfer quantifications, we examined several known inhibitors or inducers of melanosomal transfer in our system. HMB45 labelling proved to be quite coherent in the determination of both transfer inhibition and induction in all these conditions. HMB45 was thus valuable in the detection of melanosomal transfer at least in our in vitro system. Therefore, although far from physiologic conditions, this model appears suitable for obtaining homogenous melanosomal transfer.

The fact that in the present method, melanoma cells or keratinocytes can be pretreated separately allows one to distinguish which cell type is implicated in the regulation of melanosomal transfer. For example, in the case of ebselen, the pretreatment of melanocytes and not keratinocytes resulted in the transfer inhibition. This confirmed our previous observations indicating that melanocyte is the target cell for the ebselen inhibition of melanosomal transfer. These findings provide evidence for an active role of melanocytes in the transfer process.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

All mentioned authors fulfil the criteria for authorship: BK, MP, DSN, PC and OS performed the research, BK and OS designed the research study, VP and DS provided essential reagents and tools, BK, OS and JHS analysed the data, and BK and OS wrote the paper.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Melanosome immunolabelling of the cells [Cell-ELISA (CELISA)]
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Conflict of interests
  10. References
  11. Supporting Information

Data S1. Methods.

Figure S1. Co-culture scheme. (a) Beaker with cell culture inserts upside down to culture keratinocytes. (b) Medium aspiration with a pipette under vacuum. (c) BDVII keratinocyte monolayer on the lower side of a Millicell culture insert. (d) B16 melanocyte monolayer on the upper side of a Millicell culture insert. (e) Separation both cell populations with trypsin-EDTA. (f) CELISA scheme. A suspension of melanosome-containing cells is permeabilised by saponin and incubated with HMB45 antibody, then a second anti-IgG antibody coupled to HRP is applied, followed by TMB (lmax& =& 285& nm), a HRP substrate, producing a blue radical intermediate (lmax& =& 652& nm) that is then converted to a stable yellow diimino oxidation product (lmax& =& 450& nm) by addition of hydrochloric acid. The absorbance of the supernatant is read at 450& nm, and this value is proportional to the concentration of melanosomes in the starting cell suspension.

FilenameFormatSizeDescription
EXD_1356_sm_FigS1.jpg348KSupporting info item
EXD_1356_sm_Suppl-Material.doc52KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.