Isolation and enrichment of melanocytes from human corneal limbus using CD117 (c-Kit) as selection marker

Limbal melanocytes (LM) are located in the basal epithelial layer of the corneoscleral limbus and interact with adjacent limbal epithelial progenitor cells. The exploration of their biological role in the maintenance of the limbal stem cell niche has been limited by the difficulty of LM isolation and cultivation. Here, we report on a facile protocol for the efficient isolation and enrichment of pure populations of human LMs by fluorescence-activated cell sorting (FACS) using antibodies raised against the cell surface marker CD117 (c-Kit). The enriched LMs retain self-renewal capacity and sustained melanin production, and are suitable to study the potential of LMs in stem cell-based corneal tissue engineering.

Limbal melanocytes (LM) are embedded in the basal limbal epithelium and interact with adjacent limbal epithelial progenitor cells (LEPC). LM-derived melanin granules, transferred to LEPC, were suggested to serve as scavengers of reactive oxygen to prevent DNA damage 1,2 . LMs also migrate towards the central cornea after significant epithelial disruption 3,4 . Besides their protective effects, LMs also play an important role in maintaining the stemness state of LEPC both in vitro and in vivo 5,6 . Hayashi and coworkers reported that N-cadherin-positive LMs co-localized with CK15/N-cadherin-positive limbal basal epithelial cells and that N-cadherin-mediated cell-cell interactions are important in maintaining the potency of the epithelial stem cells 7 . LMs are enriched in limbal crypts which also harbor the most pigmented and immature LEPCs 5,8 . LMs from highly pigmented limbal biopsies were shown to support the stemness state and growth of LEPC and to promote the formation of multilayered epithelial cell sheets 5 . The cultivation of melanocytes has proven to be difficult due to the low percentage of melanocytes in the total epithelial cell population and their low mitotic activity compared with the other cell types 9 . This typically results in a rapid overgrowth of melanocyte cultures by keratinocytes and fibroblasts 10 .
Various methods have been suggested to obtain pure melanocyte cultures from epidermal and limbal epithelia, the most common ones rely on differential cytotoxic effects of G418 (geneticin) to prevent rapid overgrowth by epithelial cells and fibroblasts 5,[11][12][13][14] . Although most of these cytotoxic procedures provide pure melanocyte populations, subtle toxic side effects on the phenotype and functional properties of the surviving cells remain a possibility 14 . To avoid these limitations, Hayashi and coworkers used fluorescence-activated cell sorting (FACS) for isolation of LMs based on N-cadherin expression 7 . However, this approach is limited by the fact that only a small fraction of LMs expresses N-cadherin 7 . Recently, Willemsen and coworkers have reported the instant isolation of highly purified human melanocytes from freshly prepared epidermal cell suspensions by means of FACS using CD117 as a selection marker 14 . CD117 (c-Kit) is expressed on both melanocytes and several stem cell populations, and binding to its ligand stem cell factor (SCF) plays an important role in cell homeostasis, including in melanocytes 15,16 . CD117 expression was also observed in LMs, and CD117/SCF signaling was shown to play an important role in the limbal stem cell niche 5,17 .
The aim of this study was to establish and validate a facile technique for the rapid isolation of pure human LM populations from organ-cultured corneal samples by means of FACS using the CD117 selection marker. CD117 + cells were expanded on recombinant E8 fragments of laminin-511 (LN-511-E8), a basement membrane (BM) component of various stem cell niches including the limbal niche 18 . The sorted CD117 + cells were characterized by the expression of established melanocyte markers at the cellular and molecular level. The cell growth www.nature.com/scientificreports/ characteristics and functional properties of LMs also analyzed in this study. This protocol resulted in obtaining a pure population of functional LMs in a very short period of time and can be used for functional studies to further examine the role of melanocytes in the limbal stem cell niche.

Results
Localization of melanocytes and CD117 expression in situ. Immunohistochemical stains of corneoscleral tissue sections revealed that melanocytes (Melan-A + ) occurred throughout the basal limbal epithelial layer (cytokeratin (CK)-15 + ) but were more frequent in crypts as compared to non-crypt regions. Melanocytes (Melan-A + , red) colocalize with clusters of tightly packed CK15 + LEPC cells (green) at the edge of limbal crypts and not with more superficial CK3 + cells (green) (Fig. 1A). Melanocytes in the basal layer coexpress Melan-A and vimentin (green), whereas Melan-A is absent in subepithelial vimentin + cells in the limbal stroma (Fig. 1A). CD117 was expressed in cells interspersed within the basal layer of the limbal epithelium (green) as well as in blood vessels of the limbal stroma (white arrows) (Fig. 1B). Double immunostaining confirmed colocalization of the melanocyte marker Melan-A (red) and CD117 (green) (Fig. 1B). Similarly, cells coexpressing vimentin and CD117 were found in the basal epithelial layer. The epithelial progenitor marker CK15 (red) did not colocalize with CD117, but CK15 + and CD117 + cells were in close contact (Fig. 1B). Subepithelial vimentin + cells did not show expression of CD117 (Fig. 1B).
Flow cytometry analysis and sorting of CD117 + cells. Limbal cluster-derived and stromal fractionderived single cells were stained for CD117 and Melan-A. Subsequent flow cytometry analysis revealed that 4.1 ± 2.4% of limbal cluster-derived cells and 0.2 ± 0.1% of stromal fraction-derived cells expressed CD117 + / Melan-A + (n = 4) ( Fig. 2A). Interestingly, we observed CD117 + /Melan-Astaining in 1.6 ± 0.4% of the cells in the stromal fraction probably representing vascular endothelial cells described above ( Fig. 2A). Based on these observations, we used the CD117 + /Melan-A + fraction of limbal cluster-derived cells for the efficient isolation of LMs.
Growth potential and melanin synthesis of CD117 + cell populations. Pure LM cultures could be passaged for 30 generations with 100 population doublings over 10 months and doubling times of 2.78 ± 0.5 days (Fig. 4A, left column). Moreover, LMs showed similar proliferation potential from passage 2 to passage 30 (Fig. 4A, right column), although they exhibited signs of growth arrest by contact inhibition. We also observed that the LMs did not show any evident changes in morphology between passages.   www.nature.com/scientificreports/ cells compared to untreated cells (Fig. 4B-iii), a sixfold increase was observed in conditioned medium of treated cells compared to control medium (Fig. 4B-iii). These data show that enriched CD117 + LMs are functional in producing and secreting melanin into the culture medium.

Discussion
The limbal stem cell niche is a complex microenvironment composed of specific extracellular matrix (ECM) components and limbal niche cells comprising LEPC, intraepithelial LMs, subepithelial LMSCs and transmigrating immune cells [19][20][21][22][23] . Human LEPCs and LMSCs were successfully grown in culture 18,24 , but LMs have proven difficult to culture efficiently 5 . This may be due to their low numbers in the tissue and a low proliferation rate leading to frequent overgrowth by epithelial cells or fibroblasts in vitro 5,7 . Robust and reliable methods to propagate LMs in vitro would foster a better understanding of their functional roles in limbal stem cell niche maintenance and disease states. Differential trypsinization with low concentrations of trypsin has been suggested to release melanocyte-like cells and stromal fibroblast-like cells for further cultivation, leaving contaminating epithelial cells behind 25 . However, due to their high proliferation rate, contaminating fibroblasts are the main obstacle to achieve pure melanocyte cultures. Several approaches were used to remove contaminating fibroblasts, most commonly the addition of geneticin to inhibit protein synthesis 5,26-28 . At low concentration, geneticin has very limited toxicity for melanocytes showing lower protein synthesis but causes harm to actively synthesizing fibroblast and stromal cells 26,29 . Although this procedure provides pure melanocyte populations, it is very timeconsuming and may also affect the phenotype and functional properties of the melanocytes, since the action of geneticin is not cell type-specific 14,28 .
In this study, we established and validated a novel, alternative method for the isolation and efficient enrichment of melanocytes from the corneal limbus. We used the enzyme collagenase A to isolate limbal cell clusters. Earlier studies have shown that clusters derived by collagenase digestion included more epithelial progenitor cells, LMSCs, and melanocytes than dispase-isolated cell sheets 23,30 . We then targeted the antigen CD117, a cell surface marker expressed in many stem and progenitor cells including melanocytes 5,14 , to isolate LMs from limbal tissue using FACS sorting. Collagenase digestion of limbal tissue results in single-cell suspensions of the limbal stroma (stromal fraction-derived) and limbal epithelial niche cell clusters containing LEPC, LMSC, and LM 18,23 . We observed that cluster-derived cell suspensions contained more CD117 + /Melan-A + cells than the stromal fraction. Hence, we used the cluster-derived cells for FACS sorting to efficiently isolate LMs. The fraction of CD117 + cells in cluster-derived cells varied from 1 to 7.0%, most likely due to donor age, tissue quality, and duration of organ culture. The cultured cell populations still contained small amounts of contaminating stromal cells in most of the cultures (6/7 cultures). Therefore, a second CD117-based FACS sort was performed after 10 days to yield a pure population of LMs as confirmed by the expression of melanocytic markers upon qRT-PCR, immunocytochemistry, flow cytometry, and Western blotting 5 .
Since ECM regulates cell survival, proliferation, and differentiation 31,32 , niche-specific ECM substrates are needed to promote cell propagation in vitro. Laminins are the best described ECM constituents present in BM of adult stem cell niches, including the limbal stem cell niche 18,21 . We reported earlier that the LN-α5 chain is preferentially localized to the BM of the limbal epithelium compared to the corneal epithelium 18 . We have also reported that LN-α5 containing isoform, LN-511, and their biologically active E8 fragments, support the efficient expansion of LEPC 18 . As LMs also reside on the epithelial BM in close association with LEPC in vivo, we used LN-511-E8 for LM expansion in vitro. Our data clearly suggest that LN-511-E8 supports the long term expansion of LMs. Typically, the growth of melanocytes is slow and related to the age of the donor. Neonatal dermal melanocytes can undergo at least 50 population doublings during a period of 6 months, whereas dermal melanocytes from adult skin were reported to proliferate for only 1 month 33 . Uveal melanocytes from adult sources had a doubling time of 3-4 days and were passaged for 23 generations with 35 population doublings over 7 months 34 . In our current study, LMs from cadaveric corneas (average donor age 72.3 years) had a population doubling time as short as 2-3 days, and could be passaged for 30 generations with 100 population doublings over 10 months with unabated proliferative potential. These findings demonstrate that our isolation protocol and subsequent expansion on LN-511-E8 coated plates support long term survival and proliferation of LMs. On the other hand, we also observed signs of growth arrest by contact inhibition, which is a powerful anticancer mechanism that is lost in cancer cells 35 . Moreover, we did not notice any increased cell proliferation in long term LM cultures, which would be one of the associated factors of genomic instability and tumor growth 36 . These observations argue against significant genomic instability in long term LM cultures. A main function of melanocytes is melanin production, which plays a crucial role in many biological processes especially protection against the detrimental effects of UV radiation. In epidermal melanocytes, L-DOPA treatment was reported to induce a 4-fold increase in cellular melanin and a 26-fold increase in secreted melanin 14 . In line with these observations, CD117 + LMs treated with L-DOPA showed a twofold increase in cellular melanin and a sixfold increase in melanin released into the culture medium. These data show that isolated CD117 + LMs are functional in producing and secreting melanin. The protocol has the advantage of rendering pure melanocyte populations within 10 days as compared to 2-3 months needed for a sequence of geneticin treatments 37 .
In conclusion, we propose a new protocol for the isolation and propagation of highly purified populations of primary human LMs by taking advantage of their cell surface expression of CD117 and recombinant LN-511-E8 as a culture substrate. With our protocol, it is possible to obtain large numbers of purified LMs from organ cultured corneal scleral rims. Cells obtained this way expressed established melanocytic markers, showed a multi-dendritic phenotype, sustained self-renewal potential, and upregulated melanin production upon stimulation. The protocol presented carries the advantage to obtain LMs in a very short period of time and to avoid any toxic effects of commonly used selection agents. The pure LM populations can be used for stem cell-based

Materials and methods
Human donor cornea tissue with appropriate research consent was provided by the Lions Cornea Bank Baden-Württemberg after retrieval of corneal endothelial transplants. Informed consent to corneal tissue donation had been obtained from the donors or their relatives. Experiments using human tissue samples were approved by the Institutional Review Board of the Medical Faculty of the University of Freiburg (25/20) and adhered to the tenets of the Declaration of Helsinki. No organs/tissues were procured from prisoners.
Cell isolation. Limbal cells were isolated as previously described 18 . Briefly, organ-cultured corneoscleral tissue ( Table 1; mean age 72.7 ± 10.45 years; culture duration 27.2 ± 7.5 days; light pigmented donor limbal tissue) was cut into 12 three-clock-hour sectors, from which limbal segments were obtained by incisions made at 1 mm before and beyond the anatomical limbus. Limbal segments were enzymatically digested with collagenase A (Sigma-Aldrich; 2 mg/ml) at 37 °C for 18 h to generate clusters containing mixtures of epithelial, mesenchymal and melanocytic cells. Cell clusters were separated from single cells (referred to as stromal fraction-derived cells) by using reversible cell strainers with a pore size of 37 µm (Stem Cell Technologies, Köln, Germany). Subsequently, the cell clusters were dissociated into single cells (referred to as cluster-derived cells) by digestion with 0.25% trypsin and 0.02% EDTA (Pan Biotech, Aidenbach, Germany) at 37 °C for 10-15 min.
To isolate LMs from the cluster-derived single-cell suspension, we used fluorescence-activated cell sorting (FACS) to enrich CD117 + cells (see below). After sorting, CD117 + cells were seeded in LN-511-E8 (iMatrix-511, Nippo; 0.5 µg/cm 2 ) coated T75 flasks (Corning, Tewksbury, MA) and cultured in CNT-40 medium (CellnTec, Switzerland) at 37 °C, 5% CO 2 and 95% humidity. The medium was changed every other day. For further enrichment, the cultures underwent the second cycle of FACS sorting for CD117 + cells after 10 days of cultivation and subsequent expansion in the aforementioned conditions on LN-511-E8. All the following expeirments on LMs carried out on LN-511-E8 substrate, except as noted. www.nature.com/scientificreports/ To enrich LEPC and the associated mesenchymal stromal cells (LMSC), cluster-derived single-cell suspensions were seeded into T75 flasks in Keratinocyte serum-free medium (KSFM) supplemented with bovine pituitary extract, epidermal growth factor (Life Technologies), or Mesencult media (Stem Cell Technologies), respectively. The cultures were incubated at 37 °C, 5% CO 2 , and 95% humidity. The media was changed every other day.

Fluorescence-activated cell sorting (FACS).
Single-cell suspensions were incubated with FcR blocking reagent (Miltenyi Biotec, Bergisch Gladbach, Germany; 20 μl/10 6 cells) for 5 min. Subsequently, cells were washed and incubated with a mouse anti-human CD117-APC antibody (10 μl/10 6 cells) (Miltenyi Biotec) in 100 µl phosphate-buffered saline (PBS), 0.1% sodium azide and 2% fetal calf serum for 30 min at 4 °C in the dark. Cells were then washed and FACS was performed using a FACS Aria II sorter (BD Biosciences, Heidelberg, Germany) and the FACSDiva software (BD Pharmingen; BD Biosciences). Post-acquisition analysis was performed using FlowJo software (Tree Star, Inc., Ashland, OR). Real-time RT-PCR. RNA was isolated from cultured cells using the RNeasy Micro Kit (Qiagen, Hilden, Germany) including an on-column DNase digestion step according to the manufacturer's instructions. Firststrand cDNA synthesis was performed using 2 µg of RNA and Superscript II reverse transcriptase (Invitrogen, Karlsruhe, Germany) as previously described 23 . PCR reactions were run in triplicate using TaqMan Probe Mastermix (Roche Diagnostics, Mannheim, Germany), according to the manufacturers' recommendations. Primer sequences (Sigma-Aldrich) are given in Table 2. For normalization of gene expression levels, ratios relative to the housekeeping gene GAPDH were calculated by the comparative C T method (ΔΔC T ). Genes were considered as differentially expressed when their expression levels exceeded a two-fold difference in all specimens analyzed (n = 5).
Western blot. Total protein was isolated from cells using RIPA buffer (R0278, Sigma-Aldrich) containing a protease inhibitor cocktail (complete Tablets Mini, Roche, Basel, Switzerland). The total protein concentration was evaluated in a colorimetric assay (PierceTM BCA Protein Assay Kit, Thermo Fisher Scientific). Ten micrograms of total protein were separated by SDS-PAGE under reducing conditions, and immunoblot analyses were performed using antibodies (Table 3) against CD117 (1:1000), Melan-A (1:10,000), and GAPDH (1:50,000) followed by horseradish peroxidase-labeled anti-mouse or rabbit IgG (Jackson ImmunoResearch). Protein bands were visualized using the enhanced chemiluminescence Western blot detection reagent (GE Healthcare) and the fusion fx Imager/fusion software (Fusion FX7 Edge 18.05, Vilber Lourmat; https ://www.vilbe r.com/fusio n-fx/). Immunohisto-and immunocytochemistry. Corneoscleral tissue samples (mean age 75.2 ± 10.9 years) within 16 h after death were embedded in optimal cutting temperature (OCT) compound and frozen in liquid nitrogen. Cryosections of 6 μm thickness were cut from the superior or inferior quadrants and LMs cultured on 4 well-glass chamber slides (LabTek; Nunc, Wiesbaden, Germany) were fixed in 4% paraformaldehyde for 15 min, blocked with 10% normal goat serum (NGS), and incubated in primary antibodies (Table 3) diluted in 2% NGS, 0.1% Triton X-100 in PBS overnight at 4 °C or 3 h at room temperature. Antibody binding was detected by Alexa-488-and tetramethylrhodamine (TRITC)-conjugated secondary antibodies (Life Technolo- Table 2. Primers used in qRT-PCR primer assays. Primers were used for Probe based (Universal Probe Library) qRT-PCR assays with an annealing temperature of 60 °C. Population doubling assay. A population doubling (PD) assay was performed as described previously 24 .
Briefly, the assay was performed on LMs from passages 2 to passages 30. Cells were seeded (1 × 10 5 cells) at each passage and trypsinized after 10 days. The population doubling of cells was calculated as: where y is the final density of the cells and x is the initial seeding density of the cells. The cumulative population doublings are the sum of PDs in all passages. Doubling time is calculated from the cell number and the time of cell counting, using the following formula: where t, t 0 represents the time at cell counting; y equals the number of cells at time t, and x equals the number of cells at time t 0 .
Melanin production. Melanin production was assessed as described earlier 14 . Briefly, cells were seeded at 1 × 10 5 cells per well in CNT-40 medium in a 12-well plate and cultured for 24 h at 37 °C in the absence or presence of 1 mM l-3,4-dihydroxyphenylalanine (DOPA) to stimulate melanin synthesis. After 24 h, the culture medium was collected and the adherent cells were washed in PBS. Next, 100 µl of 1 M sodium hydroxide was added to both cells and 100 µl culture medium to dissolve melanin at 70 °C for 90 min. Melanin concentration was determined by comparing 405 nm absorbance values in a Spark microplate reader (TECAN) from experimental samples with a standard curve ranging from 0 to 100 µg/ml generated with synthetic melanin (Sigma). Synthetic melanin was dissolved using 1 M sodium hydroxide solution in water. The fold change values were calculated as OD of the induction/OD of control (n = 7). Statistical analysis. The statistical analyses were performed as described earlier 18 . Briefly, the GraphPad InStat statistical package for Windows (Graphpad Software Inc., Version 6.0, La Jolla, CA; https ://www.graph pad.com/) was used to perform statistical analyses. Results are expressed as mean ± standard error of the mean (SEM) from individual experiments. The statistical significance (p value < 0.05) was determined with the Mann Whitney U test.