Melanosome transfer to keratinocyte in the chicken embryonic skin is mediated by vesicle release associated with Rho-regulated membrane blebbing

Abstract

During skin pigmentation in amniotes, melanin synthesized in the melanocyte is transferred to keratinocytes by a particle called the melanosome. Previous studies, mostly using dissociated cultured cells, have proposed several different models that explain how the melanosome transfer is achieved. Here, using a technique that labels the plasma membrane of melanocytes within a three-dimensional system that mimics natural tissues, we have visualized the plasma membrane of melanocytes with EGFP in chicken embryonic skin. Confocal time-lapse microscopy reveals that the melanosome transfer is mediated, at least in part, by vesicles produced by plasma membrane. Unexpectedly, the vesicle release is accompanied by the membrane blebbing of melanocytes. Blebs that have encapsulated a melanosome are pinched off to become vesicles, and these melanosome-containing vesicles are finally engulfed by neighboring keratinocytes. For both the membrane blebbing and vesicle release, Rho small GTPase is essential. We further show that the membrane vesicle-mediated melanosome transfer plays a significant role in the skin pigmentation. Given that the skin pigmentation in inter-feather spaces in chickens is similar to that in inter-hair spaces of humans, our findings should have important consequences in cosmetic medicine.

Figure 1. Melanosome transfer to neighboring keratinocytes in the skin of chicken (Hypeco nera) embryos.

(a) Melanocyte maturation and pigmentation during skin development. Melanocytes are represented as black dendritic cells colored by melanin pigments. Keratinocytes surrounding melanocytes are shown as white cells stained nucleus with DAPI. Arrowheads indicate melanin particles in keratinocytes. (b) Average number of transferred-melanin particles per keratinocyte. 400 keratinocytes were quantified in each embryo. Data are presented as the mean ± SEM. (c) Ex vivo live imaging using a skin of E8 chicken embryo. A skin tissue dissected from embryos was placed in a glass bottom dish with culture medium containing agarose gel, and was immediately observed by confocal microscope. (d) Time-lapse imaging (bright field) revealed melanosomes actively moving in melanocytes. Colored lines indicate trajectories of melanosomes, which were tracked manually using Image J (1 second-intervals, 43 second-duration) (Supplementary video 1).

Figure 2. Gene manipulation and high-resolution imaging analyses of melanocyte precursors.

(a) Plasmids used for stable transgenesis of neural crest/melanocytes by Tol2-mediated gene transfer. (b) Experimental procedure for confocal microscopy with gapEGFP-labeled melanocytes in the skin of Hypeco nera embryos. Tol2-gapEGFP gene was electroporated into the neural tube/neural crest of E2 (HH13) embryos. From manipulated embryos of E7, E9 or E12, a skin was pealed off and placed in a glass bottom dish with agarose-containing culture medium, and was observed by confocal microscope. (c) gapEGFP-labeled melanocytes in the skin at E7. (d and e) Representative images selected from movies (Supplementary video 2,3,4) showing dynamic changes in dendrite behaviors at E7 (d) and E9 (e). Images were obtained using the confocal microscope Carl Zeiss LSM5 PASCAL. White arrows indicate the tip of dendrite (E7) and bleb (E9), respectively.

Figure 3. Melanosome transfer is mediated by plasma membrane vesicles.

(a) A melanocyte dendrite at E12. Lower panels show magnified images of a square selected from time-lapse movies by Carl Zeiss LSM5 PASCAL (Supplementary video 5), and surface-rendering images by IMARIS 7.6 (Bitplane). Arrows show a membrane bleb that was eventually released from the dendrite. (b) From bleb to vesicle changes of the melanocyte plasma membrane during development. Blebbing cells emerge around E6 onward, followed by appearance of vesicle-releasing melanocytes around E10. 30 melanocytes were assessed for each embryo. The definition was made by morphological criteria of membrane tethered EGFP-expressing melanocytes using high-resolution images obtained by confocal microscopy. We define the vesicle-releasing melanocytes as the cells adjacent to which released EGFP-positive vesicles are observed. Blebbing melanocytes are those that exhibit blebs in the plasma membrane. And the rest of the cells are classified as non-blebbing melanocytes. Data are presented as the mean ± SEM. (c) An orthogonal confocal image (Nikon A1R) shows an EGFP-positive membrane vesicle (arrow) incorporated into a keratinocyte, whose shape was visualized by phalloidin staining. (d) Among 185 vesicles examined in E12 embryos, a majority was found within keratinocytes. In each fraction of inside and outside keratinocytes, the number of pigment-containing vesicles was smaller than that of vacant vesicles. (e) A range of different sizes of membrane vesicles with and without a melanosome. 30 vesicles were assessed for each of three embryos (E12). Data are presented as the mean ± SEM. (f) Melanocyte-derived gapEGFP+ membrane vesicles are enriched with phosphatidylserine in the outer lipid bilayer (arrows). Membrane vesicles were isolated from the skin of E12 embryos, and subjected to staining with PSvue 550 (red) (See Material and Methods). Lower panels show magnified images of boxed areas in the upper panels.

Figure 4. Rho activity is required for the membrane vesicle formation and skin pigmentation.

(a) Temporally controlled inhibition of Rho activity using the tet-on system. Doxycycline (Dox) was administered into embryos at E7 and E10 to turn on the DN-RhoA- or C3 genes prior to the stages when blebbing and vesicle release would normally start. Effects by Rho inhibition for respective events were evaluated at the stages indicated. (b–e) Rho inhibition resulted in a marked reduction of both membrane blebbing (b,c) and release of membrane vesicles (arrows) (d,e). The photos in (b) are selected images from Supplementary video 6 (control gapEGFP) and Video 7 (C3). (f) Experimental procedure for pigmentation assay. pT2A-CAGGS-TetOn3G-IRES-Neo^r, pCAGGS-T2TP, and pT2A-TRE-gapEGFP/pT2A-TRE-GAPEGFP-C3 genes were electroporated into the neural tube of Hypeco nera (pigmented strain) at E2. A skin tissue taken from E9 embryo was dissociated into single cells, and gene-electroporated melanocytes were enriched in G418-containing culture medium. Subsequently, an aggregate of enriched melanocytes was transplanted into host embryos of White leghorn (non-pigmented) at E2. The C3 gene turned on by Dox administration at E10. See Material and Methods. (g) Melanin particles in 1 mm² of the skin were quantified by Nikon software NIS-elements. All values of statistical data are shown as the mean ± SEM. Statistical significance was calculated using Student’s t-test: *P < 0.05, **P < 0.005.