Dataset on gene expression profiling of multiple murine hair follicle populations

The murine hair follicle contains several different keratinocyte progenitor populations within its compartments. By using antibodies against CD34, Itgα6, Sca-1 and Plet-1, we have isolated eight populations and compared their Krt10 and Krt14 expressions using fluorescence microscopy. This improved panel was used in our associated article doi:10.1016/j.scr.2016.06.002 (A.P. Gunnarsson, R. Christensen, J. Li, U.B. Jensen, 2016) [1] and the present dataset describes the basic controls for the FACS. We also used imaging flow cytometry to visualize the identified populations as control. A more detailed analysis of the global gene expression profiling is presented, focusing on the pilosebaceous unit. Murine whole-mounts were stained for heat shock protein Hspa2, which is exclusively expressed by keratinocytes with low or no expression of the four selection markers (IRK). Whole-mount labeling was also conducted to visualize Krt79 and Plet-1 coexpression within the hair follicle and quantification on the distribution of Krt79 positive keratinocytes is presented.


a b s t r a c t
The murine hair follicle contains several different keratinocyte progenitor populations within its compartments. By using antibodies against CD34, Itgα6, Sca-1 and Plet-1, we have isolated eight populations and compared their Krt10 and Krt14 expressions using fluorescence microscopy. This improved panel was used in our associated article http://dx.doi.org/10.1016/j.scr.2016.06.002 (A.P. Gunnarsson, R. Christensen, J. Li, U.B. Jensen, 2016) [1] and the present dataset describes the basic controls for the FACS. We also used imaging flow cytometry to visualize the identified populations as control. A more detailed analysis of the global gene expression profiling is presented, focusing on the pilosebaceous unit. Murine whole-mounts were stained for heat shock protein Hspa2, which is exclusively expressed by keratinocytes with low or no expression of the four selection markers (IRK). Whole-mount labeling was also conducted to visualize Krt79 and Plet-1 coexpression within the hair follicle and quantification on the distribution of Krt79 Table   Subject area Biology.

More specific subject area
Epidermal stem cells.
Type of data Images (fluorescence microscopy), diagrams. How data was acquired Images were acquired using confocal LSM 710 and Leica Letiz DMRB microscopes. Venn diagrams were generated using GeneSpring GX (Agilent) from global microarray gene expression profiling.

Data format
Raw, merged, filtered and analyzed.

Experimental factors
Keratinocytes were isolated from murine dorsal skin after trypsinization for 16 h at 4°C. Murine tail whole-mounts were obtained by keeping the skin in PBS with 5 mM EDTA for 5 h and 37°C.

Experimental features
Murine tail whole-mounts and isolated populations were stained and visualized using fluorescence microscopy. Differently expressed genes were obtained from microarray data using GeneSpring GX software and ANOVA with Benjamin-Hochberg multiple correction (p-value o0.01).

Data source location
AROS Biotechnology (Skejby, Denmark). Aarhus University (Aarhus Denmark). Data accessibility Data is within this article.

Value of the data
To give further insight on how the murine hair follicle keratinocytes can be categorized into different populations.
To deepen the knowledge on how the populations relate to each other by their gene expression profiles.
To understand where the different hair follicle populations are located.

Data
Dorsal keratinocytes were isolated from female C57Bl/6 mice in the age of 7-9 weeks and sorted using flow cytometry. Cells were stained with conjugated antibodies and visualized by fluorescence microscopy. Global microarray data was used to generate Venn diagrams, which show the number of shared differently expressed genes (DEGs) between the different populations and a reference. See Figs. 1-6.

Isolation of murine dorsal keratinocytes
Female 7-9 weeks old C57Bl/6 mice were sacrificed by cervical dislocation. Their backs were shaved using razor machine and the dorsal skin peeled off with sterile forceps and scissors. The dermal fat was scraped away using sterile scalpel and the skin was disinfected in 1% Betadine for 30 s, 70% EtOH for 30 s and washed in PBS. With the epidermal side facing upwards, the skins were kept floating on 0,25% trypsin supplemented with 5 mM EDTA over night at 4°C. Following day, the epidermal cells were gently scraped away from dermis using sterile scalpels into chilled DMEM with 10% FCS and 100x penicillin-streptomycin, centrifuged at 200 g, 4°C for 10 min and washed in PBS supplemented with 0.1% Bovine Serum Albumin.

Flow cytometry and immunofluorescence staining
To isolate the different populations using flow cytometry, freshly isolated keratinocyte cell suspension was stained for 30 min on ice with surface antibodies Brilliant Violet 421-conjugated CD34 (clone RAM34; BD Biosciences); PE-conjugated Itgα6 (CD49f, clone GoH3; BD Biosciences); PE-Cy/7conjugated Sca-1 (Ly-6A/E, clone D4; BD Biosciences); APC-conjugated Plet-1 (clone 33A10; Mubio) using Lighting-link labeling kit (Innova Biosciences). This panel of four antibodies has been shown to increase the number of populations within the murine hair follicle and in particular increase the resolution of the pilosebaceous region by flow cytometry [1]. To ensure that our multipanel of antibody-fluorochromes does not generate false positive cells from spectral overlap, we conducted a fluorochrome minus one (FMO) assay. In essence, by staining the keratinocyte suspension with all antibodies except one, any false emission signal for that fluorochrome will be revealed. This assay of antibody-exclusion was made for all antibodies in separate tubes (Fig. 1).
Using FACSAria III cell sorter (405 nm violet, 488 nm blue, 561 nm yellow-green and 633 nm red lasers), keratinocytes were first gated on viability ( Fig. 2A) and small size (Fig. 2B) after which the different epidermal populations were identified and sorted on slides or into tubes with chilled lysisbuffer. Large cells of the sebaceous glands with intermediate levels of Sca-1 were avoided due to the size-exclusion (Fig. 2C-E). In order to acquire single cell high-magnification images of from each population, freshly isolated dorsal keratinocytes stained against Sca-1, CD34, Itgα6 and Plet-1 were analyzed though a ImageStream flow cytometer (Fig. 3). After sorting the populations onto glass slides, the cells were fixated for 30 s in methanol and stained with anti-Keratin 10 or anti-Keratin 14 primary antibodies with secondary antibody Alexa Fluor 488 and visualized using fluorescence microscopy (Leica Letiz DMRB) (Fig. 4A and B).   Freshly isolated murine dorsal keratinocytes were stained with surface-antibodies against Sca-1, CD34, Itgα6 and Plet-1, after which single cell images were acquired using ImageStream flow cytometer. Propidium iodide (PI) was used to exclude dead cells. Polarized Itgα6 staining can be visualized on the cell surface, representing the part of the keratinocytes that is connected to the basal layer [2,3]. In accordance to the flow cytometry dot plots (Fig. 2D) The population Plet-1 negative population IFE generally shows stronger Sca-1 staining than the Plet-1 positive population PletA6Sca. In addition, Plet-1 staining were more intense in the Plet population compared to PletA6 and PletA6Sca. A polarized cell-surface localization of the CD34 protein is seen at the at the opposite side of Itgα6, which can be visualized in Bulge.

Venn diagrams and the number of shared differently expressed genes (DEGs)
After sorting each of the populations into separate tubes containing lysis-buffer, their total RNA were isolated using Maxwell 16 LEV simplyRNA kit (Promega) and profiled on MouseWG-6 v2 BeadChips (Illumina). By generating Venn diagrams, the numbers of upregulated-and downregulated differently expressed genes (DEGs) were visualized between one reference population and the four other populations (Fig. 5A-D).

Whole-mount preparation and visualization of Hspa2-and Krt79-expression
Murine tail skin was isolated and peeled of the bone after performing a longitudinal section using a scalpel. The skin was cut into 0.5 Â 0.5 cm 2 pieces and kept in PBS with 5 mM EDTA at 37°C for 5 h after which whole-mount epidermal sheets were gently separated from dermis using forceps. The pieces were fixated in 2% neutral buffered formaldehyde Cellpath Ltd) for 10 min and kept in PBS at 4°C until further processing. The sheets were stained with gentle movement for 1 h using primary antibodies PE-conjugated anti-Itgα6 (CD49f, clone GoH3, and BD Biosciences), anti-Hspa2 (Clone EPR4596; Abcam), anti-Keratin 79 (clone Y-17; Santa Cruz biotechnology) and secondary Alexa Fluor 488 antibody. Images were acquired using fluorescence microscope (Leica Letiz DMRB) or LSM 710 confocal microscope (405 nm, 488 nm, and 633 nm lasers; Carl Zeiss) (Fig. 6A1-F).