Here we describe a protocol for the systematic cultivation of epidermal spheroids in 3D suspension culture. This protocol has wide-ranging applications for use in a variety of epithelial tissue types and for the modeling of several human diseases and conditions.
Epithelial dysregulation is a node for a variety of human conditions and ailments, including chronic wounding, inflammation, and over 80% of all human cancers. As a lining tissue, the skin epithelium is often subject to injury and has evolutionarily adapted by acquiring the cellular plasticity necessary to repair damaged tissue. Over the years, several efforts have been made to study epithelial plasticity using in vitro and ex vivo cell-based models. However, these efforts have been limited in their capacity to recapitulate the various phases of epithelial cell plasticity. We describe here a protocol for generating 3D epidermal spheroids and epidermal spheroid-derived cells from primary neonatal human keratinocytes. This protocol outlines the capacity of epidermal spheroid cultures to functionally model distinct stages of keratinocyte generative plasticity and demonstrates that epidermal spheroid re-plating can enrich heterogenous normal human keratinocytes (NHKc) cultures for integrinα6hi/EGFRlo keratinocyte subpopulations with enhanced stem-like characteristics. Our report describes the development and maintenance of a high throughput system for the study of skin keratinocyte plasticity and epidermal regeneration.
The mammalian stratified epithelium is the most complex epithelial architecture in all living systems and is most often subject to damage and injury. As a protective tissue, stratified epithelium has evolved to generate a complex and effective tissue damage response. Upon injury, these cells must activate lineage plasticity programs, which enable them to migrate to the injured site and carry out repair1,2,3. This multifaceted response occurs in several sequential steps which remain poorly understood.
A major obstacle in studying the intricate process of epithelial regeneration lies in the dearth of high throughput model systems that can capture dynamic cellular activities at defined stages of cell regeneration. While in vivo mouse models offer relevant insight into wound healing and most closely recapitulate the human regenerative process, their development require laborious efforts and significant cost, limiting their throughput capacity. There exists, therefore, a critical need for establishing systems that enable functional investigation of human epithelial tissue regeneration at high throughput scale.
In recent years, several attempts have been made to meet the scalability challenge. This is seen through great expansion of innovative in vitro and ex vivo cell-based models that closely mimic the in vivo regenerative context. This include advances in organ-on-chip4, spheroid5, organoid6, and organotypic cultures7. These 3D cell-based systems each offer unique advantages and present distinct experimental limitations. To date, spheroid culture remains the most cost-effective and widely used 3D cell culture model. And while several reports have indicated that spheroid cultures can be used to study skin stem cell characteristics, these studies have largely been conducted with animal tissue8,9, or with dermal fibroblasts10, with virtually no reports thoroughly characterizing the regenerative properties of human epidermal spheroid cultures. In this protocol we detail the functional development, culture, and maintenance of epidermal spheroid cultures from normal human keratinocytes (NHKc). We equally describe the utility of this system to model the sequential phases of epidermal regeneration and keratinocyte stem cell plasticity in vitro.
The protocol for the collection and handling of skin specimens and isolation of human keratinocytes has been reviewed by the University of South Carolina (UofSC) IRB and classified as "research not involving human subjects", as the foreskin specimens were surgical discards produced during routine surgical procedures (circumcision of neonate boys) and were completely devoid of identifying information. The protocol was also reviewed and approved by the UofSC Biosafety Committee on a regular basis, and all laboratory personnel underwent laboratory biosafety training. All procedures were conducted in concordance to the safety and ethics standards of UofSC.
1. Isolation and culture of human keratinocytes from neonatal foreskin tissue
2. Generating skin epidermosphere cultures in vitro
3. Epidermal spheroid re-plating assay
4. 3D fluorescence cell tracking
5. Characterization of spheroid-derived (SD) sub-populations by FACS
6. Immunofluorescence and staining of epidermospheres for basal stem cell markers
7. Transcriptional analysis of epidermosphere cultures
8. Assessing SD-NHKc colony-forming efficiency
9. Determine SD-NHKc population doublings
During the skin epidermosphere assay, NHKc cultures are seeded in agarose-coated wells of a 96-well plate (Figure 1A). Spheroid-forming cells should self-aggregate within 48h. Autonomous spheroid formation can be assessed as early as 24 h using a standard inverted phase-contrast microscope. skin epidermosphere formation and re-plating assay model various phases of epidermal tissue regeneration (Figure 1B). Figure 2 shows high resolution images of various NHKc strains assayed for epidermal spheroid forming ability in 3D culture. It is important to examine the cells for dense sphere-shape aggregation, as this is a hallmark of spontaneous spheroid formation. We found it necessary to use more than 2 x 104 cells to ensure proper spontaneous aggregation. Non-spherical cell aggregation, such as seen in strains Figure 2A, is not considered adequate epidermosphere formation. Plating non-spheroid forming cell suspensions back in 2D monolayer culture seldom results in viable NHKc cell growth. However, plating of epidermospheres in 2D culture results in the proliferation of small-sized viable NHKc (Figure 3). Images of epidermospheres and SD-NHKc can be viewed and monitored using a standard inverted phase-contrast microscope. It is important to maintain these cultures below 100% confluency as this can considerably impar their growth and stem cell state in culture (Figure 3E-F). The process of epidermal spheroid formation can be functionally tracked at the single cell level by transfecting cells with a fluorescent reporter (Figure 4A-C). Under optimal conditions, keratinocyte subpopulation primarily enriched in SD-NHKc cultures are Integrinα6hi/EGFRlo cells. These cells generally make up about 25% of all SD-NHKc cultures and can be readily isolated by FACS (Figure 4D). However, it is important to establish forward side scatter area (FSC-A) and side scatter area (SSC-A) gates to exclude doublets (Figure 4D). Further characterization of this stem-like keratinocyte subpopulation can be achieved by immunofluorescent staining analysis of epidermal stem cell marker expression, such as basal cytokeratin 14 (K14) and tumor protein 63 (P63) (Figure 4F; Table 2).
Figure 1: Cultivation of NHKc epidermospheres in 3D suspension. (A) Schematic representation of the epidermal spheroid re-plating assay adapted from 11. (B) Representative phase contrast images of NHKc cultures, epidermal spheroids, and SD-NHKc at each sequential step of the epidermal spheroid re-plating assay. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 2: Assessing skin epidermosphere growth. (A) Images of six individual spheroid non-forming and (B) spheroid-forming NHKc strains in floating 3D suspension. (C) Epidermospheres obtained using various amounts of NHKc. (D) Quantification of mean epidermosphere size obtained using different quantities of NHKc in floating 3D suspension culture. Bars indicate standard deviation. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 3: Growing spheroid-derived cultures. (A) Time course phase-contrast imaging of SD-NHKc monolayer cultures propagating from an attached epidermosphere 24 h, (B) 48 h, (C) 72 h, and (D) 96 h after re-plating in 2D plastic culture. (E) SD-NHKc cultures at 80% confluency and (F) 100% confluency 15 and 20 days after replating in 2D plastic culture, respectively. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 4: Characterization of spheroid-derived (SD) sub-populations. (A) Schematic representation of 3D cell tracking assay of epidermospheres in vitro as described by Woappi et al. 202012. (B) EGFP-expressing epidermospheres 2 h and (C) 24 h after seeding in 3D culture. (D) Fluorescence activated cell sorting (FACS) of SD-NHKc subpopulations. Approximately 1/4th of all cells should be integrinα6hi/EGFRlo. (E) Integrinα6hi/EGFRlo subpopulations produce keratinocyte holoclones. (F) Immunofluorescent staining analysis of basal epithelial stem cell marker expression in Integrinα6hi/EGFRlo cells. Scale bar = 50 µm. Please click here to view a larger version of this figure.
Primer sequence (5’-3’) | ||
Gene name | Forward Primer | Reverse Primer |
ALDH1 | GCACGCCAGACTTACCTGTC | CCTCCTCAGTTGCAGGATTAAAG |
EGFR | AGGCACGAGTAACAAGCTCAC | ATGAGGACATAACCAGCCACC |
GAPDH | GGAGCGAGATCCCTCCAAAAT | GGCTGTTGTCATACTTCTCATGG |
K14 | TGAGCCGCATTCTGAACGAG | GATGACTGCGATCCAGAGGA |
KI-67 | ACGCCTGGTTACTATCAAAAGG | CAGACCCATTTACTTGTGTTGGA |
KLF4 | CCCACATGAAGCGACTTCCC | CAGGTCCAGGAGATCGTTGAA |
TP63 | GGACCAGCAGATTCAGAACGG | AGGACACGTCGAAACTGTGC |
β-Actin | CATGTACGTTGCTATCCAGGC | CTCCTTAATGTCACGCACGAT |
ΔN TP63 | ATGTTGTACCTGGAAAACAATGCC | CAGGCATGGCACGGATAAC |
Table 1. Outlines PCR primer sequences used for the detection of select genes involved in neonatal keratinocyte plasticity.
Culture Condition | Phase Contrast Appearance | Immunostaining | FACS Analysis | Transcriptomic signature | |
Homeostatic maintenance | 2D monolayer | Small sized cells < 30 µm | K14, P63 | ITGα6med/EGFPmed | Epidermal maintenance (K14, P63, IVL, K10) |
Differentiation halt/reversal | 3D spheroid | Compact multicellular aggregate > 50 µm | K14, P63, Ki-67 | N/A | De-differentiation: NANOG, SOX2, OCT4, KLF4, K14, P63, Ki-67 |
Stress lineage mobility | 3D-to-2D spheroid attachment | Diffusing small-sized cells (< 20 µm) from spheroid edge | K14, P63, Ki-67 | ITGα6hi/EGFPmed | Proliferation of skin cells: K14, K16, K17, Ki-67 |
Tissue restoration | 3D-to-2D spheroid monolayer | Propagating small-sized cells < 20 µm | K14, P63, IVL | ITGα6hi/EGFPlo and ITGα6hi/EGFPhi | Formation of epidermis: K14, IVL, FLG |
Table 2. Strategies for phenotypic and molecular characterization of the epidermal spheroid replating assay.
The use of 3D spheroid culture systems has had broad utility in assessing cell stemness. These systems have been demonstrated to enhance enrichment of tissue stem cells13, yet their utility for the study of human epidermal stem cells has been limitedly explored. Here, we describe a strategy for enriching human keratinocyte stem cells using 3D culture techniques. In this system, NHKc are cultivated as self-assembling multicellular spheroid suspensions, comprised of several keratinocyte subtypes suspended on top of agarose beds containing KSFM-scm. The setup for this protocol is time sensitive as agar polymerizes rapidly at room temperature. Preheating serological pipettes, micropipette tips, and the agar/cell mixing tube, as well as the media to 42 °C, can dramatically reduce premature polymerization. We observed that placing the 96-well plate in 4 °C shortly after adding agar/media mix to wells can considerably speed up polymerization and ensure that the agar cushion is sufficiently firm for subsequent seeding of cells. It is important to maintain the plate) level at all times during the polymerization process, as poor polymerization of the agar/media mix will result in cells seeding or growing inside the soft agar, voiding the assay.
Also outlined in this protocol, we present a strategy for propagating epidermospheres in 2D monolayer culture to generate stem-like spheroid-derived cells. The epidermal spheroid re-plating assay enriches for a stem cell-like subpopulation of integrinα6hi/EGFRlo keratinocytes. These cells can be used to study epidermal reconstruction, psoriasis, or cellular neoplasia11,12,14. Integrinα6hi/EGFRlo keratinocytes can also be readily isolated by FACS and characterized by immunofluorescent staining. When conducting such experiments, we found it helpful to use unsorted autologous 2D monolayer cells as controls., skin SCC cell lines are a good alternative If these are unavailable.
In summary, this report demonstrates that human epidermal spheroid re-plating models keratinocyte regenerative plasticity in vitro, as it captures each of the four phases of regeneration: homeostatic maintenance, differentiation reversal, stress lineage mobility, and tissue restoration. However, one limitation of agar-based spheroid self-assembly is that not all NHKc stains are capable of spontaneously forming spheroids. The hanging drop method15 is a good alternative strategy to overcome this challenge and to force induce multicellular spheroid formation. This assay can also be multiplexed with muscle, stromal, or immune cells to gain further insight into the contribution of various cell populations on epidermal regeneration. it would be interesting to explore whether addition of Matrigel into the agar could enhance epidermosphere survival and potency.
The authors have nothing to disclose.
The UofSC School of Medicine Instrumentation Resource Facility (IRF) provided access to imaging and cell sorting equipment and technical assistance. This work was supported in part by grant 1R21CA201853. The MCF and the IRF receive partial support from NIH grant P20GM103499, SC INBRE. The MCF also receives support from NIH grant P20GM109091. Yvon Woappi was supported in part by NIH grants 2R25GM066526-06A1 (PREP) and R25GM076277 (IMSD), and by a Fellowship by the Grace Jordan McFadden Professors Program at UofSC. Geraldine Ezeka and Justin Vercellino were supported by NIH grants 2R25GM066526-10A1 (PREP) at UofSC. Sean M. Bloos was supported by the 2016 Magellan Scholar Award at UofSC.
Affymetrix platform | Affymetrix | For microarray experiments | |
Affymetrix’s HuGene-2_0-st library file | Affymetrix | Process | |
Agilent 2100 Bioanalyzer | Agilent | For microarray experiments | |
All Prep DNA/RNA Mini Kit | Qiagen | 80204 | Used for RNA isolation |
Analysis Console Software version 3.0.0.466 | analyze cell type specific transcriptional responses using one-way between-subject analysis of variance | ||
BD FACSAria II flow cytometer | Beckman | For flow cytometry | |
Console Software version 3.0.0.466/Expression console Software | Affymetrix/Thermo Fisher Scientific | For confirming data quality | |
Cytokeratin 14 | Santa Cruz Biotechnology | sc-53253 | 1:200 dilution |
Dispase | Sigma-Aldrich | D4818 | For cell media |
FITC-conjugated anti-integrinα6 | Abcam | ab30496 | For FACS analysis |
GeneChip Command Console 4.0 software | Affymetrix/Thermo Fisher Scientific | For confirming data quality | |
GeneChip Fluidics Stations 450 (Affymetrix/Thermo Fisher Scientific) | Affymetrix/Thermo Fisher Scientific | For washing and staining of hybridized arrays | |
GeneChip HuGene 2.0 ST Arrays | Affymetrix/Thermo Fisher Scientific | For hybridization and amplifycation of total RNA | |
GeneChip Hybridization Oven 640 | Thermo Fisher Scientific | For hybridization and amplifycation of total RNA | Amplify labeled samples | |
GeneChip Hybridization Wash, and Stain Kit (Affymetrix/Thermo Fisher Scientific). | Affymetrix/Thermo Fisher Scientific | For washing and staining of hybridized arrays | |
GeneChip Scanner 3000 7G system | Affymetrix/Thermo Fisher Scientific | Scanning hybridized arrays | |
GeneChip WT PLUS Reagent Kit | Affymetrix/Thermo Fisher Scientific | For amplifycation of biotinylating total RNA | |
Human Basic Fibroblast Growth Factor (hFGF basic/FGF2) | Cell Signaling Technology | 8910 | For cell media |
Human Epidermal Growth Factor (hEGF) | Cell Signaling Technology | 8916 | For cell media |
Human Insulin | Millipore Sigma | 9011-M | For cell media |
iQ SYBR Green Supermix (Bio-Rad) | Bio-Rad | 1708880 | Used for RT-qPCR |
iScript cDNA Synthesis Kit | Bio-Rad | 1708890 | Used for RT-qPCR |
KSFM | ThermoFisher Scientific | 17005041 | Supplemented with 1% Penicillin/Streptomycin, 20 ng/ml EGF, 10 ng/ml basic fibroblast growth factor, 0.4% bovine serum albumin (BSA), and 4 µg/ml insulin |
KSFM-scm | ThermoFisher Scientific | 17005042 | Supplemented with 1% Penicillin/Streptomycin, 20 ng/ml EGF, 10 ng/ml basic fibroblast growth factor, 0.4% bovine serum albumin (BSA), and 4 µg/ml insulin |
MCDB 153-LB basal medium | Sigma-Aldrich | M7403 | MCDB 153-LB basal media w/ HEPES buffer |
NEST Scientific 1-Well Cell Culture Chamber Slide, BLACK Walls on Glass Slide, 6/PK, 12/CS | Stellar Scientific | NST230111 | For immunostaining |
P63 | Thermo Scientific | 703809 | 1:200 dilution |
PE-conjugated anti-EGFR ( San Jose, CA; catalog number ) | BD Pharmingen | 555997 | For FACS analysis |
pMSCV-IRES-EGFP plasmid vector | Addgene | 20672 | For transfection |
Promega TransFast kit | Promega | E2431 | For transfection |
Qiagen RNeasy Plus Micro Kit | Qiagen | For microarray experiments | |
Thermo Scientific™ Sterile Single Use Vacuum Filter Units | Thermo Scientific | 09-740-63D | For cell media |
Zeiss Axionvert 135 fluorescence microscope | Zeiss | Use with Axiovision Rel. 4.5 software |