Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Aug 17;548(7667):334-337.
doi: 10.1038/nature23304. Epub 2017 Aug 2.

Correction of aberrant growth preserves tissue homeostasis

Samara Brown  1 , Cristiana M Pineda  1 , Tianchi Xin  1 , Jonathan Boucher  1 , Kathleen C Suozzi  1 , Sangbum Park  1 , Catherine Matte-Martone  1 , David G Gonzalez  1 , Julie Rytlewski  2 , Slobodan Beronja  2 , Valentina Greco  1   3   4   5   6
Affiliations

Correction of aberrant growth preserves tissue homeostasis

Samara Brown et al. Nature. .

Abstract

Cells in healthy tissues acquire mutations with surprising frequency. Many of these mutations are associated with abnormal cellular behaviours such as differentiation defects and hyperproliferation, yet fail to produce macroscopically detectable phenotypes. It is currently unclear how the tissue remains phenotypically normal, despite the presence of these mutant cells. Here we use intravital imaging to track the fate of mouse skin epithelium burdened with varying numbers of activated Wnt/β-catenin stem cells. We show that all resulting growths that deform the skin tissue architecture regress, irrespective of their size. Wild-type cells are required for the active elimination of mutant cells from the tissue, while utilizing both endogenous and ectopic cellular behaviours to dismantle the aberrant structures. After regression, the remaining structures are either completely eliminated or converted into functional skin appendages in a niche-dependent manner. Furthermore, tissue aberrancies generated from oncogenic Hras, and even mutation-independent deformations to the tissue, can also be corrected, indicating that this tolerance phenomenon reflects a conserved principle in the skin. This study reveals an unanticipated plasticity of the adult skin epithelium when faced with mutational and non-mutational insult, and elucidates the dynamic cellular behaviours used for its return to a homeostatic state.

PubMed Disclaimer

Figures

Extended Data Figure 1
Extended Data Figure 1. Increasing the number of mutant cells increases the severity of aberrancies
a, Diagram of the increasing severity of aberrancies that develop as the tamoxifen dosage (and number of mutant cells) increases. b–d, Representative hair follicles induced with increasing amounts of tamoxifen (TAM), upon induction (top), and their resulting phenotype (bottom): b, 10 µg of tamoxifen results in morphologically normal hair follicles with no outgrowths; c, 200 µg of tamoxifen induces about 3–15 recombined cells out of the roughly 80 HFSCs in the resting hair follicles, results in outgrowths; d, 6 mg of tamoxifen (1 mg of tamoxifen per day for 6 days) results in the formation of tumours. e, Quantification of outgrowths that arise from increasing number of labelled cells (mean ± s.e.m., n = 285 hair follicles in four mice). Scale bars, 50 µm.
Extended Data Figure 2
Extended Data Figure 2. Developing mutant outgrowths are largely composed of a mixture of mutant and wild-type cells
a, Percentage distribution of new bud composition, with 71% of developing outgrowths made up of a mixture of mutant and wild-type cells (n = 112 new buds across three mice). b, Representative budding outgrowths in tamoxifen-treated (200 µg) K19creER;β-catnflox(ex3) mice. All epithelial nuclei are labelled with K14–H2BmCherry (red). Nuclei of mutant cells are labelled with TCF–H2BGFP (green). All scale bars, 50 µm.
Extended Data Figure 3
Extended Data Figure 3. Small population of mutant cells can generate a varying number of outgrowths that regress
a, Wide-field view of emerging outgrowths. b, Outgrowth frequency per hair follicle generated by a small population of mutant cells (n = 718 hair follicles across three mice). c, Diagram of outgrowth regression.
Extended Data Figure 4
Extended Data Figure 4. Mutant β-catenin cells differentiate and are subsequently lost from the tissue
a, Whole-mount β-catenin immunofluorescence staining (blue) showing increased nuclear β-catenin staining corresponds to the Cre and Wnt fluorescent reporters. b, β-Catenin whole-mounts. tdTom+ cells (red) along the core of the outgrowth express higher levels of β-catenin (green). c, Revisits of the same outgrowth, showing the progressive elimination of the mutant core (TCF–H2BGFP, green). Epithelial nuclei are labelled with K14–H2BmCherry (red). d, Reverse transcribed PCR quantification of mRNA expression of the floxed mutant β-cateninΔex3 allele, normalized to GAPDH, during the development and regression of outgrowths (mean ± s.e.m., n = 3 mice per time point, ****P < 0.0001). e, Whole-mount Ki67 staining. f, Mitotic events in tdTomato negative (tdTom−) and positive (tdTom+) cells, during different outgrowth stages (mean ± s.e.m., n = 150 outgrowths across three mice). g, Revisit of a single outgrowth showing the morphologically differentiated core (arrowheads, left) that is expelled from the tissue over time (right). h, Whole-mount staining of inner root sheath differentiation marker Gata3 (ref. 35) co-localizes with mutant core labelled with tdTomato (left) and TCFH2BGFP (right). Scale bars, 50 µm.
Extended Data Figure 5
Extended Data Figure 5. Apoptosis and cell extrusion occurs during both growth and regression of aberrancies
a, Quantifications of different cellular behaviours occurring during growth and regression (mean ± s.e.m.). b, Representative outgrowths during growth (top, lower left) and regression (lower right) showing nuclear fragmentation, a morphological hallmark of apoptosis, pseudocoloured pink. c, Examples of outgrowths during growth (top) and regression (bottom) showing epithelial cells extruded into the dermis (pseudocoloured green). d, Revisits of extruded cells in the dermis, showing cells become undetectable over time. e, Quantifications showing the amount of time cells are detectable in the dermis once extruded. All scale bars, 50 µm.
Extended Data Figure 6
Extended Data Figure 6. Outgrowths with limited contribution of wild-type cells develop into proliferative cysts
a, K19creER; B-catnflox(Ex3);tdTomato Wlsfl/fl whole-mount of Ki67 staining. b, K19creER;β-catnflox(Ex3);K14rtTA;cdkn1b whole-mount of Ki67 staining. c, Haematoxylin and eosin image of K19creER;β-catnflox(Ex3) backskin at low- (left) and high-power (right) magnifications. d, Haematoxylin and eosin image of backskin from K19creER;β-catnflox(Ex3);Wlsfl/fl showing the formation of small follicular cysts at low- (left) and high-power (right) magnification. e, Haematoxylin and eosin image of backskin from K19creER;β-catnflox(Ex3);K14rtTA;cdkn1b showing a dilated follicular cyst at low- (left) and high-power (right) magnifications. All scale bars, 50 µm.
Extended Data Figure 7
Extended Data Figure 7. Endogenous and ectopic cellular behaviours are used in the regression of the outgrowth
Diagram of observed cellular behaviours during the growth and regression of the outgrowths.
Extended Data Figure 8
Extended Data Figure 8. Remaining cells from regressed outgrowths can develop into functional appendages
a, Revisits of the same outgrowth in the non-cycling compartment. K14–H2BGFP labels epithelial nuclei (green). The non-hair cycling compartment is outlined in white. The cycling hair follicle, out of focal plane at later time points, is represented by a white dotted line. Insets show elimination of mutant core (tdTomato, red). b, Quantification of the phases of growth and regression of the outgrowths from the hair cycling epithelium of tamoxifen-treated (200 µg) Lgr6creER;β-catnflox(Ex3)/+;tdTomato;mTmG;K14–H2BGFP mice (mean ± s.e.m.). Pre-phenotypic plot accounts for hair follicles before the development of outgrowths to demonstrate variability in timing of formation (n = 33 outgrowths in three mice). c, Quantification of the two different fates of non-cycling epithelial outgrowths 173 days after induction (n = 24 outgrowths). d, Revisits from a tamoxifen-treated (200 µg) Lgr6creER;β-catnflox(Ex3)/+;tdTomato;mTmG;K14–H2BGFP mouse showing that, after the mutant core has differentiated out of the tissue (left), the remaining nodule of cells has converted into an extranumerary hair follicle. Subsequent revisits capture the representative extranumerary follicle in the early stage of catagen (centre) and telogen (right). e, Diagram representing the conversion into extranumerary structures. f, Revisits of representative Lgr6creER;β-catnflox(Ex3)/+;tdTomato;mTmG;K14–H2BGFP mouse after the administration of a high-dose (2 mg) of tamoxifen (n = 4 mice). Large benign tumours result in a macroscopic phenotype of severely wrinkled skin (day 21, left and inset) that resolves over time as the outgrowths regress (day 208, right and inset). g, Quantification of the hair follicles with tumours (mean ± s.e.m., n = 803 hair follicles across three mice). h, Quantification of hair follicles with supernumerary sebaceous gland lobes and de novo hair follicles after outgrowth regression (mean ± s.e.m., n = 580 hair follicles across three mice). i, Representative images of normal (left) and extranumerary (right) sebaceous glands present in the tissue 151 days after induction. Normal sebaceous glands are outlined in white, extranumerary structures are outlined in blue. j, Oil Red O staining on tail whole-mounts of Lgr6creER;β-catnflox(Ex3)/+ mouse 1 year after the administration of a high dose (2 mg) of tamoxifen. Note the Oil Red O staining in the extranumerary sebaceous gland lobes, indicating these structures are functional and actively producing lipids. k, Whole-mount Pcad IF staining on Lgr6creER;β-catnflox(Ex3)/+ ear skin 1 year after the administration of a high dose (2 mg) of tamoxifen, showing the extranumerary hair follicle has a Pcad enrichment at the base of the structure, similar to the hair germ of the main axis. All scale bars, 50 µm.
Extended Data Figure 9
Extended Data Figure 9. Broad Hras activation and mutation-independent systems reveal a wide range of corrective abilities
a, Representative phenotypes in LV-cre HrasG12V broad activation model (59% are hyperthickened and 10% are structurally deformed). b, Quantification of follicular bulge width demonstrates that Hras follicles are significantly hyperthickened during the first rest (P < 0.0001) and growth phases (P < 0.0001), but that hyperthickening is resolved by the second rest phase (P = 0.0860, n = 492 wild-type follicles, 550 Hras follicles, three mice per genotype). c, Haematoxylin and eosin images demonstrating regions of epidermal hyperplasia (upper arrowhead) and dermal cysts (lower arrowhead). d, Keratinized cyst resolves over time. e, Top: x–y view of macroscopic growth in Hras mouse demonstrates that hyperthickening and hyperkeratinization are resolved in 6 weeks. Hair follicles (A–C) are used as landmarks. Bottom: y–z view of macroscopic growth at starting (p38) and ending (p80) time points (n = 3 regressing macroscopic growths observed in two mice).
Extended Data Figure 10
Extended Data Figure 10. Non-mutational insult to wild-type follicles can be corrected
a, Correction of a deformed hair follicle post-ablation (right) compared with its non-ablated counterpart (left). b, Frequency of ablated follicle recovery (mean ± s.e.m., n = 38 follicles in five mice) c, Representative image of an ablated follicle that adopts a deformed structure, corrects it, and begins to cycle again. d, Quantification of the cycling ability of follicles 4 weeks post-ablation revealed 24% of ablated follicles were able to cycle again (mean ± s.e.m., n = 38 follicles in five mice). Asterisks denote region of autofluorescence produced post-ablation. All scale bars, 50 µm.
Figure 1
Figure 1. β-Catenin mutant cells generate aberrant growths that regress
a, b, K14–H2BGFP labels epithelial nuclei (green). β-catenin-induced growths (200 µg tamoxifen) are pseudocoloured purple. a, Representative outgrowths in two hair follicles (1, 2) irreversibly regress over 4 weeks. b, Quantification of outgrowth regression over time (mean ± s.e.m.). Eighty-five per cent of outgrowths regress during the first 4 weeks, while the rest are eliminated within the next month (n = 286 outgrowths across three mice). c, Quantification of the growth and regression phases (mean ± s.e.m.). Pre-phenotypic plot accounts for quantified hair follicles before outgrowth development (n = 286 across three mice).
Figure 2
Figure 2. Wild-type cells are required for the active elimination of mutant cells
a, Revisits of the same β-catenin outgrowth. b, c, β-Catenin/Wls outgrowth phenotypes (mean ± s.e.m., number of outgrowths = 168 Wls+/+ in three mice; n = 66 Wlsfl/+ in two mice; n = 233 Wlsfl/fl in four mice). d, Schematic of β-catenin and CDKN1b induction (top) and outgrowth revisits (bottom). e, Quantification of growth and regression (mean ± s.e.m., n = 70 outgrowths in four β-catenin/CDKN1b mice and n = 126 in four β-catenin mice). f, Dilated follicular cyst with stratified squamous epithelial lining (black arrow head) displaying keratinohyalin granules and dyskeratotic keratinocytes from β-catenin/CDKN1b. Scale bars, 50 µm.
Figure 3
Figure 3. Fate of regressing tissue is niche-dependent
a, Quantification of growth and regression phases of outgrowths (mean ± s.e.m.). Pre-phenotypic plot (n = 73 outgrowths across three mice) accounts for hair follicles before outgrowth development. b, Outgrowth remaining cells are converted into extranumerary sebaceous gland lobe (left) or hair follicle (right). Extranumerary structures are outlined in blue. c, Quantification of the different fates of non-hair cycling epithelial outgrowths (mean ± s.e.m., n = 73 outgrowths). d, Representative images of distinct epidermal regions over time, after a high dose (2 mg) of tamoxifen (n = 4 mice). Insets show representative hair follicles and derived tumours at progressive time points. Scale bars, 50 µm.
Figure 4
Figure 4. Tissue correction occurs in mutant Hras model
K14–H2BGFP labels epithelial nuclei (green). a, Representative hyperthickened K19creER;HrasG12V hair follicle. b, Quantification of HFSC compartment width during hair follicle growth and rest phases (mean ± s.e.m., number of hair follicles 267 wild type, 214 Hras, 3 mice per genotype, ****P < 0.0001). c, Phenotypes in LV-cre HrasG12V broad activation model (59% hyperthickened, 31% normal, and 10% structurally deformed (n = 264 follicles, three mice)). d, Representative Hras hair follicle lacking the lower stem-cell compartment corrects over time. e, Frequency of deformity correction in hyperthickened (n = 189, three mice) and structurally deformed (n = 38, three mice) follicles during a single hair cycle (mean ± s.e.m.).
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Ling G, et al. Persistent p53 mutations in single cells from normal human skin. Am. J. Pathol. 2001;159:1247–1253. - PMC - PubMed
    1. Martincorena I, et al. High burden and pervasive positive selection of somatic mutations in normal human skin. Science. 2015;348:880–886. - PMC - PubMed
    1. Laurie CC, et al. Detectable clonal mosaicism from birth to old age and its relationship to cancer. Nat. Genet. 2012;44:642–650. - PMC - PubMed
    1. Pineda CM, et al. Intravital imaging of hair follicle regeneration in the mouse. Nat. Protocols. 2015;10:1116–1130. - PMC - PubMed
    1. Rompolas P, et al. Live imaging of stem cell and progeny behaviour in physiological hair-follicle regeneration. Nature. 2012;487:496–499. - PMC - PubMed

Publication types