Although metastasis is the most lethal consequence of tumor progression, comparatively little is known regarding the molecular machinery governing this process. In many carcinomas, there is a robust correlation between the expression of the transcription factor Snail and a poor prognosis, but the contribution of this protein to the metastatic process remains unresolved. Interestingly, the prolonged expression of Snail in epidermal keratinocytes is sufficient to recapitulate early features of metastasis. However, it does so without inducing a complete epithelial-mesenchymal transition (EMT), a developmental phenomenon mediated by Snail that is extensively invoked as the mechanism fueling tumorigenesis. Instead, we found that the local invasiveness of keratinocytes is the consequence of the recruitment and activity of macrophages. Moreover, keratinocyte proliferation is the product of an IL-17/IL-6/Stat3 signaling module initiated by activated resident γδT cells in the transgenic skin. Together, these phenotypes prime the transgenic skin for the formation and metastasis of tumors in response to chemically induced carcinogenesis. Thus, the contribution of Snail to the progression of carcinomas is largely through the creation of a hyperproliferative and inflammatory niche that facilitates tumor development and dissemination. Cancer Res; 70(24); 10080–9. ©2010 AACR.

A decisive factor in the multistage process of metastasis is the early step of local invasion of carcinoma cells (1). However, the mechanisms coordinating the increased motility of proliferating cancer cells remain elusive. Members of the Snail family of transcription factors have garnered widespread interest in this context, as they are expressed in a variety of carcinomas (2, 3) and are associated with recurring or metastasizing tumors (4). Snail proteins have a well-established role in embryogenesis during which they mediate a process known as an epithelial-mesenchymal transition (EMT) to facilitate tissue formation (5). The Snail-mediated EMT causes cells to lose their epithelial characteristics such as E-cadherin–mediated adhesion and polarity while adopting phenotypes of mesenchymal cells such as an increased migratory capacity (6). Given the similarities to its effects in development, it is widely extrapolated that Snail functions similarly in the metastasis of somatic cells during tumorigenesis (7). The ability of Snail to induce an EMT in multiple cancer cell lines supports this notion, but the absence of in vivo evidence that this likewise occurs in cancerous cells of the body (8) suggests that additional mechanisms are operational in this disease.

We previously found that Snail is expressed during budding morphogenesis of the hair follicle, wherein proliferating keratinocytes in the epidermis invade into the underlying dermal compartment of the skin (9). Moreover, transgenic expression of Snail in the epidermis of young mice leads to features commonly seen in cutaneous cancers (10). This includes the activation of the Ras-MAPK signaling axis, involution of the epidermis, reduction in intercellular adhesion, and degradation of the basement membrane separating the epidermis from the underlying dermis (9). These attributes suggest that the Snail transgenic mouse may be a prime system with which to elucidate the regulatory network specifically controlling the critical initial stage of the metastatic cascade (1, 11).

Generation of Snail transgenic mice

Mice engineered to express the Snail transgene in the epidermis was previously described (9). All animal work was approved and adhered to the guidelines of IACUC.

Transwell cell migration and invasion assays

For the cell migration assay, 5 × 104 Raw246.7 cells (ATCC) were seeded in Transwell inserts with an 8.0-μm pore (Corning) with DMEM + 10% heat-inactivated FBS. CM from the epidermis of P7 mice was added to the bottom chamber at a 1:2 dilution in medium with or without 1.6 μg/mL of M-CSF neutralizing antibody (R&D Systems) or goat IgG control antibody. After 8-hour incubation, cells were stained with 0.1% Crystal Violet. The membrane inserts were removed and mounted on a slide. For the keratinocyte invasion assay, the Transwell inserts were coated with Collagen I (Sigma) and 1 × 105 HaCat cells (ATCC) were used. CM from thioglycollate-elicited peritoneal macrophages stimulated with (TAM CM) or without (Mac CM) 20 ng/mL of IL-4 (eBioscience) was collected and added to the bottom chamber and incubated for 14 hours.

Proliferation assays

After overnight starvation, primary mouse keratinocytes were trypsinized and resuspended in mouse keratinocytes medium with 2% FBS. A total of 1,500 cells were inoculated in 96-well dishes with or without 10 ng/mL of IL-6 (eBioscience), 25 ng/mL of IL-17 (R&D Systems), 2 μg/mL of IL-6 neutralizing antibody (R&D Systems), or 50 μmol/L of STAT3 inhibitor peptide (Calbiochem). Cell numbers were measured by Cell TiTer 96 Aqueous One Solution (Promega) as suggested by the manufacturer.

Dexamethasone and indomethacin treatment

Mice were injected subcutaneously with a mixture of 0.05 mg/kg of dexamethasone (Sigma-Aldrich) and 0.05 mg/kg of indomethacin (Fluka) starting from newborn mice for 5 consecutive days. Control mice were injected with vehicle (0.4% ethanol in PBS).

Histology, in situ hybridization, and immunohistochemistry

Mouse skin or lymph nodes from WT and Snail transgenic animals were either frozen in OCT (Tissue-Tek) or embedded in paraffin depending on the application. Paraffin sections were prepared for histology and counterstained with hematoxylin and eosin-Y (H&E). Antibodies used were anti-phospho-Akt, c-Jun, CD31, and phospho-STAT3 (Tyr 705) all from Cell Signaling, CD44, fibronectin, phospho-NFκB (Ser276; Cell Signaling), CD3 (BD Biosciences), MAC-1 (BD Biosciences), and CD206. For nuclear staining, Hoechst 33342 (Calbiochem) was added in a final concentration of 1 mg/mL to the secondary antibody dilution. Immunofluorescence (IF) was detected using rhodamine-X or FITC-conjugated secondary antibodies (Jackson Immunoresearch) or expression was developed using the Vectastain ABC kit (Vector Labs) according to the manufacturer's instructions. Images were acquired on an Olympus Bx51 microscope with an Olympus DP70 camera. A 40× 1.3 UPlan FL N objective (Olympus) was used for acquisition.

Quantitative real-time PCR

Total RNA was extracted from whole skin of WT (n = 5) and Snail transgenic (Tg; n = 5) mice at postnatal day 7 (P7) using Trizol reagent (Invitrogen) according to manufacturer's instructions. Similarly, epidermis from P7 WT (n = 5) and Tg (n = 5) mice was isolated with dispase treatment and total RNA was isolated using the Trizol protocol. cDNA was synthesized by reverse transcription using oligo-dT as primers (Superscript III kit; Invitrogen). Real-time PCR analysis was carried out with previously described primers (12). Experiments were carried out in triplicate from cDNA isolated from 5 different animals.

Zymography

Presence of active MMP-9 and MMP-2 was detected using a previously described protocol (13).

Two-stage chemically induced skin carcinogenesis

Seven- to 8-week-old mice were subjected to the 2-stage skin chemical carcinogenesis protocol as previously described (14), using 400 nmol of DMBA as the initiating agent and 10 nmol of TPA as the promoting agent.

The extensive attention paid to the role of Snail in tumorigenesis is partly due to its expression in numerous carcinomas (3) including those of the skin (Supplementary Fig. S1). Interestingly, we found that a transgenic mouse engineered to overexpress Snail in epidermal keratinocytes shares some features with these carcinomas including an elevated proliferative index and local invasiveness (9). We, therefore, investigated the extent to which the Snail transgenic skin recapitulates the biochemical features of the metastatic program. We found that several factors known to promote tumor dissemination are upregulated in the Snail transgenic skin of neonatal mice (Fig. 1A). Among these are the activated Akt kinase, which is necessary for many events of the metastatic pathway (15), and c-Jun/AP-1, which has been linked to invasive properties of aggressive breast cancer cells (16).

Figure 1.

Presence of metastasis-associated markers in the Snail transgenic skin. A, WT (left) and Snail transgenic (Snail Tg; right)skin sections were subjected to IF with antibodies recognizing keratin 5 (K5) in red and phospho-Akt, c-Jun, and the blood vessel marker CD31 in green. Dotted lines denote the basement membrane, which separates the epidermis (epi) and hair follicle(hf) from the dermis (der). B, reverse transcription PCR of VEGF (left) from RNA extracted from 2 WT and 3 transgenic skin samples, with GAPDH as a loading control, and zymography of MMP-9 and MMP-2 activity (right column). C, immunohistochemistry of phosphorylated Stat3 (pStat3). D, IF of CD44 expression. Bars, 30 μm.

Figure 1.

Presence of metastasis-associated markers in the Snail transgenic skin. A, WT (left) and Snail transgenic (Snail Tg; right)skin sections were subjected to IF with antibodies recognizing keratin 5 (K5) in red and phospho-Akt, c-Jun, and the blood vessel marker CD31 in green. Dotted lines denote the basement membrane, which separates the epidermis (epi) and hair follicle(hf) from the dermis (der). B, reverse transcription PCR of VEGF (left) from RNA extracted from 2 WT and 3 transgenic skin samples, with GAPDH as a loading control, and zymography of MMP-9 and MMP-2 activity (right column). C, immunohistochemistry of phosphorylated Stat3 (pStat3). D, IF of CD44 expression. Bars, 30 μm.

Close modal

The growth of tumors within somatic tissues imposes an increased metabolic burden that is met by an increased supply of nutrients delivered by the targeted growth of new blood vessels to the cancer cells. Moreover, these vessels provide a route for metastatic dissemination by providing a site for entry into the circulation (17). The level of CD31 staining that denotes endothelial vessels is markedly increased in the transgenic dermis, indicating augmented vascularization (Fig. 1A). This phenotype is assisted by an increase in the level of the proangiogenic protein, vascular endothelial growth factor (VEGF), and matrix metalloproteinase-9 (MMP-9), which facilitates remodeling of the local extracellular matrix and directional growth of the blood vessel (Fig. 1B; ref. 18). This restriction of MMP-9 to the dermal compartment of the skin is unexpected, as it has previously been shown that introduction of Snail into an epithelial cell line is sufficient to induce expression of MMP-9 (19). In addition, MMP-2, which is also known to facilitate angiogenesis, has relatively low activity in both the wild-type (WT) and transgenic dermis (Fig. 1B).

Increasing evidence shows that the transcription factor Stat3 plays an indispensable role in various aspects of oncogenesis including keratinocyte proliferation and tumor angiogenesis, invasion, and metastasis (20). Consonant with its varied roles in tumor progression, Stat3 (Fig. 1C) and its transcriptional target c-myc (ref. 21; Supplementary Fig. S2) are activated in the transgenic skin. Moreover, c-myc has recently been shown to induce a transcriptional program leading to the propagation of cancer stem cells (22). In line with this hypothesis, keratinocytes expressing the marker keratin 5 coexpress the epithelial stem cell marker CD44 (23) in the Snail transgenic epidermis (Fig. 1D), suggesting an expansion of this pool of cells.

Given the ability of Snail to singularly induce multiple features of the early stages of metastasis, we focused our investigation on elucidating the mechanisms by which it promotes 2 phenomena of this complex process—keratinocyte proliferation and local invasion. Keratinocytes in the basal layer of the epidermis are cuboidal, whereas suprabasal cells are flattened. In contrast, keratinocytes in the transgenic epidermis display an elongated and spindle-like morphology (Figs. 1A and 2D), which implies an enhanced migratory capacity. Together with our previous finding that transgenic epidermis exhibited lower levels of E-cadherin in regions expressing Snail (9), these results support the notion that Snail induces an EMT for tumor progression (7). However, despite the continual expression of Snail in the transgenic epidermis in neonatal mice, keratin expression is maintained and there is no induction of the mesenchymal marker fibronectin in the epidermal cells (Fig. 2A). This is particularly surprising, as the constitutive expression of exogenous Snail in epithelial cell lines is sufficient to induce mesenchymal marker expression (6). Moreover, previous reports have also suggested that cells stably transfected with Snail have decreased cell proliferation (24) whereas epidermal keratinocytes expressing Snail were found to be proliferative (9). These results suggest that a complete or permanent EMT, which occurs during development, is not required for the early stages of metastasis of cancer cells derived in vivo.

Figure 2.

Inflammation in the Snail transgenic skin. A, expression of the mesenchymal marker fibronectin (FN), activated NFκB (pNFκB), and the pan T-cell marker CD3 in WT (left) and Snail transgenic (Snail Tg, right) skin. B, profile of the Th1 cytokine IFNγ and Th2 cytokines IL-4 and IL-13 via reverse transcription PCR with GAPDH as a loading control. C, infiltration of macrophages detected by Mac-1 staining in the dermis of WT and Snail Tg skin (top) and the presence of M2/TAMs marked by a subset of macrophages (Mac-1; green) coexpressing the mannose receptor recognized by the CD206 antibody in red. D, histologic analysis of the effect of PBS vehicle control or dexamethasone–indomethacin (DI) immunosuppressive cocktail on the transgenic phenotype. Bars, 30 μm.

Figure 2.

Inflammation in the Snail transgenic skin. A, expression of the mesenchymal marker fibronectin (FN), activated NFκB (pNFκB), and the pan T-cell marker CD3 in WT (left) and Snail transgenic (Snail Tg, right) skin. B, profile of the Th1 cytokine IFNγ and Th2 cytokines IL-4 and IL-13 via reverse transcription PCR with GAPDH as a loading control. C, infiltration of macrophages detected by Mac-1 staining in the dermis of WT and Snail Tg skin (top) and the presence of M2/TAMs marked by a subset of macrophages (Mac-1; green) coexpressing the mannose receptor recognized by the CD206 antibody in red. D, histologic analysis of the effect of PBS vehicle control or dexamethasone–indomethacin (DI) immunosuppressive cocktail on the transgenic phenotype. Bars, 30 μm.

Close modal

Accumulating data suggest that secondary inflammatory responses may instead be responsible for the invasiveness of cancer cells (25). The elevated levels of nuclear NFκB (which is a classic regulator of proinflammatory genes) in the upper layers of the epidermis as well as the dermis (Fig. 2A) suggest that inflammation may contribute to the invasive phenotype of the transgenic skin. Furthermore, Stat3 (Fig. 1C) and CD44 (Fig. 1D) staining in the transgenic dermis are consistent with their expression in infiltrating immune cells. The transgenic dermis indeed has elevated levels of T cells, which is illustrated by the pan T-cell marker CD3 (Fig. 2A) and include CD4+, CD8+, and γδT cells (Supplementary Fig. S3A). The cytokines secreted by these T cells are skewed toward the TH2 factors (IL-4, 13), which promotes an inflammatory microenvironment (26) at the expense of IFNγ, a TH1 cytokine and tumor antagonist (Fig. 2B). Cells of the innate immune system are also recruited into the transgenic skin, including macrophages (Fig. 2C) and granulocytes (Supplementary Fig. S3A). One level of cross talk between the adaptive and innate immune systems is illustrated by the ability of TH2-derived cytokines to polarize macrophages along the M2 lineage into tumor-associated macrophages (TAM). Evidence of this is seen in the Snail transgenic skin by the coexpression of the lectin CD206 on a subset of macrophages (Fig. 2C) and the presence of other markers of TAMs (Supplementary Fig. S3B and C). Interestingly, it has been reported that TAMs are a source of MMP-9, which is found exclusively in the transgenic dermis (Fig. 1B), and this enzyme contributes to their role in the metastatic cascade (27). The importance of these various immune cells in manifesting the changes in the Snail transgenic skin was shown by the ability of an immunosuppressive cocktail to significantly reduce the epidermal involution and hyperplasia (Fig. 2D) and cutaneous inflammation (Supplementary Fig. S4) in the mutant mouse.

Among these infiltrating immune cells, macrophages have garnered extensive attention for their remarkable ability to promote tumor proliferation and metastasis (28). Thus, the mechanism by which Snail expressing keratinocytes induces their recruitment into the skin becomes an important problem to resolve. We observed that the transgenic epidermis had an elevated level of monocyte colony-stimulating factor-1 (CSF-1), which is a well-known chemoattractant for macrophages (Fig. 3A; ref. 29). The induction of CSF-1 seems to be cell autonomous, as transfection of Snail into primary keratinocytes is sufficient to elicit CSF-1 expression (Fig. 3A). Importantly, we found, using a Transwell assay, that conditioned medium (CM) from epidermal explants of transgenic mice is capable of recruiting macrophages (Fig. 3B). An inhibitory antibody against this cytokine shows that CSF-1 is a required component of the CM to stimulate macrophage mobilization. Because the epidermal explants used to condition the media are a heterogeneous population of cells, we tested whether Snail expression in keratinocytes is directly involved in the recruitment of macrophages by reconstituting this process completely in vitro. Primary keratinocytes transfected with Snail are capable of synthesizing and secreting CSF-1 to promote macrophage recruitment (Fig. 3C). These macrophages can, in turn, potently stimulate invasion of primary keratinocytes through an extracellular matrix (Fig. 3D). Moreover, the cytokine milieu present in the transgenic skin favors the polarization of the macrophages into TAMs (Fig. 2B and C; Supplementary Fig. 3B and C), which have been localized to areas of metastasis and shown to promote tumor cell invasion (30, 31). Consistent with this scenario, we found that TAMs can stimulate keratinocyte invasion at even higher levels than the classically activated macrophages (Fig. 3D).

Figure 3.

Recruitment and function of macrophages in the transgenic skin. A, analysis of the expression of CSF-1 and Snail in WT epidermis (WT epi) or Snail transgenic epidermis (Tg epi) and primary mouse keratinocytes transfected with either empty vector (mKT + vector) or Snail (mKT + Snail) via reverse transcription PCR using GAPDH as a loading control. B, migration of macrophages in a Transwell assay when treated with conditioned media from WT epidermis (WT epi CM) or Snail transgenic epidermis (Tg epi CM) and either a control immunoglobulin (Con IgG) or a CSF-1 inhibitory antibody (CSF-1 Inh. Ab). Results are expressed as a fold increase over macrophages treated with normal growth media. C, effect of CM from mouse keratinocytes (mKT) transfected with vector or Snail on macrophage migration in the presence of control or CSF-1 inhibitory antibody. D, quantification of keratinocyte invasion through extracellular matrix in a Transwell assay. Primary keratinocytes were exposed to growth media or CM from macrophages or tumor associated macrophages (Mac CM or TAM CM, respectively).

Figure 3.

Recruitment and function of macrophages in the transgenic skin. A, analysis of the expression of CSF-1 and Snail in WT epidermis (WT epi) or Snail transgenic epidermis (Tg epi) and primary mouse keratinocytes transfected with either empty vector (mKT + vector) or Snail (mKT + Snail) via reverse transcription PCR using GAPDH as a loading control. B, migration of macrophages in a Transwell assay when treated with conditioned media from WT epidermis (WT epi CM) or Snail transgenic epidermis (Tg epi CM) and either a control immunoglobulin (Con IgG) or a CSF-1 inhibitory antibody (CSF-1 Inh. Ab). Results are expressed as a fold increase over macrophages treated with normal growth media. C, effect of CM from mouse keratinocytes (mKT) transfected with vector or Snail on macrophage migration in the presence of control or CSF-1 inhibitory antibody. D, quantification of keratinocyte invasion through extracellular matrix in a Transwell assay. Primary keratinocytes were exposed to growth media or CM from macrophages or tumor associated macrophages (Mac CM or TAM CM, respectively).

Close modal

Given TAMs ability to stimulate proliferation of breast carcinoma cells (32), we hypothesized that they may also be responsible for the elevated keratinocyte proliferation found in the Snail transgenic skin (Supplementary Fig. S5; ref. 9). Surprisingly, we found that both M1 macrophages and M2/TAMs have no effect on the growth rate of keratinocytes (Supplementary Fig. S6). To decipher the mechanism of increased keratinocyte proliferation in the transgenic skin, we focused on the activation of Stat3 (Fig. 1C), which is a prerequisite for keratinocyte proliferation during carcinogenesis (20). Profiling of the cytokines released from the transgenic epidermis revealed an increased level in the amount of secreted IL-6, which can stimulate the phosphorylation and nuclear translocation of Stat3 (Fig. 4A). Moreover, IL-6 can increase the rate of keratinocyte growth in a Stat3-dependent fashion (Fig. 4A). IL-6 is not normally expressed in the WT epidermis, and transfection of Snail into keratinocytes is incapable of inducing its expression (Fig. 4A). However, it is found in inflammatory skin diseases such as psoriasis (33), suggesting that the stimulus may be derived from an activated leukocyte. A clue into this mechanism came from reports that IL-17 can induce IL-6 expression in both autoimmune (34) and tumor settings (35). While profiling cytokine expression, we observed that IL-17 was induced in the transgenic skin and that the activated γδT cells in the transgenic dermis are a source of this cytokine (Fig. 4B). The analysis of the biological activity of this cytokine on epidermal keratinocytes revealed that IL-17 can promote keratinocyte proliferation via IL-6 signaling (Fig 4B), and this cascade is competent to activate Stat3 (Fig 4C). As noted earlier, c-myc is a target of Stat3 and the increased expression of this proto-oncogene upon treatment of keratinocytes with IL-17 verified that the IL-17/IL-6/Stat3 signaling cascade was operational (Fig. 4C).

Figure 4.

Leukocyte-derived IL-17 stimulates IL-6 production in keratinocytes to promote proliferation. A, quantification of IL-6 secreted from WT or Snail transgenic (Snail Tg) epidermis (column 1) and nuclear translocation of phosphorylated Stat3 (pStat3) in keratinocytes treated with buffer control or recombinant IL-6 (column 2). Proliferation rate of keratinocytes treated with growth media (blue), recombinant IL-6 (orange), or IL-6 + Stat3 inhibitor (green) using the XTT assay (column 3). IL-6 expression (column 4) detected via reverse transcription PCR of RNA extracted from WT epidermis (WT epi) or transgenic epidermis (Tg epi) and keratinocytes transfected with empty vector (mKT + vector) or Snail (mKT + Snail). B, coexpression of γδT-cell marker (green) and IL-17 (red) in column 1. Right, enlarged image of the area marked by the arrowhead in the left panel. Column 2 displays the proliferation rate of keratinocytes treated with media (blue), IL-17 (orange), or IL-17 + IL-6 inhibitory antibody (green). C, nuclear translocation of phosphorylated Stat3 in keratinocytes treated with buffer, IL-17 + control antibody, or IL-17 + IL-6 inhibitory antibody (left). Right, quantitative real-time PCR of c-myc expression in WT or transgenic epidermis or keratinocytes treated with buffer or IL-17 and/or inhibitors. Note: standard deviation for each triplicate data point in the proliferation assays was less than 3%. Bars, 30 μm.

Figure 4.

Leukocyte-derived IL-17 stimulates IL-6 production in keratinocytes to promote proliferation. A, quantification of IL-6 secreted from WT or Snail transgenic (Snail Tg) epidermis (column 1) and nuclear translocation of phosphorylated Stat3 (pStat3) in keratinocytes treated with buffer control or recombinant IL-6 (column 2). Proliferation rate of keratinocytes treated with growth media (blue), recombinant IL-6 (orange), or IL-6 + Stat3 inhibitor (green) using the XTT assay (column 3). IL-6 expression (column 4) detected via reverse transcription PCR of RNA extracted from WT epidermis (WT epi) or transgenic epidermis (Tg epi) and keratinocytes transfected with empty vector (mKT + vector) or Snail (mKT + Snail). B, coexpression of γδT-cell marker (green) and IL-17 (red) in column 1. Right, enlarged image of the area marked by the arrowhead in the left panel. Column 2 displays the proliferation rate of keratinocytes treated with media (blue), IL-17 (orange), or IL-17 + IL-6 inhibitory antibody (green). C, nuclear translocation of phosphorylated Stat3 in keratinocytes treated with buffer, IL-17 + control antibody, or IL-17 + IL-6 inhibitory antibody (left). Right, quantitative real-time PCR of c-myc expression in WT or transgenic epidermis or keratinocytes treated with buffer or IL-17 and/or inhibitors. Note: standard deviation for each triplicate data point in the proliferation assays was less than 3%. Bars, 30 μm.

Close modal

In light of the effect of Snail expression in the skin, we investigated whether this transgene renders mice more susceptible to inflammation-driven skin cancer. To test this hypothesis, we subjected both WT and transgenic mice to the 2-stage chemical carcinogenesis protocol using 7,12-dimethylbenz(a)anthracene (DMBA) as the mutagen and 12-O-tetradecanoylphorbol-13 acetate (TPA) as the promoter (14). This protocol was facilitated by the fact that the neonatal phenotypes decrease as the mice reach adulthood (Fig. 5A) and Snail protein levels diminish (data not shown), thus allowing us to test whether Snail primes the epithelial cells of the skin for tumor formation. Transgenic mice have a higher frequency and incidence of tumor formation than WT littermate controls (Table 1; Supplementary Fig. S7). The transgenic skin responded with a significant epidermal hyperplasia and involution of the tissue reminiscent of migrating cells (Fig. 5A). Moreover, all of the adult transgenic mice displayed a hyperplastic sebaceous gland, an appendage of the epidermis, and treatment with DMBA + TPA led to the development of sebaceous carcinomas. The dermis of the DMBA + TPA-treated transgenic mouse verified a substantial increase in the number of lobular acini relative to the WT skin as marked by Oil Red O staining (Fig. 5B). Histologic analysis shows that these sebaceous carcinomas invade the blood vessels (Fig. 5B), induce epidermal ulceration, and invasion into both the adipose and stromal tissues of the skin (Fig. 5C). These sebaceous carcinomas were indeed metastatic, as 15 of 16 transgenic mice (Table 1) had sebocytes in their lymph nodes (Fig. 5D) that were positive for keratin 5 expression (Fig. 5E).

Figure 5.

Tumor development and metastasis in the Snail transgenic mouse. Seven- to 8-week-old mice were treated with DMBA + TPA unless otherwise noted. A, histologic staining with H&E of WT and Snail transgenic (Tg) ± DMBA and TPA. B, Oil Red O staining to detect sebaceous glands in WT (left) and transgenic (middle) skin. Right, transgenic skin section stained with H&E–arrowheads point to blood vessel. C, H&E staining of Snail transgenic skin showing an epidermal ulcer (left), invasion into adipose tissue, and stromal tissue (asterisk). D, top, H&E staining of lymph nodes from WT (left) and transgenic (middle, 10×; right, 40×) mice treated with DMBA and TPA. Bottom, IF of keratin 5 (red) in lymph nodes. Bars, 30 μm.

Figure 5.

Tumor development and metastasis in the Snail transgenic mouse. Seven- to 8-week-old mice were treated with DMBA + TPA unless otherwise noted. A, histologic staining with H&E of WT and Snail transgenic (Tg) ± DMBA and TPA. B, Oil Red O staining to detect sebaceous glands in WT (left) and transgenic (middle) skin. Right, transgenic skin section stained with H&E–arrowheads point to blood vessel. C, H&E staining of Snail transgenic skin showing an epidermal ulcer (left), invasion into adipose tissue, and stromal tissue (asterisk). D, top, H&E staining of lymph nodes from WT (left) and transgenic (middle, 10×; right, 40×) mice treated with DMBA and TPA. Bottom, IF of keratin 5 (red) in lymph nodes. Bars, 30 μm.

Close modal
Table 1.

Incidence and frequency of tumorigenesis

Adult mouse + treatmentNo. of animalsIncidence of epidermal hyperplasia, %Incidence of sebaceous hyperplasia, %Incidence of sebaceous carcinoma, %No. of sebaceous carcinoma per mouseIncidence of lymph node metastasis, %
Snail Tg 10 100 
WT + DMBA/TPA 10 30 
Snail Tg + DMBA/TPA 16 100 100 100 20.56 ± 10.3 93.8 
Adult mouse + treatmentNo. of animalsIncidence of epidermal hyperplasia, %Incidence of sebaceous hyperplasia, %Incidence of sebaceous carcinoma, %No. of sebaceous carcinoma per mouseIncidence of lymph node metastasis, %
Snail Tg 10 100 
WT + DMBA/TPA 10 30 
Snail Tg + DMBA/TPA 16 100 100 100 20.56 ± 10.3 93.8 

NOTE: Quantitation of phenotypes in WT and transgenic mice with or without chemical carcinogenesis. Data are compiled from mice that are 8 weeks after the promotion phase of the 2-step chemical carcinogenesis protocol.

A model summarizing the epithelial-leukocytic cross talk stimulated by the expression of Snail in epidermal keratinocytes to promote an early metastatic phenotype is presented in Figure 6. In the aggregate, these findings highlight the non–cell autonomous role that Snail has in promoting oncogenesis in vivo. We found no evidence of a complete or permanent EMT but can attribute many of the phenotypes to the recruitment and activity of immune cells recruited to the skin of the Snail transgenic mouse. This is consonant with the lack of reports of definitive evidence of an EMT occurring in any carcinoma cells in vivo (8). The intermediate phenotype of the Snail transgenic mouse may therefore be more appropriately referred to as “EMT like” (36). Our data suggest that the cell autonomous function of Snail during carcinogenesis in vivo may be to maintain the undifferentiated state of a metastasizing cell (23, 37) as it disseminates to new tissues, thereby contributing to the maintenance of the “cancer stem cell” pool. On the other hand, the driving force for the local invasion of these cancer cells is the reciprocal interactions between the carcinoma cells and leukocytes. These findings concur with recent reports implicating Snail in mediating inflammation (38, 39). Interestingly, the epidermal hyperplasia and cutaneous inflammation that is prominent in neonatal mice significantly dissipate in the adult mice (Fig. 5). This is likely due to the inherent instability of Snail expression (6) and reduction in protein levels despite the continual transcription driven by the keratin-14 promoter in epidermal keratinocytes. This reversibility of the phenotype implies that continual intercellular signaling stimulated by Snail is required to preserve the changes we documented in the transgenic skin.

Figure 6.

Signaling in the Snail transgenic skin. Expression of Snail in epidermal keratinocytes leads to a program that maintains the undifferentiated state of the transgenic cells. Snail also transcriptionally upregulates CSF-1 to cause homing of macrophages (MΦ) into the skin. A subset of these macrophages is polarized along the M2/TAM lineage. Together, these macrophages then stimulate the local invasion of the keratinocytes into the underlying dermis. Snail expressing keratinocytes also leads to the wound-like activation of resident γδT cells that migrate into the dermis and begin secreting IL-17. Epidermal keratinocytes respond to IL-17 by inducing expression of the cytokine IL-6, which works in a paracrine fashion to activate the transcription factor Stat3. Stat3 activation contributes to tumorigenesis by augmenting proliferation, cell survival, and angiogenesis.

Figure 6.

Signaling in the Snail transgenic skin. Expression of Snail in epidermal keratinocytes leads to a program that maintains the undifferentiated state of the transgenic cells. Snail also transcriptionally upregulates CSF-1 to cause homing of macrophages (MΦ) into the skin. A subset of these macrophages is polarized along the M2/TAM lineage. Together, these macrophages then stimulate the local invasion of the keratinocytes into the underlying dermis. Snail expressing keratinocytes also leads to the wound-like activation of resident γδT cells that migrate into the dermis and begin secreting IL-17. Epidermal keratinocytes respond to IL-17 by inducing expression of the cytokine IL-6, which works in a paracrine fashion to activate the transcription factor Stat3. Stat3 activation contributes to tumorigenesis by augmenting proliferation, cell survival, and angiogenesis.

Close modal

At first glance, the activity of resident γδT cells in the Snail transgenic skin seems to contradict their anticancer capability (40). However, γδT cells are activated and play a critical role in wound healing (41), which shares many processes in common with tumorigenesis such as inflammation/cytokine signaling and cell proliferation. In fact, tumor promotion and development are often found at sites of wounds/tissue damage (42). The activation and function of γδT cells in the transgenic skin likely occur along the lines of a wound-healing program to elicit a protumor response from these cells. Another important subset of T cells in this metastatic cascade seems to be TH2 cells that contribute to the inflammatory microenvironment and polarizes the macrophages into M2/TAMs. These TH2 cells are probably recruited to the skin by the chemokines CCL18, 22, and 27, which are upregulated in the epidermis of the transgenic skin (data not shown).

We postulate that the sebaceous gland carcinoma induced via chemical carcinogenesis is likely the response to the fact that the alteration in the sebocytes is the strongest remaining phenotype in adult transgenic mice. Moreover, these hyperplastic cells seem to be positive for keratin 15 (data not shown), which is another marker for sebaceous neoplasms as well as the sebaceous carcinoma of Murre–Torre syndrome (43), which is a subtype of hereditary nonpolyposis colorectal cancer, and sebaceous differentiation (Fig. 5B and C) seems to be a strong phenotypic marker of this disease. Other genes associated with hair follicle morphogenesis that were expressed in the basal layer of the epidermis, such as a mutated Lef1, also generated sebaceous skin tumors (44). Altogether, a clearer picture is emerging regarding the mechanism by which genes associated with budding morphogenesis of the hair follicle can be usurped by carcinomas to serve similar, albeit unregulated, roles in tumor development and metastasis.

No potential conflicts of interest were disclosed.

We thank members of the laboratories of David Traver, Kees Murre, Steve Hedrick, Randy Johnson, and Wendy Havran for technical advice, Shih-Wei Chen, Hannah Lee, Na Zhang, Norihiko Takeda, Masataka Asagiri, and Edward Yang for technical assistance, and members of the Jamora laboratory for critical discussions and comments.

This work was supported by grants from the NIH (NIAMS grant number 5R01AR053185-03), American Skin Association, and the Dermatology Foundation. C. Jamora is supported by a Hellman Faculty Fellowship, Y. Nakamura was supported by a postdoctoral fellowship from the Japanese Society for the Promotion of Science, and P. Lee is supported by a predoctoral fellowship from the NIH.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Christofori
G
. 
New signals from the invasive front
.
Nature
2006
;
441
:
444
50
.
2.
Hoot
KE
,
Lighthall
J
,
Han
G
, et al
Keratinocyte-specific Smad2 ablation results in increased epithelial-mesenchymal transition during skin cancer formation and progression
.
J Clin Invest
2008
;
118
:
2722
32
.
3.
Becker
KF
,
Rosivatz
E
,
Blechschmidt
K
,
Kremmer
E
,
Sarbia
M
,
Hofler
H
. 
Analysis of the E-cadherin repressor Snail in primary human cancers
.
Cells Tissues Organs
2007
;
185
:
204
12
.
4.
Blechschmidt
K
,
Sassen
S
,
Schmalfeldt
B
,
Schuster
T
,
Hofler
H
,
Becker
KF
. 
The E-cadherin repressor Snail is associated with lower overall survival of ovarian cancer patients
.
Br J Cancer
2008
;
98
:
489
95
.
5.
Hay
ED
. 
An overview of epithelio-mesenchymal transformation
.
Acta Anat (Basel)
1995
;
154
:
8
20
.
6.
Nieto
MA
. 
The snail superfamily of zinc-finger transcription factors
.
Nat Rev Mol Cell Biol
2002
;
3
:
155
66
.
7.
Thiery
JP
. 
Epithelial-mesenchymal transitions in development and pathologies.
Curr Opin Cell Biol
2003
;
15
:
740
6
.
8.
Thompson
EW
,
Newgreen
DF
,
Tarin
D
. 
Carcinoma invasion and metastasis: a role for epithelial-mesenchymal transition?
Cancer Res
2005
;
65
:
5991
5
;
discussion 5
.
9.
Jamora
C
,
Lee
P
,
Kocieniewski
P
, et al
A signaling pathway involving TGF-beta2 and Snail in hair follicle morphogenesis
.
PLoS Biol
2005
;
3
:
e11
.
10.
Dajee
M
,
Lazarov
M
,
Zhang
JY
, et al
NF-kappaB blockade and oncogenic Ras trigger invasive human epidermal neoplasia
.
Nature
2003
;
421
:
639
43
.
11.
Gupta
GP
,
Massague
J
. 
Cancer metastasis: building a framework
.
Cell
2006
;
127
:
679
95
.
12.
Lee
P
,
Lee
DJ
,
Chan
C
,
Chen
SW
,
Ch'en
I
,
Jamora
C
. 
Dynamic expression of epidermal caspase 8 simulates a wound healing response
.
Nature
2009
;
458
:
519
23
.
13.
Troeberg
L
,
Nagase
H
. 
Zymography of metalloproteinases
.
Curr Protoc Protein Sci
2004
;
Chapter 21:Unit 21 15
.
14.
Abel
EL
,
Angel
JM
,
Kiguchi
K
,
DiGiovanni
J
. 
Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications
.
Nat Protoc
2009
;
4
:
1350
62
.
15.
Qiao
M
,
Sheng
S
,
Pardee
AB
. 
Metastasis and AKT activation
.
Cell Cycle
2008
;
7
:
2991
6
.
16.
Smith
LM
,
Wise
SC
,
Hendricks
DT
, et al
c-Jun overexpression in MCF-7 breast cancer cells produces a tumorigenic, invasive and hormone resistant phenotype
.
Oncogene
1999
;
18
:
6063
70
.
17.
Zetter
BR
. 
Angiogenesis and tumor metastasis
.
Annu Rev Med
1998
;
49
:
407
24
.
18.
Mira
E
,
Lacalle
RA
,
Buesa
JM
, et al
Secreted MMP9 promotes angiogenesis more efficiently than constitutive active MMP9 bound to the tumor cell surface
.
J Cell Sci
2004
;
117
:
1847
57
.
19.
Jorda
M
,
Olmeda
D
,
Vinyals
A
, et al
Upregulation of MMP-9 in MDCK epithelial cell line in response to expression of the Snail transcription factor
.
J Cell Sci
2005
;
118
:
3371
85
.
20.
Chan
KS
,
Sano
S
,
Kiguchi
K
, et al
Disruption of Stat3 reveals a critical role in both the initiation and the promotion stages of epithelial carcinogenesis
.
J Clin Invest
2004
;
114
:
720
8
.
21.
Knoepfler
PS
. 
Myc goes global: new tricks for an old oncogene
.
Cancer Res
2007
;
67
:
5061
3
.
22.
Wong
DJ
,
Liu
H
,
Ridky
TW
,
Cassarino
D
,
Segal
E
,
Chang
HY
. 
Module map of stem cell genes guides creation of epithelial cancer stem cells
.
Cell Stem Cell
2008
;
2
:
333
44
.
23.
Mani
SA
,
Guo
W
,
Liao
MJ
, et al
The epithelial-mesenchymal transition generates cells with properties of stem cells
.
Cell
2008
;
133
:
704
15
.
24.
Kudo-Saito
C
,
Shirako
H
,
Takeuchi
T
,
Kawakami
Y
. 
Cancer metastasis is accelerated through immunosuppression during Snail-induced EMT of cancer cells
.
Cancer Cell
2009
;
15
:
195
206
.
25.
Mantovani
A
. 
Cancer: inflammation by remote control
.
Nature
2005
;
435
:
752
3
.
26.
Ellyard
JI
,
Simson
L
,
Parish
CR
. 
Th2-mediated anti-tumour immunity: friend or foe?
Tissue Antigens
2007
;
70
:
1
11
.
27.
Hiratsuka
S
,
Nakamura
K
,
Iwai
S
, et al
MMP9 induction by vascular endothelial growth factor receptor-1 is involved in lung-specific metastasis
.
Cancer Cell
2002
;
2
:
289
300
.
28.
Pollard
JW
. 
Tumour-educated macrophages promote tumour progression and metastasis
.
Nat Rev Cancer
2004
;
4
:
71
8
.
29.
Sica
A
,
Allavena
P
,
Mantovani
A
. 
Cancer related inflammation: the macrophage connection
.
Cancer Lett
2008
;
267
:
204
15
.
30.
Lewis
CE
,
Pollard
JW
. 
Distinct role of macrophages in different tumor microenvironments
.
Cancer Res
2006
;
66
:
605
12
.
31.
Green
CE
,
Liu
T
,
Montel
V
, et al
Chemoattractant signaling between tumor cells and macrophages regulates cancer cell migration, metastasis and neovascularization
.
PLoS One
2009
;
4
:
e6713
.
32.
Condeelis
J
,
Pollard
JW
. 
Macrophages: obligate partners for tumor cell migration, invasion, and metastasis
.
Cell
2006
;
124
:
263
6
.
33.
Elder
JT
,
Klein
SB
,
Tavakkol
A
,
Fisher
GJ
,
Nickoloff
BJ
,
Voorhees
JJ
. 
Growth factor and proto-oncogene expression in psoriasis
.
J Invest Dermatol
1990
;
95
:
7S
9S
.
34.
Ogura
H
,
Murakami
M
,
Okuyama
Y
, et al
Interleukin-17 promotes autoimmunity by triggering a positive-feedback loop via interleukin-6 induction
.
Immunity
2008
;
29
:
628
36
.
35.
Wang
L
,
Yi
T
,
Kortylewski
M
,
Pardoll
DM
,
Zeng
D
,
Yu
H
. 
IL-17 can promote tumor growth through an IL-6-Stat3 signaling pathway
.
J Exp Med
2009
;
206
:
1457
64
.
36.
Klymkowsky
MW
,
Savagner
P
. 
Epithelial-mesenchymal transition: a cancer researcher's conceptual friend and foe
.
Am J Pathol
2009
;
174
:
1588
93
.
37.
Mikkelsen
TS
,
Hanna
J
,
Zhang
X
, et al
Dissecting direct reprogramming through integrative genomic analysis
.
Nature
2008
;
454
:
49
55
.
38.
Wu
K
,
Bonavida
B
. 
The activated NF-kappaB-Snail-RKIP circuitry in cancer regulates both the metastatic cascade and resistance to apoptosis by cytotoxic drugs
.
Crit Rev Immunol
2009
;
29
:
241
54
.
39.
Wu
Y
,
Deng
J
,
Rychahou
PG
,
Qiu
S
,
Evers
BM
,
Zhou
BP
. 
Stabilization of snail by NF-kappaB is required for inflammation-induced cell migration and invasion
.
Cancer Cell
2009
;
15
:
416
28
.
40.
Girardi
M
,
Oppenheim
DE
,
Steele
CR
, et al
Regulation of cutaneous malignancy by gammadelta T cells
.
Science
2001
;
294
:
605
9
.
41.
Jameson
J
,
Ugarte
K
,
Chen
N
, et al
A role for skin gammadelta T cells in wound repair
.
Science
2002
;
296
:
747
9
.
42.
Schafer
M
,
Werner
S
. 
Cancer as an overhealing wound: an old hypothesis revisited
.
Nat Rev Mol Cell Biol
2008
;
9
:
628
38
.
43.
Misago
N
,
Narisawa
Y
. 
Cytokeratin 15 expression in neoplasms with sebaceous differentiation
.
J Cutan Pathol
2006
;
33
:
634
41
.
44.
Niemann
C
,
Owens
DM
,
Hulsken
J
,
Birchmeier
W
,
Watt
FM
. 
Expression of DeltaNLef1 in mouse epidermis results in differentiation of hair follicles into squamous epidermal cysts and formation of skin tumours
.
Development
2002
;
129
:
95
109
.