B7-H1 (CD274) is a T-cell coinhibitory molecule that is also often induced on human carcinoma cells, where its expression has been implicated in immune escape. Under inflammatory conditions, B7-H1 is also inducible in normal epithelial cells but little is known about its involvement in conversion of normal cells to tumor cells. Here, we show that skin-specific expression of B7-H1 accelerates inflammatory carcinogenesis in a methylcholantrene (MCA)-induced model of squamous cell carcinoma (SCC). Inflammatory responses induced by MCA or phorbol ester TPA were clearly inhibited in B7-H1 transgenic mice (B7-H1tg mice). Antibody-mediated blockade of either B7-H1 or the related molecule PD-1 revealed that their ability to limit inflammation relied on ligand interactions made by B7-H1 or PD-1. Skin keratinocytes derived from B7-H1tg mice exhibited constitutive reduction of E-cadherin, and SCC induced in B7-H1tg mice also showed loss of E-cadherin along with elevated expression of the transcription factors Slug and Twist that drive epithelial–mesenchymal transition (EMT). Our results indicate that upregulation of B7-H1 in skin epithelial cells promotes EMT and accelerates carcinogenesis, revealing insights into the significance of B7-H1 overexpression on solid tumor cells and hinting at a close relationship between EMT and immune escape signaling pathways in cancer. Cancer Res; 71(4); 1235–43. ©2010 AACR.

B7-H1 (CD274) is one of the ligands for the coinhibitory receptor PD-1 (CD279), and the B7-H1:PD-1 pathway is involved in the induction and maintenance of peripheral tolerance (1, 2). B7-H1 is widely distributed on leukocytes and nonhematopoietic cells in lymphoid and nonlymphoid tissues. Nonlymphoid tissue-associated B7-H1 is found on pancreatic islets (3), keratinocytes (KCs; ref. 4), smooth muscle cells (5), and placenta (6) at inflammatory disease sites. IFN-γ is a key cytokine in the induction of B7-H1. In addition to the expression of B7-H1 in normal tissue cells, constitutive expression of B7-H1 is found on various human cancers, including squamous cell carcinomas (SCCs) of the lung, esophagus, and head and neck; other types of carcinomas of the colon, ovaries, bladder, and breast; melanoma; and glioma (1, 7–17). Tumor-associated B7-H1 is closely correlated with poor prognosis and/or higher malignancy grade. In mouse tumor transplants, B7-H1-transduced tumors show more aggressive tumor growth and poor survival rates (7, 9, 12). Ligation of PD-1 by tumor-associated B7-H1 induces apoptosis or downregulation of effector CTL, resulting in an escape from T-cell-mediated immune surveillance. Blockade of the B7-H1:PD-1 pathway efficiently reduces tumor growth and improves survival (7, 9, 12, 18, 19). Thus, tumor-associated B7-H1 apparently dampens host antitumor immune responses.

B7-H1 expression in breast cancer is strongly associated with proliferative Ki-67 expression and cell cycle progression that is independent of host PD-1 (16). In human glioma, cell surface B7-H1 is posttranscriptionally controlled and closely correlated with loss of “phosphatase and tensin homolog on chromosome ten” (PTEN), a tumor suppressor gene (20). KC-specific Pten deficiency resulted in epidermal hyperplasia and accelerated tumor formation, suggesting that PTEN may be an important regulator of carcinogenesis in the skin (21).

We recently generated B7-H1 transgenic mice under the control of human keratin 14 promoter (B7-H1tg) in which epidermal KCs overexpressed B7-H1 (22). No obvious abnormality was seen in the skin or hair of B7-H1tg mice, even in aged mice. KC-associated B7-H1 directly regulated effector CD8+ T-cell function at the cutaneous inflammatory sites of contact hypersensitivity. Although overexpression of B7-H1 on skin tumors has been reported, little is known about the role of B7-H1 in skin cancer formation. Recently, intradermal, but not subcutaneous, injection of 3-methylcholanthrene (MCA) was shown to preferentially induce SCCs rather than sarcomas (23). Here, we used this method to examine the frequency of tumor formation in B7-H1tg mice and the phenotypic changes in KCs and SCCs.

Mice

B7-H1 transgenic mice under the control of the human keratin 14 (K14) promoter (C57BL/6 K14-B7-H1tg; ref. 22) were backcrossed with BALB/c mice. Transgene-positive and transgene -negative mice were used as B7-H1+/− tg (B7-H1tg/hetero) and littermate (Lm) control mice, respectively. B7-H1tg/hetero female and male mice were bred and B7-H1+/+ tg (B7-H1tg/homo) mice were obtained. BALB/c mice were purchased from Japan SLC (Shizuoka) and used as wild-type (wt) control mice for B7-H1tg/homo. PD-1-deficient mice with a BALB/c background (24) were kindly provided by Dr. Tasuku Honjo through RIKEN RBC and were maintained in our facility. All procedures were reviewed and approved by the Animal Care and Use Committee of Tokyo Medical and Dental University.

Induction of skin tumors

Skin tumors were induced as described by Wakita et al. (23) with minor modification. Mice were injected intradermally on the shaved right abdominal skin with 50 μL olive oil (Wako) containing 500 μg of MCA (Sigma-Aldrich) and monitored every week for the development of tumors. Skin tissues at the MCA injection sites were surgically removed at 3 and 7 days or at 7 weeks after MCA injection. Histological examination was performed. In separate experiments, the survival rate was monitored until 28 weeks.

TPA-induced skin inflammation

To induce skin inflammation, 12-O-tetradecanoylphorbol 13-acetate (TPA, 10 μg/200 μL in acetone; Alexis) was painted onto the shaved abdominal skin once or twice. For mAb blocking experiments, each group of mice received intraperitoneal (i.p.) injections of either control IgG (Cappel), anti-PD-1 (RMP1–14; ref. 25), or anti-B7-H1 (MIH5) mAb (200 μg/mouse; ref. 12) at days –1, 0, and 1 for 3 times. The painted sites of skin tissues were removed at 48 hours after the last painting.

Separation of epidermal sheets and isolation of epidermal cells

Dorsal halves of ear skin from intact or TPA-painted mice were incubated in 0.25% trypsin in PBS at 37°C for 45 minutes. The epidermal sheets were separated from the dorsal ear halves and used for the isolation of total RNA. For isolate single-cell suspensions of epidermal cells, the epidermal sheets were minced mechanically and filtered through nylon mesh. Single-cell suspensions of epidermal cells contained approximately 1% CD45+CD49f lymphocytes and 99% CD45CD49f+ KCs, as assessed by flow cytometry.

Histology and immuohistochemistry

Paraffin-embedded tissue sections were stained with hematoxylin and eosin (H&E). Histology was assessed by 2 independent investigators. Cryostat sections were stained with anti-B7-H1 (MIH5) mAb, and enzymatic immunohistochemistry was performed as described previously (26).

Quantitative real-time RT-PCR

Total RNA was extracted from epidermal sheets or SCC tumor mass using Isogen (Nippongene). First-strand cDNA was synthesized using oligo (dT) primers and Superscript III reverse transcriptase (Invitrogen). Real-time PCR was performed using a LightCycler instrument with a DNA Master SYBR Green I kit (Roche Diagnostics). Primer sequences are listed in the Supplementary Table S1. Data are presented as the relative expression against β-actin (Actb).

Culture of primary established tumor cells

Tumor specimens generated from MCA-injected B7-H1tg mice were mechanically minced into small tissue fragments. The fragments were placed onto culture plates and cultured with DMEM supplemented with 10% FBS and gentamicin for 10–14 days. When a large enough number of tumor cells was grown from the fragments, the cells were detached by trypsinazation and reseeded onto fresh culture plates. Experiments were performed using the cells at passages 3 to 6. The type of SCC was confirmed by histological examination. Two primary cultured SCC cells (B7-H1tg/SCC1 and B7-H1tg/SCC2) were used in this study.

Tumor inoculation and evaluation of tumor growth

Colon26 (a BALB/c-originated colon adencarcinoma cell line) was kindly obtained from Cell Resource Center for Biomedical Research, Institute of Development, Aging and Center Tohoku University. Cells were cultured, freeze-stocked, and used in accordance with the United Kingdom Coordinating Committee on Cancer Research (UKCCCR) guidelines. Colon26 (5 × 105 cells and primary cultured SCC cells originating from B7-H1tg mice (B7-H1tg/SCC1 and B7-H1tg/SCC2, 5 × 104 cells) were injected intradermally into the shaved right flank of wt, B7-H1tg, or PD‐1−/– BALB/c mice. In experiments to determine the involvement of B7-H1, 200 μg each of control IgG or ant-B7-H1 (MIH5) mAb were injected i.p. every other day after tumor inoculation. Tumor volumes were evaluated as described previously (27).

Statistics

Statistical analyses were performed using the Mann–Whitney U-test or Fisher's exact test. Values of P < 0.05 were considered significant.

B7-H1tg mice overexpress B7-H1 on epidermal KCs

To confirm the expression of B7-H1 in epidermal KCs, we examined B7-H1 mRNA and protein levels. Epidermal KCs in B7-H1tg hetero and homo mice expressed high levels of mRNA and cell surface B7-H1 (Fig. 1A and 1B). The levels of mRNA and cell surface B7-H1 in B7-H1tg/hetero mice were almost half those of B7-H1tg/homo mice. Specific overexpression in the epidermis was confirmed by immunohistochemistry (Fig. 1C).

Figure 1.

B7-H1tg mice overexpress B7-H1 on KCs. A, B7-H1 mRNA expression. Total RNA was extracted from epidermal sheets of ears from intact wt, B7-H1tg/hetero, and B7-H1tg/homo mice, and real-time PCR was performed. Values are means ± SD (n = 4). B, cell surface expression of B7-H1. Single-cell suspensions of epidermal sheets were stained with FITC-anti-CD49f (GoH3, BioLegend) and PE-anti-B7-H1 (MIH5, eBioscience), and APC-anti-CD45 (30-F11, eBioscience) or appropriate control mAbs. An electronic gate was placed on CD49f+CD45 KCs and B7-H1 expression (filled histograms) is shown with control staining (plain line histograms). The values in the upper right are the mean fluorescence intensities. C, immunohistostaining for B7-H1. Cryostat sections of respective abdominal skin were immunostained with anti-B7-H1 (MIH5) mAb. Representative images are shown. Bars, 25 μm.

Figure 1.

B7-H1tg mice overexpress B7-H1 on KCs. A, B7-H1 mRNA expression. Total RNA was extracted from epidermal sheets of ears from intact wt, B7-H1tg/hetero, and B7-H1tg/homo mice, and real-time PCR was performed. Values are means ± SD (n = 4). B, cell surface expression of B7-H1. Single-cell suspensions of epidermal sheets were stained with FITC-anti-CD49f (GoH3, BioLegend) and PE-anti-B7-H1 (MIH5, eBioscience), and APC-anti-CD45 (30-F11, eBioscience) or appropriate control mAbs. An electronic gate was placed on CD49f+CD45 KCs and B7-H1 expression (filled histograms) is shown with control staining (plain line histograms). The values in the upper right are the mean fluorescence intensities. C, immunohistostaining for B7-H1. Cryostat sections of respective abdominal skin were immunostained with anti-B7-H1 (MIH5) mAb. Representative images are shown. Bars, 25 μm.

Close modal

Earlier inflammatory responses are impaired in B7-H1tg mice

To investigate the role of B7-H1 in skin tumor formation, we used an intradermal injection of MCA to generate SCCs. No clear difference was observed between wt and B7-H1tg/homo mice before MCA injection (Fig. 2A). At 3 days after MCA injection, the epidermal thickness and number of epidermal layers were markedly increased, and these changes were more obvious in wt mice. Additionally, infiltration under the epidermal layers was more abundant in wt mice. At higher magnification, the alignment of basal cells in wt mice was well organized, whereas the basal cell alignment in the B7-H1tg mice was disturbed and more chromatin condensation was seen. At 7 days, the thickness of epidermis became comparable between wt and B7-H1tg mice. Careful observation revealed that the number of infiltrating cells in the supradermis (upper half) was clearly lower in the B7-H1tg mice, although infiltration in the deep dermis near the injected MCA/olive oil emulsion was abundant in both types of mice. The decreased inflammatory responses in B7-H1tg mice were confirmed by real-time PCR. The expression levels of IL-1α, IL-1β, IFN-γ, TNFα, and IL-6 were markedly impaired in the skin of B7-H1tg mice (Fig. 2B). In contrast, the expression of IL-10 (an antiinflammatory cytokine) was upregulated in B7-H1tg skin.

Figure 2.

Skin inflammatory responses are suppressed in B7-H1tg mice. A, H&E staining of intact or MCA-injected skin tissues at days 3 and 7. Bars, 100 μm. B, mRNA was extracted from whole skin tissues of MCA-injected sites from wt (open columns) and B7-H1tg (closed columns) mice at day 7. Cytokine expression was analyzed by real-time PCR. Values are means ± SD (n = 3). Data are representative of 2 independent experiments. C, H&E staining of skin sections at 48 h after the last TPA painting and mAb treatment. Bars, 100 μm.

Figure 2.

Skin inflammatory responses are suppressed in B7-H1tg mice. A, H&E staining of intact or MCA-injected skin tissues at days 3 and 7. Bars, 100 μm. B, mRNA was extracted from whole skin tissues of MCA-injected sites from wt (open columns) and B7-H1tg (closed columns) mice at day 7. Cytokine expression was analyzed by real-time PCR. Values are means ± SD (n = 3). Data are representative of 2 independent experiments. C, H&E staining of skin sections at 48 h after the last TPA painting and mAb treatment. Bars, 100 μm.

Close modal

Topical painting of TPA also induced rapid basal KC proliferation and skin inflammation. Histology of TPA-painted skin showed proliferation of KCs and no clear difference in epidermal thickness or layer number between wt and B1-H1tg mice (Fig. 2C). Abundant cell infiltration was seen in the subepidermis of wt mice. Similar to the MCA-induced early inflammatory responses, cell infiltration in the B7-H1tg mice was clearly impaired. To clarify the involvement of B7-H1 and PD-1 interactions in the impaired inflammatory responses, TPA-painted mice received i.p. injection with control IgG, anti-PD-1 mAb or anti-B7-H1 mAb for 3 days and then a histological analysis was conducted. Treatment with anti-PD-1 or anti-B7-H1 mAb, but not control IgG, dramatically enhanced the infiltration level in B7-H1tg mice (Fig. 2C), but the same treatment did not clearly affect wt mice (data not shown). These results suggest the involvement of the B7-H1:PD-1 pathway in the impaired inflammatory responses in B7-H1tg mice. To identify PD-1-expressing cells within the epidermis and dermis, we analyzed PD-1 expression by flow cytometry. A minor fraction of PD-1dim positive cells (∼5%) was found within CD45+CD49f epidermal lymphocytes. These cells expressed very low levels of CD3ϵ and lacked CD11c and TCRγδ expression (data not shown). No PD-1+ cells were observed in the dermis fraction. Further studies are needed to clarify the target cells involved in the B7-H1:PD-1-mediated reduction of inflammatory responses.

Epidermal tumor formation is promoted in B7-H1tg mice

MCA-injected mice were monitored weekly for the development of primary tumors, and tumor masses were resected at 7 weeks for histological analysis. Representative images of well-differentiated SCC with cancer pearl (Fig. 3A, a), undifferentiated SCC with spindle phenotype (Fig. 3A, b), and basal cell carcinoma (BCC; Fig. 3A, c) are shown. In most SCCs, scattered small colonies were observed near the cyst wall (Fig. 3A, d), suggesting that the SCC cells were derived from the epithelium of cyst walls. Overall tumor incidence was approximately threefold higher in B7-H1tg/homo and twofold higher in B7-H1tg/hetero compared with Lm control (Table 1). The incidence of SSC was significantly higher in both B7-H1tg homo and hetero mice. The incidence of basal cell tumors (BCTs), including basal cell epitheliomas and BCCs, also was increased significantly in B7-H1tg/homo mice. When we monitored the survival rates until 28 weeks after MCA injection, the final survival rate of B7-H1tg/homo mice was 20%, which was threefold less than that of wt mice (Fig. 3B). All surviving mice were tumor-free. These results demonstrate that KC-derived tumor formation was promoted in B7-H1tg mice.

Figure 3.

MCA-induced skin tumors. A, H&E staining of representative tumor tissue types 7 weeks after MCA injection. Representative images are shown of a well-differentiated SCC (a), undifferentiated SCC (b), basal cell carcinoma (BCC) (c), and scattered small colonies of SCCs near the cyst (d). Bars, 50 μm. B, survival curves after MCA injection. Data (n = 10) are representative of 2 independent experiments. C, real-time PCR for B7-H1. Intact epidermal sheets and SCCs from wt (open columns) and B7-H1tg (closed columns) mice were analyzed. Values are means ± SD (n = 3). D, immunohistostaining of B7-H1. Frozen sections of SCC tumor sites at 7 weeks after MCA injection were stained with either control rat IgG or anti-B7-H1 mAb. Bars, 250 μm.

Figure 3.

MCA-induced skin tumors. A, H&E staining of representative tumor tissue types 7 weeks after MCA injection. Representative images are shown of a well-differentiated SCC (a), undifferentiated SCC (b), basal cell carcinoma (BCC) (c), and scattered small colonies of SCCs near the cyst (d). Bars, 50 μm. B, survival curves after MCA injection. Data (n = 10) are representative of 2 independent experiments. C, real-time PCR for B7-H1. Intact epidermal sheets and SCCs from wt (open columns) and B7-H1tg (closed columns) mice were analyzed. Values are means ± SD (n = 3). D, immunohistostaining of B7-H1. Frozen sections of SCC tumor sites at 7 weeks after MCA injection were stained with either control rat IgG or anti-B7-H1 mAb. Bars, 250 μm.

Close modal
Table 1

Tumor incidence 7 weeks after MCA injection

Group(n)Tumor incidence (%)a
  Total tumors BCT SCC FS 
wt control 42 19.0 (8) 2.4 (1) 14.3 (6) 2.4 (1) 
B7-H1tg/homo 40 62.5b (25) 15.0b (6) 42.5 (17) 5.0 (2) 
      
Lm control 32 43.7 (14) 9.4 (3) 31.2 (10) 3.1 (1) 
B7-H1tg/hetero 35 68.8b (24) 11.4 (4) 57.1b (20) 0.0 (0) 
Group(n)Tumor incidence (%)a
  Total tumors BCT SCC FS 
wt control 42 19.0 (8) 2.4 (1) 14.3 (6) 2.4 (1) 
B7-H1tg/homo 40 62.5b (25) 15.0b (6) 42.5 (17) 5.0 (2) 
      
Lm control 32 43.7 (14) 9.4 (3) 31.2 (10) 3.1 (1) 
B7-H1tg/hetero 35 68.8b (24) 11.4 (4) 57.1b (20) 0.0 (0) 

aSkin tissues at the MCA-injected sites were removed after 7 weeks and histological examination was performed. Values are the data from 3 to 4 independent experiments. BCT, basal cell tumors including basal cell epithelioma and basal cell carcinoma; SCC, squamous cell carcinoma; FS, fibrosarcoma.

bStatistically different from each control group (P < 0.05).

Next, B7-H1 status was compared between wt and B7-H1tg mice. Both KCs and SCCs from wt mice expressed very low levels of B7-H1 transcripts. However, both KCs and SCCs from B7-H1tg mice showed persistently much higher levels of B7-H1 (Fig. 3C). Immunohistochemistry of B7-H1 in the B7-H1tg-SCC sections showed extremely strong B7-H1 expression in the epidermis, hair follicles, and SCC tumor sites (Fig. 3D). These results indicate that B7-H1tg-SCCs consistently possessed similar feature in terms of B7-H1 status, suggesting the conversion of KCs to SCCs.

E-cadherin is downregulated in B7-H1–overexpressing KCs

To address why B7-H1-overexpression in KCs induced higher tumor formation, we focused on molecules involved in the epithelial–mesenchymal transition (EMT; ref. 28, 29). Loss of E-cadherin and upregulation of N-cadherin have been closely correlated with migratory properties of epithelial cells to mesenchymal cells. Thus, transcripts for E-cadherin (Cdh1) and N-cadherin (Cdh2) in KCs and SCCs were compared between wt and B7-H1tg mice, using with K14 (Krt14) to assess the quantity of epidermal cells. KCs from B7-H1tg mice showed significantly lower Cdh1 expression, despite comparable levels of Krt14 (Fig. 4A). After the conversion to SCCs, Cdh1 levels in the SCCs were markedly decreased in both types of mice, but were consistently lower in B7-H1tg. In contrast, Cdh2 clearly increased after SCC conversion, with no apparent difference between wt and B7-H1tg mice. To directly confirm the preferential loss of E-cadherin in B7-H1tg-KCs, Krt14 and Cdh1 expression was examined 24 hours after topical TPA painting. TPA painting clearly enhanced the Krt14 transcripts in KCs of both wt and B7-H1tg mice, suggesting comparable expansion of KCs; however, TPA stimulation further downregulated Cdh1 in both wt and B7-H1tg.

Figure 4.

Transcripts of EMT-related molecules. A and B, total RNA was isolated from epidermal sheets of intact skin (KC), TPA-painted skin at 24 hours (TPA-KC), and SCC tumor mass (SCC) from wt (open columns) and B7-H1tg (closed columns) mice. The expression of the indicated genes was analyzed by real-time PCR. Values shown are means ± SD (n = 5–7). *, statistically significant (P < 0.05).

Figure 4.

Transcripts of EMT-related molecules. A and B, total RNA was isolated from epidermal sheets of intact skin (KC), TPA-painted skin at 24 hours (TPA-KC), and SCC tumor mass (SCC) from wt (open columns) and B7-H1tg (closed columns) mice. The expression of the indicated genes was analyzed by real-time PCR. Values shown are means ± SD (n = 5–7). *, statistically significant (P < 0.05).

Close modal

Downregulation of E-cadherin has been shown to be controlled by transcription repressors, such as Snail and Slug (29, 30). Thus, we next examined Snail (Snai1), Slug (Snai2), and Twist (Twist) transcription levels. Intact KCs did not show substantial levels of these transcripts, and no difference was observed between wt and B7-H1tg (Fig. 4B). After SCC conversion, levels of all 3 were dramatically increased, but the increased levels of Snai2 and Twist were significantly higher in the B7-H1tg-SCCs. No clear difference was seen in Snai1 expression. These results suggest that the higher incidence of skin tumor formation in B7-H1tg mice may be related to a loss of E-cadherin and the preferential upregulation of Slug and Twist.

PD-1–mediated host immune responses are involved in the enhanced tumor formation

We next examined whether B7-H1- or PD-1-mediated antitumor host responses are involved in the accelerated SCC formation. To address whether over-expressing B7-H1 on KCs affects antitumor immune responses against B7-H1negative-low tumors, Colon26 tumor cells, which expressed very low levels of B7-H1 (Fig. 5A), were inoculated into wt and B7-H1tg mice, and tumor growth was examined. No clear difference was observed between their growth curves (Fig. 5B). In addition, the analyses of regional lymph nodes 10 days after tumor inoculation showed no clear differences in the total cell number, the proportions of CD3+, CD4+, and CD8+ T cells; or IFN-γ expression in CD4+ and CD8+ T cells (data not shown). These results indicate that B7-H1 overexpression in KCs does not affect antitumor immune responses against the tumors that express endogenous low levels of B7-H1.

Figure 5.

Involvement of the B7-H1:PD-1 pathway in B7-H1–negative or –overexpressing tumor growth. A, expression of B7-H1. Colon26, B7-H1tg/SCC1, and B7-H1tg/SCC2 cell lines were stained with PE-conjugated anti-B7-H1 mAb or control IgG and analyzed by flow cytometry. Data are shown as filled histograms with control staining as plain-line histograms. B–D, Colon26 (5 × 105 cells), B7-H1tg/SCC1 or B7-H1tg/SCC2 (5 × 104 cells) were intradermally injected into syngeneic BALB/c wt (open circles), B7-H1tg (closed circles), or PD-1–/– (diamonds) mice and tumor volumes were monitored. In D, tumor-inoculated mice received an intraperitoneal injection of control rat IgG (open diamonds) or anti-B7-H1 mAb (closed diamonds) (200 μg/mouse) 3 times a week. The mean tumor volume ± SD was determined in each group of 5 mice.

Figure 5.

Involvement of the B7-H1:PD-1 pathway in B7-H1–negative or –overexpressing tumor growth. A, expression of B7-H1. Colon26, B7-H1tg/SCC1, and B7-H1tg/SCC2 cell lines were stained with PE-conjugated anti-B7-H1 mAb or control IgG and analyzed by flow cytometry. Data are shown as filled histograms with control staining as plain-line histograms. B–D, Colon26 (5 × 105 cells), B7-H1tg/SCC1 or B7-H1tg/SCC2 (5 × 104 cells) were intradermally injected into syngeneic BALB/c wt (open circles), B7-H1tg (closed circles), or PD-1–/– (diamonds) mice and tumor volumes were monitored. In D, tumor-inoculated mice received an intraperitoneal injection of control rat IgG (open diamonds) or anti-B7-H1 mAb (closed diamonds) (200 μg/mouse) 3 times a week. The mean tumor volume ± SD was determined in each group of 5 mice.

Close modal

After the malignancy conversion from KCs, the growth of B7-H1-overexpressing tumor cells may be affected by antitumor immune responses. To investigate the involvement of PD-1 and/or B7-H1 in antitumor responses against B7-H1 overexpressing tumors, we compared tumor growth between wt and PD-1−/– mice inoculated with B7-H1tg-derived SCC cells, which expressed cell surface B7-H1 at very high levels (Fig. 5A). To mimic early stage of tumors development, we intradermally inoculated 10-fold few tumor cells (5 × 104 cells) than in the experiments using colon26. The wt mice permitted the growth of B7-H1tg-derived primary tumor cells, whereas PD-1-deficient mice completely rejected the tumor engraftment (Fig. 5C). Anti-B7-H1 mAb treatment in PD-1-deficient mice canceled the tumor rejection, and tumor growth was comparable to that in wt mice (Fig. 5D). These results suggest that PD-1-mediated host responses against B7-H1-overexpressing tumors are involved in accelerated tumor formation.

We demonstrated for the first time that SCC formation was promoted in B7-H1-overexpressing skin. Our results indicate that downregulation of E-cadherin in B7-H1tg mice occurs prior to tumor development. Without stimulation, E-cadherin expression in intact KCs was constitutively downregulated in B7-H1tg mice (Fig. 4A). The disturbed alignment and atypical change of basal KCs observed in B7-H1tg mice at the earlier time point (Fig. 2A) suggest dysregulation of cell adhesion and rapid cancerous changes in basal KCs.

TGF-β1 and its related signaling have been shown to contribute greatly to the suppression of EMT or progression of tumor invasion and metastasis during skin carcinogenesis (31, 32). Although we compared TGF-β1 expression in MCA-induced inflammatory skin (Supplementary Fig. S1), intact KCs and SCCs (Supplementary Fig. S2) between wt and B7-H1tg mice, no obvious difference was observed. TGF-β1 expression was rather decreased both in the inflammatory skin and after SCC conversion. Thus, it is unlikely that the TGF-β pathway was involved in the increased carcinogenesis in B7-H1tg mice.

The inflammatory microenvironment promotes EMT-like changes in the skin and upregulation of Snail and Slug seems to occur as an upstream event prior to loss of E-cadherin (29, 30, 33). Our study revealed that inflammatory responses were markedly reduced in the skin of B7-H1tg mice in experiments with both MCA injection and TPA painting. Thus, we could not attribute the increased tumor formation in B7-H1tg mice to inflammation-induced tumor formation. Short-time TPA stimulation did not upregulate Snail, Slug, or Twist expression (data not shown). Preferential upregulation of Slug and Twist may occur in B7-H1tg during the tumor progression process after the loss of E-cadherin. Snail and Slug have been shown to contribute to cell survival and apoptosis resistance (34–36). In an ultraviolet radiation-induced murine SCC model, Snail and Slug upregulation was mediated by activation of the mitogen-activated protein kinase kinase (MAPK)/extracellular signal-regulated kinase (ERK) signaling cascade (36). Transcriptional repression of E-cadherin requires activation of MAPK/ERK signaling (37, 38). Presently, we cannot identify specific molecules involved in the B7-H1-mediated regulation of E-cadherin, Slug, and Twist expression. It is possible that overexpression of B7-H1 may transduce constitutive signals, resulting in activation of the MAPK/ERK signaling cascade. A common signaling cascade may regulate different target molecules that promote EMT during the development of tumors. We investigated whether persistent upregulation of endogenously induced B7-H1 by IFN-γ stimulation in murine SCC cell lines (NRS-1 and SCCVll) could induce downregulation of E-cadherin. The expression levels of E-cadherin were comparable before and after treatment for 2 weeks (data not shown). Cell surface expression on primary B7-H1tg/SCC cell cultures was approximately 50-fold higher than that of endogenously induced B7-H1 on SCC cell lines. Thus, the promoted carcinogenesis observed in our study may occur in only limited situations. Further studies are needed to address the signaling involved in B7-H1-mediated E-cadherin regulation.

Although the loss of Pten, a tumor suppressor gene, has been reported in B7-H1-overexpressing tumors (20), the B7-H1 transgene of B7-H1 did not alter Pten expression in B7-H1tg KCs and SCCs (Supplementary Fig. S2), Antiapoptotic reverse signaling of B7-H1 in cancer cells has been also reported (39). Therefore, we examined the effects of B7-H1-mediated reverse signaling by stimulation with immobilized PD-1Ig or anti-B7-H1 mAb-conjugated polystyrene beads. No morphological and phenotypic changes and no anti-apoptotic effects were seen in the B7-H1tg/SCC cells. The mRNA expression for cell cycle related Cyclin D1 (Ccnd1) and antiapoptotic Bcl2 (Bcl2) was also comparable between wt and B7-H1tg SCCs (Supplementary Fig. S2). These results suggest that B7-H1tg SCCs do not possess obviously distinct features in cell cycle progression and antiapoptosis.

After the atypical change or malignancy conversion from normal KCs, PD-1-mediated regulation in antitumor immune responses may strongly contribute to the promotion of tumor growth because of the much higher expression of B7-H1. Indeed, our results suggest that tumor growth of B7-H1tg-derived SCC cells was completely dependent on the B7-H1-PD-1 pathway. Overexpression of B7-H1 in KCs may promote an early stage of carcinogenesis via intrinsic cell change; however, after this process, PD-1-mediated host immune responses greatly contribute the accelerated tumor formation.

The studies using human SCC specimens of the esophagus, skin, head and neck, and oral cavity demonstrated that the decreased E-cadherin expression was associated with increased invasiveness, lymph node metastasis, and/or poor survival (40–43). Immunohistological analyses in esophageal SCCs revealed that the elevated Slug or Twist expression was significantly correlated with reduced E-cadherin and showed poor clinical outcomes (44, 45). Another study also demonstrated that Twist expression in esophageal SCCs served as an independent prognostic factor for predicting distant metastasis and survival (46). B7-H1 expression has shown to be closely correlated with poor prognosis and/or higher malignancy grade in esophageal SCCs (11). Although no parallel study for B7-H1 expression and E-cadherin or EMT-inducing transcription factors has been reported, the above results suggest a close correlation between upregulation of B7-H1 expression and reduced E-cadherin and/or enhanced Slug and Twist transcription factors in human SCCs. Further studies including in situ detection of EMT-related molecules and B7-H1 expression in the invasive front of human SCC samples are required.

In conclusion, the overexpression of B7-H1 accelerates skin tumor development, despite marked B7-H1-mediated reduction of skin inflammation. Our data show that B7-H1 serves a new role; the promotion of EMT via downregulation of E-cadherin and upregulation of Slug and Twist. Our results indicate that B7-H1 in KCs may function as an antiinflammatory molecule for short-term inflammation, but persistent overexpression of B7-H1 by continuous or repeated stimulation may cause intrinsic changes within the cell and promote carcinogenesis. Our study provides new insights into B7-H1 function in inflammation and cancer development.

No potential conflicts of interest were disclosed.

We thank Drs. K. Katsube and L.M. Sun (Tokyo Medical and Dental University, Tokyo Japan) for helpful suggestions and technical assistance with histological analysis.

This work was supported by grants from the Japan Society for the Promotion of Science (to M. Hashiguchi1 and M. Azuma) and by Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to M. Azuma).

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.
Zou
W
,
Chen
L
. 
Inhibitory B7-family molecules in the tumour microenvironment
.
Nat Rev Immunol
2008
;
8
:
467
77
.
2.
Keir
ME
,
Butte
MJ
,
Freeman
GJ
,
Sharpe
AH
. 
PD-1 and its ligands in tolerance and immunity
.
Annu Rev Immunol
2008
;
26
:
677
704
.
3.
Ansari
MJ
,
Salama
AD
,
Chitnis
T
, et al
The programmed death-1 (PD-1) pathway regulates autoimmune diabetes in nonobese diabetic (NOD) mice
.
J Exp Med
2003
;
198
:
63
9
.
4.
Youngnak-Piboonratanakit
P
,
Tsushima
F
,
Otsuki
N
, et al
The expression of B7-H1 on keratinocytes in chronic inflammatory mucocutaneous disease and its regulatory role
.
Immunol Lett
2004
;
94
:
215
22
.
5.
Koga
N
,
Suzuki
J
,
Kosuge
H
, et al
Blockade of the interaction between PD-1 and PD-L1 accelerates graft arterial disease in cardiac allografts
.
Arterioscler Thromb Vasc Biol
2004
;
24
:
2057
62
.
6.
Guleria
I
,
Khosroshahi
A
,
Ansari
MJ
, et al
A critical role for the programmed death ligand 1 in fetomaternal tolerance
.
J Exp Med
2005
;
202
:
231
7
.
7.
Dong
H
,
Strome
SE
,
Salomao
DR
, et al
Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion
.
Nat Med
2002
;
8
:
793
800
.
8.
Wintterle
S
,
Schreiner
B
,
Mitsdoerffer
M
, et al
Expression of the B7-related molecule B7-H1 by glioma cells: a potential mechanism of immune paralysis
.
Cancer Res
2003
;
63
:
7462
7
.
9.
Strome
SE
,
Dong
H
,
Tamura
H
, et al
B7-H1 blockade augments adoptive T-cell immunotherapy for squamous cell carcinoma
.
Cancer Res
2003
;
63
:
6501
5
.
10.
Konishi
J
,
Yamazaki
K
,
Azuma
M
,
Kinoshita
I
,
Dosaka-Akita
H
,
Nishimura
M
. 
B7-H1 expression on non-small cell lung cancer cells and its relationship with tumor-infiltrating lymphocytes and their PD-1 expression
.
Clin Cancer Res
2004
;
10
:
5094
100
.
11.
Ohigashi
Y
,
Sho
M
,
Yamada
Y
, et al
Clinical significance of programmed death-1 ligand-1 and programmed death-1 ligand-2 expression in human esophageal cancer
.
Clin Cancer Res
2005
;
11
:
2947
53
.
12.
Tsushima
F
,
Tanaka
K
,
Otsuki
N
, et al
Predominant expression of B7-H1 and its immunoregulatory roles in oral squamous cell carcinoma
.
Oral Oncol
2006
;
42
:
268
74
.
13.
Thompson
RH
,
Kwon
ED
. 
Significance of B7-H1 overexpression in kidney cancer
.
Clin Genitourin Cancer
2006
;
5
:
206
11
.
14.
Hamanishi
J
,
Mandai
M
,
Iwasaki
M
, et al
Programmed cell death 1 ligand 1 and tumor-infiltrating CD8+ T lymphocytes are prognostic factors of human ovarian cancer
.
Proc Natl Acad Sci USA
2007
;
104
:
3360
5
.
15.
Nomi
T
,
Sho
M
,
Akahori
T
, et al
Clinical significance and therapeutic potential of the programmed death-1 ligand/programmed death-1 pathway in human pancreatic cancer
.
Clin Cancer Res
2007
;
13
:
2151
7
.
16.
Ghebeh
H
,
Tulbah
A
,
Mohammed
S
, et al
Expression of B7-H1 in breast cancer patients is strongly associated with high proliferative Ki-67-expressing tumor cells
.
Int J Cancer
2007
;
121
:
751
8
.
17.
Yao
Y
,
Tao
R
,
Wang
X
,
Wang
Y
,
Mao
Y
,
Zhou
LF
. 
B7-H1 is correlated with malignancy-grade gliomas but is not expressed exclusively on tumor stem-like cells
.
Neuro Oncol
2009
;
11
:
757
66
.
18.
Iwai
Y
,
Terawaki
S
,
Honjo
T
. 
PD-1 blockade inhibits hematogenous spread of poorly immunogenic tumor cells by enhanced recruitment of effector T cells
.
Int Immunol
2005
;
17
:
133
44
.
19.
Hirano
F
,
Kaneko
K
,
Tamura
H
, et al
Blockade of B7-H1 and PD-1 by monoclonal antibodies potentiates cancer therapeutic immunity
.
Cancer Res
2005
;
65
:
1089
96
.
20.
Parsa
AT
,
Waldron
JS
,
Panner
A
, et al
Loss of tumor suppressor PTEN function increases B7-H1 expression and immunoresistance in glioma
.
Nat Med
2007
;
13
:
84
8
.
21.
Suzuki
A
,
Itami
S
,
Ohishi
M
, et al
Keratinocyte-specific Pten deficiency results in epidermal hyperplasia, accelerated hair follicle morphogenesis and tumor formation
.
Cancer Res
2003
;
63
:
674
81
.
22.
Ritprajak
P
,
Hashiguchi
M
,
Tsushima
F
,
Chalermsarp
N
,
Azuma
M
. 
Keratinocyte-associated B7-H1 directly regulates cutaneous effector CD8+ T cell responses
.
J Immunol
2010
;
184
:
4918
25
.
23.
Wakita
D
,
Chamoto
K
,
Ohkuri
T
, et al
IFN-γ-dependent type 1 immunity is crucial for immunosurveillance against squamous cell carcinoma in a novel mouse carcinogenesis model
.
Carcinogenesis
2009
;
30
:
1408
15
.
24.
Nishimura
H
,
Okazaki
T
,
Tanaka
Y
, et al
Autoimmune dilated cardiomyopathy in PD-1 receptor-deficient mice
.
Science
2001
;
291
:
319
22
.
25.
Yamazaki
T
,
Akiba
H
,
Koyanagi
A
,
Azuma
M
,
Yagita
H
,
Okumura
K
. 
Blockade of B7-H1 on macrophages suppresses CD4+ T cell proliferation by augmenting IFN-γ-induced nitric oxide production
.
J Immunol
2005
;
175
:
1586
92
.
26.
Chalermsarp
N
,
Azuma
M
. 
Identification of three distinct subsets of migrating dendritic cells from oral mucosa within the regional lymph nodes
.
Immunology
2009
;
127
:
558
66
.
27.
Mogi
S
,
Sakurai
J
,
Kohsaka
T
, et al
Tumour rejection by gene transfer of 4–1BB ligand into a CD80+ murine squamous cell carcinoma and the requirements of co-stimulatory molecules on tumour and host cells
.
Immunology
2000
;
101
:
541
7
.
28.
Takeichi
M
. 
Morphogenetic roles of classic cadherins
.
Curr Opin Cell Biol
1995
;
7
:
619
27
.
29.
Cano
A
,
Perez-Moreno
MA
,
Rodrigo
I
, et al
The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression
.
Nat Cell Biol
2000
;
2
:
76
83
.
30.
Bolos
V
,
Peinado
H
,
Perez-Moreno
MA
,
Fraga
MF
,
Esteller
M
,
Cano
A
. 
The transcription factor Slug represses E-cadherin expression and induces epithelial to mesenchymal transitions: a comparison with Snail and E47 repressors
.
J Cell Sci
2003
;
116
:
499
511
.
31.
Han
G
,
Lu
SL
,
Li
AG
, et al
Distinct mechanisms of TGF-β1-mediated epithelial-to-mesenchymal transition and metastasis during skin carcinogenesis
.
J Clin Invest
2005
;
115
:
1714
23
.
32.
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
.
33.
St John
MA
,
Dohadwala
M
,
Luo
J
, et al
Proinflammatory mediators upregulate snail in head and neck squamous cell carcinoma
.
Clin Cancer Res
2009
;
15
:
6018
27
.
34.
Olmeda
D
,
Montes
A
,
Moreno-Bueno
G
,
Flores
JM
,
Portillo
F
,
Cano
A
. 
Snai1 and Snai2 collaborate on tumor growth and metastasis properties of mouse skin carcinoma cell lines
.
Oncogene
2008
;
27
:
4690
701
.
35.
Newkirk
KM
,
Parent
AE
,
Fossey
SL
, et al
Snai2 expression enhances ultraviolet radiation-induced skin carcinogenesis
.
Am J Pathol
2007
;
171
:
1629
39
.
36.
Hudson
LG
,
Choi
C
,
Newkirk
KM
, et al
Ultraviolet radiation stimulates expression of Snail family transcription factors in keratinocytes
.
Mol Carcinog
2007
;
46
:
257
68
.
37.
Conacci-Sorrell
M
,
Simcha
I
,
Ben-Yedidia
T
,
Blechman
J
,
Savagner
P
,
Ben-Ze'ev
A
. 
Autoregulation of E-cadherin expression by cadherin-cadherin interactions: the roles of beta-catenin signaling, Slug, and MAPK
.
J Cell Biol
2003
;
163
:
847
57
.
38.
Pece
S
,
Gutkind
JS
. 
Signaling from E-cadherins to the MAPK pathway by the recruitment and activation of epidermal growth factor receptors upon cell-cell contact formation
.
J Biol Chem
2000
;
275
:
41227
33
.
39.
Azuma
T
,
Yao
S
,
Zhu
G
,
Flies
AS
,
Flies
SJ
,
Chen
L
. 
B7-H1 is a ubiquitous antiapoptotic receptor on cancer cells
.
Blood
2008
;
111
:
3635
43
.
40.
Kaihara
T
,
Kusaka
T
,
Kawamata
H
, et al
Decreased expression of E-cadherin and Yamamoto-Kohama's mode of invasion highly correlates with lymph node metastasis in esophageal squamous cell carcinoma
.
Pathobiology
2001
;
69
:
172
8
.
41.
Papadavid
E
,
Pignatelli
M
,
Zakynthinos
S
,
Krausz
T
,
Chu
AC
. 
Abnormal immunoreactivity of the E-cadherin/catenin (α-, β-, and γ-) complex in premalignant and malignant non-melanocytic skin tumours
.
J Pathol
2002
;
196
:
154
62
.
42.
Eriksen
JG
,
Steiniche
T
,
Sogaard
H
,
Overgaard
J
. 
Expression of integrins and E-cadherin in squamous cell carcinomas of the head and neck
.
APMIS
2004
;
112
:
560
8
.
43.
Pyo
SW
,
Hashimoto
M
,
Kim
YS
, et al
Expression of E-cadherin, P-cadherin and N-cadherin in oral squamous cell carcinoma: correlation with the clinicopathologic features and patient outcome
.
J Craniomaxillofac Surg
2007
;
35
:
1
9
.
44.
Uchikado
Y
,
Natsugoe
S
,
Okumura
H
, et al
Slug Expression in the E-cadherin preserved tumors is related to prognosis in patients with esophageal squamous cell carcinoma
.
Clin Cancer Res
2005
;
11
:
1174
80
.
45.
Sasaki
K
,
Natsugoe
S
,
Ishigami
S
, et al
Significance of Twist expression and its association with E-cadherin in esophageal squamous cell carcinoma
.
J Exp Clin Cancer Res
2009
;
28
:
158
66
.
46.
Xie
F
,
Li
K
,
Ouyang
X
. 
Twist, an independent prognostic marker for predicting distant metastasis and survival rates of esophageal squamous cell carcinoma patients
.
Clin Exp Metastasis
2009
;
26
:
1025
–32.

Supplementary data