Combination of Dacarbazine and Dimethylfumarate Efficiently Reduces Melanoma Lymph Node Metastasis

ORIGINAL ARTICLE

Combination of Dacarbazine and Dimethylfumarate Efficiently Reduces Melanoma Lymph Node Metastasis Teresa Valero1,2, Silvia Steele1, Karin Neumu¨ller1, Andreas Bracher1, Heide Niederleithner1, Hubert Pehamberger2, Peter Petzelbauer1,2 and Robert Loewe1,2 Dimethylfumarate (DMF) has been shown to reduce melanoma growth and metastasis in animal models. We addressed the question of whether DMF is as effective in its antitumor activity as the US Food and Drug Administration–approved alkylating agent dacarbazine (DTIC). We also tested the possibility of an improved antitumoral effect when both therapeutics were used together. Using our severe combined immunodeficiency (SCID) mouse model, in which xenografted human melanoma cells metastasize from primary skin sites to sentinel nodes, we show that these treatments, alone or in combination, reduce tumor growth at primary sites. Our main finding was that metastasis to sentinel nodes is significantly delayed only in mice treated with a combination of DTIC and DMF. Subsequent experiments were able to show that a combination of DTIC/DMF significantly reduced lymph vessel density in primary tumors as examined by real-time PCR and immunohistochemistry. In addition, DTIC/DMF treatment significantly impaired melanoma cell migration in vitro. In vivo, DTIC/DMF therapy significantly reduced mRNA expression and protein concentration of the promigratory chemokines CXCL2 and CXCL11. In addition, our data suggest that this xenotransplantation model is suitable for preclinical testing of various combinations of antimelanoma agents. Journal of Investigative Dermatology (2010) 130, 1087–1094; doi:10.1038/jid.2009.368; published online 26 November 2009

INTRODUCTION Metastatic melanoma has a poor prognosis. The mean survival time from first appearance of distant metastases is 6–9 months, and in most patients, distant metastases are directly responsible for cancer-related death in advancedstage melanoma. Melanoma can metastasize via three routes: (i) continuous (per continuitatem), with tumor growth extending from the primary tumor; (ii) lymphogenic, into regional (sentinel) lymph nodes; and (iii) hematogenic, into distant organs, including the lungs and brain (Balch et al., 2001). Usually, lymphogenic metastasis into regional lymph nodes (clinical stage III) precedes hematogenic metastases in distant organs (clinical stage IV; Balch et al., 2001). Of the several mechanisms that account for the development of metastases, induction of tumor cell proliferation, inhibition of apoptosis, and mobilization and invasion of tumor cells, as 1

Skin and Endothelial Research Division, Department of Dermatology, Medical University of Vienna, Vienna, Austria and 2Division of General Dermatology, Department of Dermatology, Medical University of Vienna, Vienna, Austria Correspondence: Robert Loewe, Skin and Endothelial Research Division, Department of Dermatology, Medical University of Vienna, Waehringer Guertel 18–20, A-1090 Vienna, Austria. E-mail: [email protected] Abbreviations: DMF, dimethylfumarate; DTIC, dacarbazine; SCID, severe combined immunodeficiency

Received 9 July 2009; revised 30 September 2009; accepted 15 October 2009; published online 26 November 2009

& 2010 The Society for Investigative Dermatology

well as tumor vessel formation, are the most important (Hanahan and Weinberg, 2000; Steeg, 2006). Many efforts have been made to find a therapy to improve the prognosis for metastatic melanoma, but for more than two decades the only Food and Drug Administration– approved systemic therapeutic options for stage IV melanoma have been the cytotoxic agent dacarbazine (DTIC), IL-2, and hydroxyurea (Tsao et al., 2004; Garbe and Eigentler, 2007; Mansfield and Markovic, 2009). Of the three, DTIC is by far the most often used drug because of its better effect/side effect ratio. Interestingly, although it has been in therapeutic use for more than 30 years, DTIC’s mechanism of action has not yet been fully elucidated (Eggermont and Kirkwood, 2004). Possible mechanisms by which DTIC exerts it effect are (i) alkylation of nucleic acids (Gerulath and Loo, 1972), (ii) inhibition of purine base incorporation during DNA synthesis (Hayward and Parsons, 1984), and (iii) interaction with protein sulfhydryl radicals (Saunders and Schultz, 1972). In summary, DTIC exhibits a cytotoxic effect and induces tumor cell apoptosis (Soengas and Lowe, 2003). The overall response rate of DTIC is only 15–20%, but even in patients with clinical responses these observed responses are temporary because the drug usually induces chemoresistance after a few cycles of chemotherapy. Altogether, DTIC prolongs only the disease-free interval, not median overall survival (Khan et al., 2006); therefore new therapeutic options for metastatic melanoma are urgently needed. www.jidonline.org 1087

T Valero et al. Inhibition of Melanoma Lymph Node Metastasis

RESULTS Growth of primary melanoma

A first set of experiments evaluated the therapeutic effects of different treatment strategies (DMF, DTIC, DTIC/DMF) on the growth of primary human melanomas in SCID mice. In our model using human M24met melanoma cells injected intradermally into mice, treatment was initiated when tumors were clinically apparent (palpable tumors, approximately 2 mm in diameter). Treatment with DMF or DTIC alone or in combination (DTIC/DMF) reduced tumor volumes compared with tumors grown in control animals, but the extent of the reduction was not statistically significant (Figure 1). Melanoma metastases in sentinel lymph nodes

After excision of primary tumors, mice were treated for another 10 days. Lymph nodes were then excised, and lymph node metastases were analyzed by immunohistochemistry using an antibody against human vimentin. This antibody shows no cross-reactivity with mouse vimentin (Loewe et al., 2006). The ipsilateral axillary lymph node was invariably the first lymph node (the so-called sentinel lymph node) that showed melanoma metastases in our spontaneously 1088 Journal of Investigative Dermatology (2010), Volume 130

M24met Tumor volume in mm3

Dimethylfumarate (DMF) is a fumaric acid ester used successfully in the treatment of psoriasis (Mrowietz et al., 2007). We have previously shown that DMF reduces melanoma growth and metastasis in severe combined immunodeficiency (SCID) mouse xenotransplantation models through its antiproliferative and proapoptotic effects (Loewe et al., 2006). Previous studies were able to identify inhibition of the NF-kB signaling pathway in endothelial cells and dermal fibroblasts as a possible explanation for the observed therapeutic effects (Loewe et al., 2001, 2002; Vandermeeren et al., 2001). Other studies have extended these findings by identifying similar DMF effects in other cell types (Schilling et al., 2006; Gesser et al., 2007; Meili-Butz et al., 2008). DMF is also able to interfere with the intracellular redox system by depleting glutathione and consequently inducing heme oxygenase 1 (Lehmann et al., 2007; Schmidt et al., 2007). The NF-kB signaling pathway is critical in cancer development and progression and is activated constitutively in various types of tumors, including melanoma (Yang and Richmond, 2001). Therefore, Rel transcription factors and the signaling pathways that regulate their activity are promising targets for cancer prevention and therapy (Karin, 2006; Yang et al., 2006). By using the M24met human melanoma xenotransplantation model, we sought to evaluate the therapeutic potential of DMF in comparison or in combination with the established standard chemotherapeutic DTIC. We therefore have analyzed treatment effects on primary tumor growth and lymph node metastasis in this spontaneously metastasizing mouse melanoma model. We can conclude from our data that a combination therapy of DTIC/DMF is superior to all other treatment modalities for reduction of lymph node metastasis. We were also able to show that this model is a convenient and valuable tool for testing therapeutic strategies for metastatic melanoma.

1,600 1,400 1,200 1,000 800 600 400 200 0

Control DMF DTIC DTIC+DMF

0

2

4 Days

6

8

10

Figure 1. Reduction of primary tumor growth in a severe combined immunodeficiency (SCID) mouse melanoma xenotransplantation model. Growth of human M24met melanoma cells in SCID mice before (day 1) and during treatment was measured daily. Tumor cell inoculation, treatment protocol, and tumor volume calculation were as described in Materials and Methods. n ¼ 6 animals per group; volumes are shown as mean±SEM, Po0.05.

metastasizing model. Sentinel lymph nodes within all treatment groups (n ¼ 24, total yield 100%) were examined for melanoma metastases (Figure 2). Mean sentinel lymph node volume was reduced significantly in DTIC/DMF-treated animals (Po0.05). Reduction by other treatment modalities was not significant. Effects of DTIC and DMF on tumor blood and lymph vessels

To identify the distribution of blood and lymph vessels in primary tumors, we carried out double staining of the panendothelial vessel marker CD31 and the lymphatic vessel marker Lyve-1 (see Materials and Methods). As shown in Figure 3a, immunofluorescence on tumor tissue cryosections shows that CD31 þ /Lyve-1 þ lymphatic vessels outnumber CD31 þ /Lyve-1 blood vessels in the primary melanomas of our xenotransplantation model. By analyzing the mRNA expression of the panendothelial marker CD31 by reverse transcriptase (RT)–PCR, no treatment condition (DMF, DTIC, DTIC/DMF) significantly altered CD31 mRNA expression (Figure 3b). Expression levels of CD144 and Prox-1 mRNAs were also analyzed. CD144 (VEcadherin) is a junctional protein predominantly expressed in blood endothelial cells as well as, to a lesser extent, on lymphatic endothelial cells, and Prox-1 is a highly specific marker for lymphatic endothelial cells. Both CD144 and Prox-1 mRNA expression levels were reduced by 40% after DTIC/DMF treatment; nevertheless, the reduction in mRNA expression remained statistically insignificant (CD144: P ¼ 0.11; Prox-1: P ¼ 0.083). Treatment with DMF or DTIC had no influence on CD144 mRNA expression, whereas both DMF and DTIC single treatments were able to reduce Prox-1 mRNA. However, this reduction was less distinct than that achieved by DTIC/DMF treatment (Figure 3b). Because Prox-1 mRNA expression was most effectively reduced after DTIC/DMF treatment, we next evaluated possible changes in lymph vessel density in primary tumors. Using immunohistochemistry with antibodies against mouse podoplanin, Lyve-1, and Prox-1, we identified changes in the amount of antibody-stained lymphatic vessels (podoplanin

T Valero et al. Inhibition of Melanoma Lymph Node Metastasis

500 µm

500 µm

1 mm

Sentinel lymph node volume in mm3

800 600 400 200

*

Control DMF DTIC DTIC+DMF

0

Figure 2. A combination of dacarbazine/dimethylfumarate (DTIC/DMF) is most effective in the reduction of lymph node metastases. Lymph nodes of M24met-bearing mice were collected after 18 days of treatment and stained immunohistochemically with an antihuman vimentin antibody (see Materials and Methods). Lymph nodes were grouped into macrometastases (macroscopically enlarged lymph nodes), micrometastases (normal-sized lymph nodes with melanoma cell infiltration stained red), and tumor-free lymph nodes. Bars ¼ 1 mm (left) and 500 mm (center and right). Volumes of lymph node metastases in ipsilateral axillary lymph nodes (sentinel lymph nodes) were assessed by measuring vimentin-positive areas (red). n ¼ 6 animals per group; volumes were expressed as mean±SEM, *Po0.05.

and Lyve-1) as well as in the amount of Prox-1-stained nuclei of lymphatic endothelial cells (Figure 3c). Lyve-1 and Prox-1 immunostaining showed significant reduction of Lyve-1 and Prox-1 positivity in DTIC/DMF-treated primary tumors (Figure 3c) when compared with control tumors or with tumors treated with DMF or DTIC alone. Staining with antibodies against mouse podoplanin did not reveal significant differences in the amount of positive staining among the different treatment groups (Figure 3c). Effects of DTIC and DMF on cell motility and stromal chemokine expression

Tumor metastasis strongly depends on the ability of tumor cells to migrate to and into the vasculature upon appropriate chemotactic signals. To test whether various treatment conditions also influence tumor cell motility, we performed a wounding assay in vitro. In this assay, a confluent monolayer of human M24met melanoma cells was disrupted, and the amount of regrowth into this area was measured after the indicated time points (see Materials and Methods). Treatment with DTIC/DMF led to a significant reduction in tumor cell migration; after 72 hours less than 50% of the wounded area was recovered by tumor cells. This result differed from results obtained in experiments performed with DTIC or DMF alone. In both treatment groups, the wounded area was recovered with cells after 72 hours, which was also true for the untreated control cells (Figure 4a). To analyze whether changes in chemotactic gradients may lead to altered migration of melanoma cells, we analyzed the expression of promigratory chemokines in the stroma of primary tumors in our mouse xenotransplantation model. DTIC/DMF treatment significantly reduced mRNA expression of chemokines CXCL2 and CXCL11, as well as of the

promigratory cytokine tumor necrosis factor-a (TNF-a; data not shown), as analyzed by RT-PCR. Treatment with DTIC or DMF alone showed a reduction in mRNA expression of CXCL2, CXCL11, and TNF-a when compared with untreated control mice, but these therapies were less effective than a combination therapy of DTIC/DMF (Figure 4b). Protein levels of these factors were analyzed by ELISA in the lysates of primary tumor samples taken from our M24met melanoma model (Figure 4c). Murine CXCL2 protein levels were significantly reduced with DTIC alone or with a DTIC/ DMF combination. In this assay, DTIC most efficiently reduced CXCL2 protein levels. Reduction in CXCL11 protein levels was greater with DMF/DTIC treatment than with DMF or DTIC alone as well as in control animals (Figure 4c). DISCUSSION Several mechanisms are responsible for the development of metastases: (i) increased tumor growth, (ii) inhibition of apoptosis, (iii) neoangiogenesis of blood and lymph vessels, (iv) direct interactions with the surrounding stroma, and (v) increased concentrations of chemoattractants. DTIC is antiproliferative (Alley et al., 1984) and proapoptotic (Soengas and Lowe, 2003) in vitro and in vivo. We have previously shown that DMF is also antiproliferative and proapoptotic and that it has an inhibitory effect on the NF-kB signaling pathway (Loewe et al., 2006). We postulated that these two substances might have an additive antiproliferative and proapoptotic effect. In addition, given that inhibition of NF-kB is known to reduce chemoresistance, we speculated that DMF might enhance and/or prolong the therapeutic effect of DTIC (Aggarwal, 2004). Interestingly, in our experiments we were not able to identify additive effects on tumor cell proliferation or on apoptosis. The number of www.jidonline.org 1089

T Valero et al. Inhibition of Melanoma Lymph Node Metastasis

a

CD31

LYVE-1

Merge

200 µm

CD31

CD144

Prox-1

140 120 100 80 60 40 20 0

c

Control

DTIC

DMF

DTIC+DMF

Podoplanin

Lyve-1

50 µm

Prox-1

50 µm

Positive pixel (% of total pixel)

10 8 6

*

4 2 0 Control

DTIC

DMF

DTIC+DMF

50 µm

Prox-1+ nuclei per HPF

Rel. mRNA expression (%)

b

100 80 60

* *

40 20 0

Figure 3. Dacarbazine/dimethylfumarate (DTIC/DMF) treatment significantly reduces lymphatic vessel density in primary tumors. (a) M24met primary tumor cryosections were stained by immunofluorescence with the panendothelial vessel marker CD31 (red) and the lymphatic vessel marker Lyve-1 (green). Lymphatic vessels were double positive (**), blood vessels were only CD31 þ (*). Cell nuclei were counterstained with 40 ,6-diamidino-2-phenylindole dihydrochloride (blue; see Materials and Methods). Bar ¼ 200 mm. (b) Real-time PCR analysis of mRNA expression of mouse blood and lymphatic vessel markers CD31, CD144, and Prox-1, measured on tissue samples from mouse xenotransplantation experiments. (c) Immunohistochemistry of primary tumors with antibodies against mouse podoplanin, Lyve-1, and Prox-1. Representative photomicrographs of control tumors stained with indicated antibodies are shown (see Materials and Methods; vessel lumina in Prox-1-stained image are marked with red asterisk; bar ¼ 50 mm). Positive pixel count in percentage of total pixel number is given for stainings against podoplanin and Lyve-1. Amount of Prox-1 positivity is given by number of positive nuclei per high-power field (HPF). Statistical significance was set at Po0.05 (*).

proliferating tumor cells as assessed with Ki-67 immunohistochemistry did not differ among various treatment groups (data not shown). Regarding effects on tumor cell apoptosis, DMF-, DTIC-, and DTIC/DMF-treated tumors did not show relevant differences in the extent of tumor cell apoptosis to sufficiently explain the observed differences in lymph node metastasis (data not shown). In addition, the ratio of necrotic to viable tumor tissue did not reveal any differences between primary tumors of treated animals when compared with control tumors (data not shown). Therefore, an increased 1090 Journal of Investigative Dermatology (2010), Volume 130

inhibition of proliferation or induction of apoptosis cannot solely account for the observed significant differences in lymph node metastases in the DTIC/DMF group in comparison to the other treatment groups. Neoangiogenesis of blood and lymph vessels in and around growing primary tumors is an important contributor to the development of lymph node metastases (Carmeliet and Jain, 2000). In melanoma, a direct correlation between increased tumor lymphangiogenesis and increased lymphatic metastasis has been established in animal models

T Valero et al. Inhibition of Melanoma Lymph Node Metastasis

0h 72 h

Area recovered (%)

Control

DTIC

DMF

DTIC+DMF

120 100 Control DMF DTIC DTIC+DMF

80 60

*

40 20

*

*

*

CXCL2 pg mg–1 protein

Control DMF 30 25 20 15 10 5 0

*

*

Control DMF

CXCL11 Rel. mRNA expression (%)

140 120 100 80 60 40 0

140 120 100 80 60 40 20 0

*

DTIC DTIC+DMF

CXCL11 pg mg–1 protein

CXCL2 Rel. mRNA expression (%)

0

14 12 10 8 6 4 2 0

* *

DTIC DTIC+DMF

Figure 4. Dacarbazine/dimethylfumarate (DTIC/DMF) treatment inhibits tumor cell migration and expression of stromal chemokines. (a) In vitro wounding assay. Confluent monolayers of human M24met melanoma cells were scratched (see Materials and Methods) and regrowth of cells into the cell-free area was measured after indicated time intervals. Percentages of recovered areas are expressed as mean ± SEM, *Po0.05. (b) Reverse transcriptase PCR analysis of mRNA expression of mouse stromal chemokines CXCL2 and CXCL11. Statistical significance was set at Po0.05 (*). (c) Quantification of CXCL2 and CXCL11 protein levels in tumor stroma by ELISA. Protein lysates from primary M24met melanomas grown intradermally in severe combined immunodeficiency mice and treated with indicated therapies for 10 days were analyzed (see Materials and Methods). Statistical significance was set at Po0.05 (*).

(He et al., 2004) and in clinical studies on melanoma patients (Dadras et al., 2005; Massi et al., 2006). Conversely, as recent studies have already elucidated, reduced vessel density may account for inhibited primary tumor growth as well as impaired metastasis (Liang et al., 2006). Immunohistochemistry revealed that lymphatic vessels outnumber blood vessels in and around primary tumors in our xenotransplantation model. In our experiments, DTIC/DMF treatment reduced lymph vessel density in primary tumors, as seen by reduction of mRNA expression of the lymphatic marker Prox-1 as well as by a significant reduction of Lyve-1 þ and Prox-1 þ vessels as assessed by immunohistochemistry. DTIC/DMF treatment was significantly more effective than therapy with DMF or DTIC alone. The fact that stromal fibroblasts also stained positive for podoplanin might explain why no clear

differences were observed in the podoplanin staining of treated and untreated primary tumors. As mentioned above, various chemoattractants are also crucial in the successful establishment of distant metastases. These substances are able to mobilize tumor cells, increase tumor cell motility, and target and guide tumor cells into specific host organs (Hanahan and Weinberg, 2000; Cabioglu et al., 2005; Steeg, 2006; Lin and Karin, 2007; Shields et al., 2007). We tested the regulation of chemo- and cytokines known to influence tumor cell migration. Treatment with DTIC/DMF significantly and most potently reduced mRNA expression of several such factors, such as TNF-a and the chemokines CXCL11 and CXCL2 in vivo. Interestingly, expression of all of these factors is under the transcriptional control of NF-kB (Loewe et al., 2001; Chang et al., 2009; Tang et al., 2009), which is known to be inhibited by DMF (Loewe et al., 2001). Chemokines CXCL11 and CXCL2 exhibit their chemotactic and motility-increasing effects on human tumor cells through their cognate receptors CXCR3 and CXCR2, respectively (Kawada et al., 2004; Proost et al., 2007; Richmond et al., 2009). M24met melanoma cells express both receptors, CXCR3 and CXCR2, as shown by RTPCR and immunohistochemistry (data not shown). By analyzing protein levels in tumor lysates, we found a significant reduction of CXCL2 and CXCL11 protein levels in the tumor stroma, thus further confirming our mRNA expression data. CXCL11 has also been described as exhibiting angiostatic potential (Richmond et al., 2009). Nevertheless, in our experimental model, reduction of CXCL11 mRNA expression did not lead to increased vessel density or increased mRNA expression of vascular markers (see Figure 3). The proinflammatory cytokine TNF-a also exhibits potent promigratory effects on human melanoma cells; a significant reduction in TNF-a mRNA expression has been shown to reduce tumor cell motility (Katerinaki et al., 2003; Zhu et al., 2004). We were able to show reduced mRNA expression of murine TNF-a, but in our assay, unfortunately, we were not able to detect TNF-a protein reliably in tumor lysates. This may be due to protein instability. In line with the observed reduction of promigratory chemokines in vivo is the finding that DTIC/DMF treatment was most potent among all treatments tested in inhibiting migration of human M24met cells in vitro. Taken together, these data favor the possibility that DTIC/DMF treatment inhibits melanoma lymph node metastasis by reducing lymph vessel density in primary tumors as well as influencing migration of tumor cells. Because DMF is an approved, welltolerated drug that has been in use for the treatment of psoriasis for many years, a therapy combining DTIC and DMF appears to be a reasonable new possibility to broaden therapeutic options for patients with stage IV melanoma. MATERIALS AND METHODS Cell lines, antibodies, and therapeutics Human melanoma cell lines M24met (kindly provided by RA Reisfeld; Mueller et al., 1991) and A375 (ATCC cell line, purchased from LGC Promochem, Wesel, Germany) were cultivated in Iscove’s www.jidonline.org 1091

T Valero et al. Inhibition of Melanoma Lymph Node Metastasis

modified Dulbecco’s medium supplemented with 10% fetal calf serum (Life Technologies, Vienna, Austria), 2 mMl1 glutamine (Life Technologies), and 50 IU ml1 penicillin–streptomycin (Life Technologies). Monoclonal mouse antihuman vimentin antibody (clone V9) and isotype control antibodies were from Dako Denmark A/S (Glostrup, Denmark). Rat monoclonal antibody against mouse CD31 (PECAM-1) was purchased from BD Pharmingen (San Jose, CA). Hamster polyclonal anti-mouse podoplanin, rabbit polyclonal antimouse Lyve-1, and rabbit polyclonal anti-mouse Prox-1 antibodies were purchased from Acris Antibodies (Hiddenhausen, Germany). Tetramethylrhodamine isothiocyanate–labeled second-step antibodies were from Jackson ImmunoResearch (West Grove, PA), Alexa Fluor 488-labeled second-step antibodies were from Invitrogen (Eugene, OR). For mouse experiments, DMF (20 mg kg1; Sigma, Vienna, Austria) was suspended in 0.8% methylcellulose and administered orally daily through gavage (Loewe et al., 2006). DTIC (80 mg kg1; Medac, Wedel, Germany) was given intraperitoneally in cycles of 5 consecutive days per week (Heere-Ress et al., 2005). For in vitro assays, DMF was solubilized in methanol as a 70 mM stock solution and diluted in Iscove’s modified Dulbecco’s medium for final concentrations. Stock solutions were stored at 4 1C until needed and then used within 24 hours. DTIC (stock concentration 10 mg ml1 distilled water) was light activated for 60 minutes before use and further diluted to the final concentration in Iscove’s modified Dulbecco’s medium.

SCID mouse xenotransplantation model Female, pathogen-free, 4- to 6-week-old CB17 SCID (scid/scid) mice were obtained from Charles River (Sulzfeld, Germany) and housed as described previously (Loewe et al., 2006). All procedures were carried out in accordance with the Association for Assessment and Accreditation of Laboratory Animal Care guidelines and the Guide for the Care and Use of Laboratory Animals (US Department of Health and Human Services, National Institutes of Health, publication no. 86–23). In addition, all experiments were approved by the ethics committee of the Medical University of Vienna and by the Austrian government committee on animal experimentation (reference no. GZ 66.009/33-BrGT/2006). A spontaneously metastasizing model using human M24met melanoma cells as described previously was used (Loewe et al., 2006). Briefly, 1  106 of M24met human melanoma cells were injected intradermally into the right flank of each animal. When tumors were palpable, treatment was initiated. Tumors were measured with calipers on a daily basis, and volumes were calculated as previously described (Loewe et al., 2006). Primary tumors were excised at volumes of 500–900 mm3, and treatment was continued. Ten days later, animals were killed and lymph nodes were removed. The tumor tissues were divided. One part was fixed with 4% paraformaldehyde and embedded in paraffin; the second was embedded in OCT (Tissue-Tek, Sakura Finetek Europe, Zoeterwoude, The Netherlands) and snap-frozen for cryosections and immunofluorescence; and the third was stored in RNAlater (Ambion, Austin, TX) for mRNA analysis. Lymph nodes were fixed in 4% paraformaldehyde for immunohistochemistry.

Assessment of lymph node metastases Formalin-fixed and paraffin-embedded lymph nodes were stained with hematoxylin and eosin according to standard procedures. 1092 Journal of Investigative Dermatology (2010), Volume 130

Lymph nodes were grouped into macroscopically enlarged (macrometastases) and regular-sized lymph nodes. Macrometastatic lymph nodes were cut at their largest diameter before processing. For immunohistochemistry, serial sections were deparaffinized according to routine procedures and incubated in citrate buffer (pH 6.0; Dako Denmark A/S) overnight at 80 1C. Staining with an antibody against human vimentin (clone V9), which does not cross-react with mouse vimentin, and an appropriate isotype control was performed using a DAKO TechMate Horizon automated staining system according to the manufacturer’s protocol and as described previously (Loewe et al., 2006). Tumor volumes of lymph node metastases were calculated from hematoxylin-and-eosin-stained slides using an AxioCam MRc5 digital camera (Zeiss, Vienna, Austria) attached to an AH3-RFCA microscope (Olympus, Vienna, Austria) and AxioVision Rel 4.4 software (Zeiss).

Analysis of blood and lymph vessels Anti-mouse CD31 and Lyve-1 double staining was performed with immunofluorescence on 4 mm cryosections, fixed in acetone and airdried for 30 minutes. After incubation with respective secondary antibodies, 40 ,6-diamidino-2-phenylindole dihydrochloride (Sigma, St Louis, MO) was used for nuclear counterstaining. Sections were analyzed on a confocal laser scan microscope (LSM 510; Zeiss). For immunohistochemical staining of murine lymphatic tumor vessels, formalin-fixed and deparaffinized primary tumor sections were treated with citrate buffer (pH 6.0) for antigen retrieval by heating using an autoclave at either 100 1C for 10 minutes or 80 1C overnight. Binding of primary antibodies was detected with a biotinylated anti-hamster antibody and a streptavidin–biotin–horseradish peroxidase complex (for podoplanin antibody detection) or by using an anti-rabbit horseradish peroxidase UltraVision LP Detection System (Thermo Scientific, Fremont, CA) for Prox-1 antibody detection. Finally, counterstaining of both stainings was performed using Papanicolaou’s hematoxylin solution (Merck, Darmstadt, Germany). Lyve-1 staining was performed in a DAKO TechMate Horizon automated staining system following the manufacturer’s instructions. Quantification of positive-stained tissue areas/positivestaining nuclei was assessed using ImageScope software (Aperio Technologies, Vista, CA).

mRNA expression in vivo RNA was prepared from tissue samples stored in RNAlater using RNeasy Mini RNA kits (Qiagen, Valencia, CA) according to the manufacturer’s instructions. RNA was reverse transcribed into firststrand cDNA using Revert Aid M-MulV Reverse Transcriptase (Fermentas, Vienna, Austria) and hexamer primers (Roche, Vienna, Austria) for 90 minutes at 42 1C. Real-time PCR was performed in doublets as previously described (Loewe et al., 2006). The primer sets used for detection of specific target gene mRNA were Assays-on-Demand (Applied Biosystems, Vienna, Austria; Table 1). For normalization, a b-2 microglobulin (b-2m) primer set was used (Applied Biosystems; Table 1). Reactions were run on an ABI Prism 7700 Sequence Detector (PerkinElmer, Vienna, Austria), and data were analyzed with the SDS 1.9.1 software (Applied Biosystems). Changes in mRNA concentration were determined by subtracting the Ct (cycle threshold) of the target gene from the Ct of the housekeeping gene (b-2m) (D ¼ Ct gene–Ct b-2m). The mean of D control was subtracted from the D target gene reaction

T Valero et al. Inhibition of Melanoma Lymph Node Metastasis

Table 1. Primer sets used for the detection of mouse mRNA expression in vivo CD31

Mm01242584_m1

CD144

Mm00486938_m1

Prox-1

Mm00435969_m1

CXCL2

Mm00436450_m1

CXCL11

Mm00444662_m1

TNF-a

Mm99999068_m1

b-2 Microglobulin

Mm00437762_m1

ACKNOWLEDGMENTS This study was supported in part by grants 11912 and 13046 from the ¨ sterreichische Nationalbank and by the Chamber of Jubila¨umsfonds of the O Commerce of Vienna (‘‘Wirtschaftskammerpreis 2007’’). We thank Udo Losert and the staff of the Biomedical Sciences Center, Medical University of Vienna, Austria, and Marion Groeger for helpful discussions.

REFERENCES Aggarwal BB (2004) Nuclear factor-kappa-B: the enemy within. Cancer Cell 6:203–8 Alley MC, Powis G, Appel PL et al. (1984) Activation and inactivation of cancer chemotherapeutic agents by rat hepatocytes cocultured with human tumor cell lines. Cancer Res 44:549–56

(mean D control – D target gene). The difference was calculated as 2(DDCt).

Balch CM, Buzaid AC, Soong SJ et al. (2001) Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol 19:3635–48

In vitro wounding assay

Cabioglu N, Yazici MS, Arun B et al. (2005) CCR7 and CXCR4 as novel biomarkers predicting axillary lymph node metastasis in T1 breast cancer. Clin Cancer Res 11:5686–93

Cultured M24met tumor cells (300,000 per well) were plated onto six-well plates (Costar; Corning, Corning, NY) in 3 ml medium and were allowed to grow until they had formed a compact monolayer. Using a pipette tip, the cellular monolayer was scratched twice per well to obtain cell-free lines followed by repeated washing to remove cellular debris. Cells were then cultivated in medium containing the indicated concentrations of DMF, DTIC, DTIC/DMF, or medium alone, which was renewed every 24 hours. A black line drawn on the undersurface of the well plate was used to identify the wounded area during re-epithelialization. For documentation, photographs were taken immediately after wounding and 48 and 72 hours later. The amount of migration was quantified by placing the measured wound size area at 72 hours in relation to the respective original wound size using AxioVision Rel 4.4 software.

Mouse chemokine ELISA Primary tumors grown intradermally in SCID mice and treated with the indicated conditions were removed, and tumor sections (50 mg) were snap-frozen in liquid nitrogen. Tissue was extracted into 750 ml of lysis buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.5% NP-40, 1 mM EDTA, 10 ml ml1 Protease Inhibitor Cocktail (Sigma, St Louis, MO) by tissue homogenization using Mixer Mill MM200 (2 minutes, 20 per second; Retsch, Haan, Germany) and centrifugation at 15,000 r.p.m. for 10 minutes at 4 1C. Supernatants were collected and were either used immediately or stored at 80 1C. Mouse CXCL2/CXCL11 DuoSet ELISA was purchased from R&D Systems (R&D Systems Europe, Abingdon, UK) and the assay was performed according to the manufacturer’s instructions. Results were normalized to total protein concentration using a modified Bradford assay (Bio-Rad Laboratories, Hercules, CA).

Statistical analyses Statistical comparison was performed with two-sided Student’s t-test. P-values o0.05 were considered statistically significant. Calculations were performed with SPSS 16 software (SPSS, Chicago, IL). Statistical results are presented as mean±standard error of the mean (SEM). CONFLICT OF INTEREST The authors state no conflict of interest.

Carmeliet P, Jain RK (2000) Angiogenesis in cancer and other diseases. Nature 407:249–57 Chang CC, Liu Z, Vlad G et al. (2009) Ig-like transcript 3 regulates expression of proinflammatory cytokines and migration of activated T cells. J Immunol 182:5208–16 Dadras SS, Lange-Asschenfeldt B, Velasco P et al. (2005) Tumor lymphangiogenesis predicts melanoma metastasis to sentinel lymph nodes. Mod Pathol 18:1232–42 Eggermont AM, Kirkwood JM (2004) Re-evaluating the role of dacarbazine in metastatic melanoma: what have we learned in 30 years? Eur J Cancer 40:1825–36 Garbe C, Eigentler TK (2007) Diagnosis and treatment of cutaneous melanoma: state of the art 2006. Melanoma Res 17:117–27 Gerulath AH, Loo TL (1972) Mechanism of action of 5-(3,3-dimethyl-1triazeno)imidazole-4-carboxamide in mammalian cells in culture. Biochem Pharmacol 21:2335–43 Gesser B, Johansen C, Rasmussen MK et al. (2007) Dimethylfumarate specifically inhibits the mitogen and stress-activated kinases 1 and 2 (MSK1/2): possible role for its anti-psoriatic effect. J Invest Dermatol 127:2129–37 Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100:57–70 Hayward IP, Parsons PG (1984) Epigenetic effects of the methylating agent 5(3-methyl-1-triazeno) imidazole-4-carboxamide in human melanoma cells. Aust J Exp Biol Med Sci 62(Part 5):597–606 He Y, Rajantie I, Ilmonen M et al. (2004) Preexisting lymphatic endothelium but not endothelial progenitor cells are essential for tumor lymphangiogenesis and lymphatic metastasis. Cancer Res 64:3737–40 Heere-Ress E, Boehm J, Thallinger C et al. (2005) Thalidomide enhances the anti-tumor activity of standard chemotherapy in a human melanoma xenotransplantation model. J Invest Dermatol 125:201–6 Karin M (2006) Nuclear factor-kappaB in cancer development and progression. Nature 441:431–6 Katerinaki E, Evans GS, Lorigan PC et al. (2003) TNF-alpha increases human melanoma cell invasion and migration in vitro: the role of proteolytic enzymes. Br J Cancer 89:1123–9 Kawada K, Sonoshita M, Sakashita H et al. (2004) Pivotal role of CXCR3 in melanoma cell metastasis to lymph nodes. Cancer Res 64:4010–7 Khan MA, Andrews S, Ismail-Khan R et al. (2006) Overall and progressionfree survival in metastatic melanoma: analysis of a single-institution database. Cancer Control 13:211–7 Lehmann JCU, Listopad JJ, Rentzsch CU et al. (2007) Dimethylfumarate induces immunosuppression via glutathione depletion and subsequent induction of heme oxygenase 1. J Invest Dermatol 127:835–45 Liang WC, Wu X, Peale FV et al. (2006) Cross-species vascular endothelial growth factor (VEGF)-blocking antibodies completely inhibit the growth

www.jidonline.org 1093

T Valero et al. Inhibition of Melanoma Lymph Node Metastasis

of human tumor xenografts and measure the contribution of stromal VEGF. J Biol Chem 281:951–61

imidazole-4-(5)carboxamide in Bacillus subtilis. Biochem Pharmacol 21:2065–76

Lin WW, Karin M (2007) A cytokine-mediated link between innate immunity, inflammation, and cancer. J Clin Invest 117:1175–83

Schilling S, Goelz S, Linker R et al. (2006) Fumaric acid esters are effective in chronic experimental autoimmune encephalomyelitis and suppress macrophage infiltration. Clin Exp Immunol 145:101–7

Loewe R, Valero T, Kremling S et al. (2006) Dimethylfumarate impairs melanoma growth and metastasis. Cancer Res 66:11888–96 Loewe R, Holnthoner W, Groger M et al. (2002) Dimethylfumarate inhibits TNF-induced nuclear entry of NF-kappa B/p65 in human endothelial cells. J Immunol 168:4781–7

Schmidt TJ, Ak M, Mrowietz U (2007) Reactivity of dimethyl fumarate and methylhydrogen fumarate towards glutathione and N-acetyl-l-cysteine— preparation of S-substituted thiosuccinic acid esters. Bioorg Med Chem 15:333–42

Loewe R, Pillinger M, de Martin R et al. (2001) Dimethylfumarate inhibits tumor-necrosis-factor-induced CD62E expression in an NF-kappa Bdependent manner. J Invest Dermatol 117:1363–8

Shields JD, Emmett MS, Dunn DB et al. (2007) Chemokine-mediated migration of melanoma cells towards lymphatics—a mechanism contributing to metastasis. Oncogene 26:2997–3005

Mansfield AS, Markovic SN (2009) Novel therapeutics for the treatment of metastatic melanoma. Future Oncol 5:543–57

Soengas MS, Lowe SW (2003) Apoptosis and melanoma chemoresistance. Oncogene 22:3138–51

Massi D, Puig S, Franchi A et al. (2006) Tumour lymphangiogenesis is a possible predictor of sentinel lymph node status in cutaneous melanoma: a case–control study. J Clin Pathol 59:166–73

Steeg PS (2006) Tumor metastasis: mechanistic insights and clinical challenges. Nat Med 12:895–904

Meili-Butz S, Niermann T, Fasler-Kan E et al. (2008) Dimethyl fumarate, a small molecule drug for psoriasis, inhibits nuclear factor-kappaB and reduces myocardial infarct size in rats. Eur J Pharmacol 586: 251–8 Mrowietz U, Altmeyer P, Bieber T et al. (2007) Treatment of psoriasis with fumaric acid esters (Fumaderm). J Dtsch Dermatol Ges 5:716–7 Mueller BM, Romerdahl CA, Trent JM et al. (1991) Suppression of spontaneous melanoma metastasis in scid mice with an antibody to the epidermal growth-factor receptor. Cancer Res 51:2193–8 Proost P, Mortier A, Loos T et al. (2007) Proteolytic processing of CXCL11 by CD13/aminopeptidase N impairs CXCR3 and CXCR7 binding and signaling and reduces lymphocyte and endothelial cell migration. Blood 110:37–44 Richmond A, Yang J, Su Y (2009) The good and the bad of chemokines/ chemokine receptors in melanoma. Pigment Cell Melanoma Res 22:175–86 Saunders PP, Schultz GA (1972) Role of 4-diazoimidazole-5-carboxamide in the action of the antitumor agent 5(4)-(3,3-dimethyl-1-triazeno)

1094 Journal of Investigative Dermatology (2010), Volume 130

Tang W, Wang W, Zhang Y et al. (2009) Tumour necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced chemokine release in both TRAIL-resistant and TRAIL-sensitive cells via nuclear factor kappa B. FEBS J 276:581–93 Tsao H, Atkins MB, Sober AJ (2004) Management of cutaneous melanoma. N Engl J Med 351:998–1012 Vandermeeren M, Janssens S, Wouters H et al. (2001) Dimethylfumarate is an inhibitor of cytokine-induced nuclear translocation of NF-kappa B1, but not RelA in normal human dermal fibroblast cells. J Invest Dermatol 116:124–30 Yang J, Amiri KI, Burke JR et al. (2006) BMS-345541 targets inhibitor of \{kappa\}B kinase and induces apoptosis in melanoma: involvement of nuclear factor \{kappa\}B and mitochondria p. Clin Cancer Res 12:950–60 Yang JM, Richmond A (2001) Constitutive I kappa B kinase activity correlates with nuclear factor-kappa B activation in human melanoma cells. Cancer Res 61:4901–9 Zhu N, Lalla R, Eves P et al. (2004) Melanoma cell migration is upregulated by tumour necrosis factor-alpha and suppressed by alpha-melanocytestimulating hormone. Br J Cancer 90:1457–63

Copyright of Journal of Investigative Dermatology is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.