Merkel Cell Polyomavirus–Positive Merkel Cell Carcinoma Cells Do Not Require Expression of the Viral Small T Antigen

ORIGINAL ARTICLE

Merkel Cell Polyomavirus–Positive Merkel Cell Carcinoma Cells Do Not Require Expression of the Viral Small T Antigen Sabrina Angermeyer1, Sonja Hesbacher1, Ju¨rgen C Becker2, David Schrama2 and Roland Houben1 Increasing evidence suggests that Merkel cell carcinoma (MCC) is caused by the Merkel cell polyomavirus (MCV). The viral sequence encodes for two potential oncoproteins, i.e., the small T antigen (sT) and the large T antigen (LT). Indeed, sT has recently been shown to bear transforming activity. Here, we confirm this observation by demonstrating focus formation upon expression of MCV sT in NIH3T3 fibroblasts. On the other hand, however, we provide evidence that established MCC cells do not require sT for growth and survival. Silencing of sT protein expression by two different sT-specific short hairpin RNAs (shRNAs) leads to variable degrees of growth retardation in MCV-positive MCC cell lines. However, these effects are not sT specific, as proliferation of MCVnegative cell lines is similarly affected by these sT shRNAs. Furthermore, ectopic expression of shRNA-insensitive sT does not revert the growth inhibition implicated by sT silencing. Finally, the unambiguous and specific growth inhibition induced by means of an shRNA targeting both T antigens, can be completely rescued by ectopic expression of LT alone, thus demonstrating a dispensable role of sT. Altogether, our results indicate that MCV LT is more relevant in maintaining the proliferation and survival of established MCC cell lines. Journal of Investigative Dermatology (2013) 133, 2059–2064; doi:10.1038/jid.2013.82; published online 4 April 2013

INTRODUCTION Merkel cell carcinoma (MCC) is the most aggressive skin cancer with one-third of the affected patients dying from their disease (Becker, 2010). MCC is a very rare malignancy, although its incidence is rising (Hodgson, 2005). Risk factors for developing MCC are immunosuppression, UV exposure, and old age (Becker et al., 2009b). The molecular pathogenesis of MCC has been elucidated by the discovery of a new polyomavirus, termed Merkel cell polyomavirus (MCV), integrated into the genome of most MCCs (Feng et al., 2008). Many groups have confirmed the presence of MCV in about 80% of MCC tumors (Kassem et al., 2008; Becker et al., 2009a; Garneski et al., 2009), but this might actually be an underestimation as sensitivity of MCV detection is higher when fresh or frozen samples (Martel-Jantin et al., 2012) or improved immunohistochemistry and PCR methods are used (Rodig et al., 2012). MCV, like other polyomaviruses, encodes the T antigens (TAs), i.e., the small and the large T antigen (sT and LT; Shuda 1

Department of Dermatology, University Hospital Wu¨rzburg, Wu¨rzburg, Germany and 2Department of General Dermatology, Medical University Graz, Graz, Austria

Correspondence: Roland Houben, Department of Dermatology, University Hospital Wu¨rzburg, Josef-Schneider-Street 2, D-97080 Wu¨rzburg, Germany. E-mail: [email protected] Abbreviations: cDNA, complementary DNA; GFP, green fluorescent protein; MCC, Merkel cell carcinoma; MCV, Merkel cell polyomavirus; LT, large T antigen; PBS, phosphate-buffered saline; shRNA, short hairpin RNA; sT, small T antigen; TA, T antigen Received 18 July 2012; revised 22 January 2013; accepted 24 January 2013; accepted article preview online 25 February 2013; published online 4 April 2013

& 2013 The Society for Investigative Dermatology

et al., 2008, 2009, 2011). These are derived by alternative splicing from a common gene locus; the first 78 N-terminal amino acids are shared, whereas the C terminus is different. In contrast to wild-type LT, most of the MCC-associated LTs are truncated deleting a part of LT necessary for virus replication. This is caused either by integration break points within the LT sequence or by stop codon mutations (Shuda et al., 2008; Schmitt et al., 2012). The retinoblastoma protein-binding motif, however, is generally preserved in MCC suggesting an essential function of the interaction of LT with the cell cycle regulator protein retinoblastoma protein for MCC tumorigenesis. Knockdown of TA expression using short hairpin RNAs (shRNAs) targeting both TAs induces cell cycle arrest and apoptosis of MCV þ MCC cell lines in vitro and regression of established xenograft tumors (Houben et al., 2010, 2012). Consequently, TAs are regarded as attractive therapeutic targets to treat MCC. In order to develop such a therapeutic approach, the extent to which sT and LT contribute to the oncogenic phenotype of MCV þ MCC cells has to be established. Notably, the TA shRNA-induced anti-proliferative effects can be rescued by ectopic expression of shRNA insensitive TA. This rescue, however, is abrogated by a point mutation in LT interfering with its ability to bind to the retinoblastoma protein (Houben et al., 2012). Moreover, specific knockdown of LT alone inhibits proliferation of MCV þ MCC, which again can be rescued by re-expression of LT (Houben et al., 2012). Here, we scrutinize to which extent MCV þ MCC cell lines are addicted to MCV sT for continuous proliferation. To this end, silencing of sT protein expression by sT-specific shRNAs www.jidonline.org 2059

S Angermeyer et al. MCV Small T Antigen in Merkel Cell Carcinoma

sT2

sT1

TA

shRNA

Scr

Tubulin

WaGa

LT Tubulin

Relative ratio shRNA infected/uninfected cells

sT2

sT

293T pIH-sT

sT1

TA

shRNA

Scr

MCV– cell lines Jurkat

1.2 1

1

0.8

0.8

0.6

Scr

0.6

0.4

TA sT1

0.4

0.2

MCC13

1.2

0.2

sT2

0

0 4

9

14 Days

19

24

4

9

14 Days

19

24

19

24

Relative ratio shRNA infected/uninfected cells

MCV+ cell lines WaGa

1.2

MKL-2

1.2

1

1

1

0.8

0.8

0.8

0.6

0.6

0.6

0.4

0.4

0.4

0.2

0.2

0.2

0

0 4

9

14 Days

19

24

MKL-1

1.2

0 4

9

14 Days

19

24

4

9

14 Days

Figure 1. Variable response of Merkel cell polyomavirus (MCV)-positive Merkel cell carcinoma (MCC) cell lines to two small T antigen (sT)-specific short hairpin RNAs (shRNAs). (a) HEK 293T cells were co-transfected with an expression construct for V5-tagged MCV sT and the indicated shRNA constructs targeting either T antigen (TA), large T antigen (LT), or sT (two different shRNAs: sT1 and sT2). A scrambled shRNA (Scr) was used as control. Total cell lysates harvested after 2 days of incubation were subjected to immunoblot analysis using a V5 antibody to visualize sT expression. WaGa cells infected with the same shRNA constructs were analyzed on day 5 after infection for expression of MCV LT. (b and c) The indicated MCV-negative (b) and MCV-positive (c) cell lines were infected with the lentiviral shRNA constructs also coding for GFP and were then mixed with uninfected parental cells. Ratios of fluorescent/non-fluorescent cells were measured over time. Mean values (± SD) of 8 (WaGa), 3 (MKL-2), 6 (MKL-1 and MCC13), or 7 (Jurkat) independent experiments are depicted.

led to variable degrees of growth inhibition of both MCV  and MCV þ MCC cell lines. Furthermore, ectopic expression of shRNA-insensitive sT did not revert the growth inhibition induced by sT silencing, whereas ectopic expression of only LT could rescue the TA knockdown phenotype. RESULTS AND DISCUSSION Shuda et al. (2011) recently reported that MCV encoded sT bears transforming activity. Moreover, using the MCV þ MCC cell line, MKL-1, they demonstrated that shRNA knockdown of sT impeded cellular metabolic activity. Indeed, as measured by Wst-1 assay, the inhibiting effect was as strong as upon knockdown of both TAs (Shuda et al., 2011). However, in contrast to TA knockdown no increased cell death and only moderate cell cycle effects were observed upon sT silencing; thus the modus operandi of the inhibitory effects of sT knockdown were left in uncertainty. Prompted by this report, we scrutinized the dependence of MCV þ MCC cells on sT expression in more detail. For this purpose, we used one established TA-specific shRNA targeting both LT and sT and two different new sT-specific shRNAs; the latter is important as shRNAs frequently exhibit off-target effects (Jackson et al., 2006). All three shRNAs were effective in knocking down sT expression, whereas only the TA-specific shRNA reduced LT expression (Figures 1a and 2). As the used lentiviral shRNA 2060 Journal of Investigative Dermatology (2013), Volume 133

vectors also encode green fluorescent protein (GFP), the shRNA induced changes in the proliferation of three MCV þ and two MCV  cell lines could be pinpointed in mixed cultures with unmarked control cells. Although the TA-specific shRNA uniformly inhibited the growth of all three MCV þ cells, the effect of sT-specific shRNAs was very variable. Furthermore, sT-specific shRNAs also exerted some growth inhibitory effects on one of the MCV  cell lines (Figure 1b and c). These observations already suggested off-target effects of the sT-specific shRNAs. To formally distinguish between specific and off-target effects, we performed rescue experiments by stably expressing sT rendered insensitive to either of the sT-specific shRNA in MCV þ MCC cell lines (Figure 2). Notably, virtually no rescue of the shRNA-induced growth inhibition was observed. The interpretation of these results, however, is complicated by the fact that due to the low expression levels of sT (Houben et al., 2010; Shuda et al., 2011), we did not succeed to measure endogenous sT protein: even when using three different antibodies targeting either an epitope in exon 1 common to sT and LT (2T2, CM8E6) or an sT-specific epitope (CM5E1) no sT could be detected by immunoblot analyzing lysates from WaGa (Figure 3) or any other MCV þ MCC cell line (data not shown). As control, sT overexpressed in 293 T cells was readily detectable (Figure 3). Anyhow, such a low sT expression per se argues against an essential role of sT for the

S Angermeyer et al. MCV Small T Antigen in Merkel Cell Carcinoma

MCC cells. Nevertheless, to redress the inability to monitor endogenous sT protein expression additional rescue experiment were performed. We have previously demonstrated that the used TA shRNA effectively reduces LT and sT levels (Figures 1a and 4a; Houben et al., 2010). Notably, this TA knockdown-induced growth inhibition is completely rescued by expression of a TA shRNA insensitive LT (Figure 4). Indeed, the rescue by expression of shRNA-insensitive LT is as efficient as the rescue by an shRNA-insensitive TA (Figure 4). It should further be noted that the rescue capacity of a shRNA insensitive TA expression vector is completely abrogated by a mutation affecting only LT (Houben et al., 2012). The abovedescribed results suggest that MCV þ MCC cells do not essentially require sT expression for proliferation and survival. It should be noted, however, that in contrast to the low expression levels of sT in MCC cell lines in vitro and in MCC tissues in situ, sT is more frequently detected in MCC tissues by immunohistochemistry than LT (Shuda et al., 2011); this notion argues for an important role of sT in tumor maintenance. Then again, a recent publication using a new LT antibody demonstrates LT expression in virtually all MCC samples by immunohistochemistry (Rodig et al., 2012). Our observations suggest that LT is more relevant for proliferation and survival of the MCV þ MCC cells than sT, as the strict oncogene addiction is only apparent for LT (Figures 1, 2 and 4). Still, MCV sT bears a higher transforming capacity on fibroblasts than LT (Shuda et al., 2011) implying that MCV sT might be the dominant oncogene in MCC tumorigenesis. This is not necessarily a contradiction as it is conceivable that sT is more involved in the initial oncogenic transformation, whereas LT is more important for the maintenance of established MCC cells. To confirm this split in importance of the two TAs in these two processes, we reevaluated the transforming capability of the MCV TAs by comparison to established oncogenic proteins such as the SV40 TAs and BRAFV600E (Ali and DeCaprio, 2001; Hahn et al., 2002; Houben et al., 2004). To this end, NIH3T3 fibroblasts were infected with high-titer lentiviral expression vectors leading to stable integration of the expression constructs in at least 30% of the cells (data not shown). Consistent with the results of Shuda et al., we observed induction of focus formation only by the MCV sT expression construct, whereas the vector carrying an MCC-derived truncated MCV LT complementary DNA (cDNA) did not induce foci (Figure 5 a and b). Focus formation in MCV sTexpressing fibroblasts, however, was very weak compared with SV40 early region or BRAFV600E. Actually, although for MCV sT construct only small foci just started to form 25 days after infection, for the BRAFV600E or SV40 early region constructs the complete plate was covered with morphologically transformed cells, which had lost contact inhibition (Figure 5b). The difficulty to recapitulate transforming capacity of the MCV TAs in vitro is not unique to MCV as, e.g., Kaposi’s sarcoma-associated herpes virus does not induce full transformation when analyzed in vitro (Mesri et al., 2010). Such an effect may be due to the lack of microenvironmental cofactors in the cell culture system (Madkan et al., 2007; Boccardo et al., 2010). Alternatively, the inability of MCV

LT to induce focus formation may be attributable to the fact that fibroblasts are not the ideal target cells. To date, only MCC has been convincingly demonstrated to be caused by MCV although MCV is widespread (Pastrana et al., 2009), can be detected in many different organs (Matsushita et al., 2012), and seems to have a broad entry tropism (Schowalter et al., 2012). These observations and the fact that MCC is a very rare disease suggest that only a limited number of cell types can be transformed by MCV; thus, only in this respective cell type the genuine transforming ability of the MCV TAs can be established. In summary, the data presented here confirm transforming activity of MCV sT in mouse fibroblasts, but suggest that established MCC cells do not essentially depend on sT expression. In contrast, MCV encoded LT, which cannot transform fibroblasts, is strictly required for growth and survival of MCV þ MCC cells. Thus, MCV LT appears to be the more suitable target for therapeutic approaches to treat patients with clinically evident MCC. MATERIALS AND METHODS Cell culture The MCV-positive MCC cell lines WaGa (Houben et al., 2010), MKL-1 (Rosen et al., 1987), and MKL-2 (Van Gele et al., 2002) as well as the MCV-negative MCC cells MCC13 (Leonard et al., 1995) and the T-cell lymphoma cell line Jurkat (Gillis and Watson, 1980) were used. All cell lines were grown in RPMI 1640 supplemented with 10% fetal calf serum, 100 U ml  1 penicillin, and 0.1 mg ml  1 streptomycin. MKL-1 and MKL-2, which grow as spheroids, were dissociated with Trypsin-EDTA solution before lentiviral infections, cell counting, or flow cytometric analysis. Cell lines stably expressing shRNAinsensitive LT or sT were raised by infection of the cells with retroviral expression constructs (pIH) containing a hygromycinresistance gene and the respective cDNA. Hygromycin-resistant cells were selected by a 3-week culture in the presence of 100 mg ml  1 hygromycin B.

shRNA constructs The lentiviral shRNA vector KH1, which drives constitutive coexpression of an shRNA and GFP (Verhaegen et al., 2006), was used to achieve knockdown of sT or of both MCV TAs. For the latter, an shRNA sequence (sense strand 50 -ATCCACAAGCTCAGAAG TGACTTCTCTATTCAAGAGATAGAGAAGTCACTTCTGAGCTTGTG GATTTTTTT-30 ) designed to target the nucleotides 391–419 (accession number: EU375803) present in all T-antigen messenger RNAs (see Figure 1) was used. To specifically target sT either the shRNA sequence sT1 (sense strand 50 -GCTTAAGCAACTTAGAGATTCTAT CAAGAGTAGAATCTCTAAGTTGCTTAAGCTTTTTT-30 ) corresponding to the nucleotides 525–547 (accession number: EU375803) or sT2 (sense strand 50 -GGAAGAATATGGAACTTTATCAAGAGTA AAGTTCCATATTCTTCCTTTTTT-30 ) corresponding to the nucleotides 450–468 (accession number: EU375803), respectively, were cloned into the KH1 vector. As control, a scrambled KH1 construct targeting neither human nor MCV transcripts was used.

TA expression constructs For generation of MCC cell lines stably expressing ectopic shRNAinsensitive sT, LT, or TA, the respective cDNAs were cloned into the www.jidonline.org 2061

S Angermeyer et al. MCV Small T Antigen in Merkel Cell Carcinoma

MKL-1

Ectopic sT

LT Tubulin

sT (sT1 shRNA insensitive) Vector

Relative ratio shRNA infected/uninfected cells

Scr sT1 sT2

sT sT sT1 shRNA sT2 shRNA insensitive insensitive Scr sT1 sT2

shRNA

Vector Scr sT1 sT2

Ectopic expression

sT (sT2 shRNA insensitive)

1 0.8 0.6 0.4 0.2

sT1 shRNA

sT2 shRNA

0 4

9

14 Days

19

WaGa

Ectopic sT

LT Tubulin

4

9

14 Days

19

24

sT (sT1 shRNA insensitive) Vector

Relative ratio shRNA infected/uninfected cells

Scr sT1 sT2

sT sT sT1 shRNA sT2 shRNA insensitive insensitive Scr sT1 sT2

shRNA

Vector Scr sT1 sT2

Ectopic expression

24

sT (sT2 shRNA insensitive)

1.2 1

sT2 shRNA

0.8 0.6 0.4 0.2

sT1 shRNA

0 4

9

14 Days

19

24

4

9

14 Days

19

24

Figure 2. No rescue of small T antigen (sT) short hairpin RNA (shRNA)-induced growth inhibition by ectopic expression of Merkel cell polyomavirus (MCV) sT. MKL-1 and WaGa cells were stably transduced with empty vector or with expression constructs for V5-tagged MCV sT. The sT messenger RNAs encoded by these constructs were rendered insensitive to either sT1 shRNA or sT2 shRNA by introducing six silent mutations in the respective shRNA target sequences. Subsequently, the cells carrying these constructs were infected with the indicated lentiviral shRNA expression constructs coding also for GFP. Left side: Expression of ectopic sT (anti-V5 antibody), MCV large T antigen (LT; antibody: CM2B4), and tubulin was measured by immunoblot in cell lysates harvested 5 days after shRNA infection. Right side: The ratios of a mixed population of green fluorescent shRNA-infected cells with uninfected, non-fluorescent cells were determined over time. Mean values (± SD) of three independent experiments are depicted.

sT (2T2)

CM8E6

LT (2T2)

2T2

293T pCDH-TA CM5E1

LT (CM2B4)

TA shRNA

Scr shRNA

TA shRNA

Scr shRNA

WaGa

LT

LT (CM8E6) sT

sT (CM5E1)

Tubulin

sT (CM8E6)

Figure 3. Expression level of Merkel cell polyomavirus (MCV) small T antigen (sT) in Merkel cell carcinoma (MCC) cells is very low compared with MCV large T antigen (LT). Total cell lysates of WaGa cells infected either with Scrambled (Scr) or T antigen (TA) short hairpin RNA (shRNA) or cell lysates of 293T cells transiently transfected with an MCV TA expression construct were analyzed by immunoblot. The LT-specific antibody CM2B4, the sT-specific antibody CM5E1, and the two antibodies targeting epitopes present in sT and LT (2T2; CM8E6) were applied. The gray arrowheads indicate the region where not visible sT bands would be expected.

retroviral vector pIH containing a hygromycin-resistance gene (Houben et al., 2012). The TA gene derived from MCC339 (accession number EU375804) inserted in the pcDNA6 vector served as 2062 Journal of Investigative Dermatology (2013), Volume 133

template for PCR amplification of an sT cDNA. For cloning of LT, we used cDNA generated from messenger RNA derived from 293T cells transfected with a MCC339 derived TA gene construct carrying a stop codon mutation (C1461T according to accession number EU375803) truncating the protein after amino acid 278. sT primers to generate the appropriate restriction sites for cloning into pIH were designed to add a C-terminal V5 tag to the sT cDNA. To render ectopically expressed mRNAs insensitive to a specific shRNA silent mutations in the respective shRNA target sequence were introduced using the quick change mutagenesis kit (Stratagene, La Jolla, CA). The TAs shRNA target sequence was modified by changing the sequence from ATCCACAAGCTCAGAAGTG ACTTCTCTAT to ATTCATAAACTCAGGAGCGACTTCTCGAT. The sT1 and the sT2 shRNA target sequences were modified by changing the sequence from 50 -GCTTAAGCAACTTAGAGATTCTA-30 to 50 -GCTCAAACAGCTTAGGGACTCAA-30 and from 50 -GGAAG AATATGGAACTTTA-30 to 50 -GGAGGAGTACGGTACCCTT-30 , respectively. For the transformation assays, the respective inserts were sub-cloned into the lentiviral expression vector pCDH coding for GFP as well (System Biosciences, Mountain View, CA).

Retro- and lentiviral infection Retro- and lentivirus containing supernatants were generated by transient transfection of HEK293T cells using two (pHIT60 and pHIT456) and three (pRSV rev, pHCMV-G and pMDLg/pRRE) helper constructs, respectively. Two days following transfection, virus supernatants were harvested. Following filtration through

S Angermeyer et al. MCV Small T Antigen in Merkel Cell Carcinoma

LT

LT

Tubulin

Tubulin

Tubulin WaGa

TA shRNA

1.2

MKL-2

TA shRNA

TA shRNA

1.2

1.2

1

1

1

0.8

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.6 0.4

Vector TA

0.2

LT

0

0

0 4

9

14 Days

19

24

LT

LT

MKL-1

Relative ratio shRNA infected/uninfected cells

shRNA

scr TA scr TA scr TA

shRNA

Ectopic Vector TA expression

LT

scr TA scr TA scr TA

shRNA

Ectopic Vector TA expression

LT

scr TA scr TA scr TA

Ectopic Vector TA expression

4

9

14 Days

19

4

9

14 Days

19

24

Figure 4. T antigen (TA) knockdown is rescued by re-expression of large T antigen (LT). The indicated Merkel cell carcinoma (MCC) cell lines were stably transduced with empty vector or with expression constructs containing either a truncated Merkel cell polyomavirus (MCV) LT complementary DNA (stop codon 279) or MCV TA gene coding for sT and LTstop279. Owing to introduction of six silent mutations in the short hairpin RNA (shRNA) target sequence, the messenger RNAs encoded by these constructs are insensitive to the TA shRNA. Subsequently, the cells stably carrying these constructs were infected with a lentiviral TA shRNA construct or with a Scrambled (Scr) shRNA control. (a) Immunoblot for LT of total cell lysates obtained 5 days after infection. (b) The ratios of a mixed population of green fluorescent TA shRNA-infected cells with uninfected, non-fluorescent cells were determined over time. Mean values (± SD) of three independent experiments are depicted. Growth of the corresponding Scr shRNA controls was not affected (data not shown).

Vector

Vector

BRAFV600E

SV40 ER

MCV sT

MCV sT

MCV LT

SV40 ER

MCV sT

MCV LT

Figure 5. Weak transforming activity of Merkel cell polyomavirus (MCV) small T antigen (sT) but not truncated MCV large T antigen (LT) in NIH3T3 cells. NIH3T3 cells were infected with lentiviral expression constructs for the indicated proteins. The MCV LT complementary DNA carried a stop codon mutation at position 279. Focus formation was monitored 25 days after infection. (a) Crystal violet staining; bar ¼ 10 mm. (b) Phase contrast microscopy. ER, early region. Bar ¼ 50 mm.

0.45-mm pore size filters and supplementation with 1 mg ml  1 polybrene, the virus containing supernatant was added to the target cells. Following overnight incubation, cells were washed twice with medium. As KH1 also encodes GFP the infection rate could be determined by flow cytometry (FACSCanto; BD Biosciences, Heidelberg, Germany).

GFP assay GFP expression by KH1-infected cells was used to compare the behavior of infected and uninfected cells: infected cells were mixed with approximately 20% of uninfected cells. Changes in the frequency of GFP-positive cells in this mixture were determined by flow cytometry over time. www.jidonline.org 2063

S Angermeyer et al. MCV Small T Antigen in Merkel Cell Carcinoma

Immunoblotting Cells were lysed in 0.6% SDS, 1 mM EDTA, 10 mM Tris-HCl (pH 8.0), 2 mM NaF, and 2 mM NaVO3 supplemented with a protease inhibitor cocktail (Roche Diagnostics, Basel, Switzerland). Samples were resolved by SDS-PAGE and transferred to nitrocellulose membranes. Following 1 h blocking with phosphate-buffered saline (PBS) containing 0.05% Tween-20 and 5% powdered skim milk, membranes were incubated overnight with primary antibody, washed three times with PBS with 0.05% Tween-20 (PBS/Tween), and then incubated with a peroxidase-coupled secondary antibody. Following three washes with PBS/Tween, bands were detected using a chemiluminescence detection kit (Thermo Fisher Scientific, Rockford, IL). The antibodies used in this study were directed against MCV LT (CM2B4; Santa Cruz Biotechnology, Santa Cruz, CA), MCV sT (CM5E1; gift from Patrick Moore), MCV TA (2T2; gift from Christopher Buck, and CM8E6;

gift from Patrick Moore), the V5 tag (Abcam, Cambridge, MA), or b-tubulin (Sigma-Aldrich, Taufkirchen, Germany). Transformation assay A total of 2  105 NIH3T3 cells were seeded in 6-well plates and TA expression constructs were stably introduced by infection with lentiviral supernatants generated in HEK293T cells. One day after infection, cells were transferred to T75 flasks and on day four, infection rates determined by flow cytometry analysis varied between 30 and 70%. On day 25 after infection, focus formation was monitored by microscopy and crystal violet staining. CONFLICT OF INTEREST

Hodgson NC (2005) Merkel cell carcinoma: changing incidence trends. J Surg Oncol 89:1–4 Houben R, Adam C, Baeurle A et al. (2012) An intact retinoblastoma proteinbinding site in Merkel cell polyomavirus large T antigen is required for promoting growth of Merkel cell carcinoma cells. Int J Cancer 130: 847–56 Houben R, Becker JC, Kappel A et al. (2004) Constitutive activation of the RasRaf signaling pathway in metastatic melanoma is associated with poor prognosis. J Carcinog 3:6 Houben R, Shuda M, Weinkam R et al. (2010) Merkel cell polyomavirusinfected Merkel cell carcinoma cells require expression of viral T antigens. J Virol 84:7064–72 Jackson AL, Burchard J, Schelter J et al. (2006) Widespread siRNA "off-target" transcript silencing mediated by seed region sequence complementarity. RNA 12:1179–87 Kassem A, Schopflin A, Diaz C et al. (2008) Frequent detection of Merkel cell polyomavirus in human Merkel cell carcinomas and identification of a unique deletion in the VP1 gene. Cancer Res 68:5009–13 Leonard JH, Dash P, Holland P et al. (1995) Characterisation of four Merkel cell carcinoma adherent cell lines. Int J Cancer 60:100–7 Madkan VK, Cook-Norris RH, Steadman MC et al. (2007) The oncogenic potential of human papillomaviruses: a review on the role of host genetics and environmental cofactors. Br J Dermatol 157:228–41 Martel-Jantin C, Filippone C, Cassar O et al. (2012) Genetic variability and integration of Merkel cell polyomavirus in Merkel cell carcinoma. Virology 426:134–42 Matsushita M, Kuwamoto S, Iwasaki T et al. (2012) Detection of Merkel Cell Polyomavirus in the human tissues from 41 Japanese autopsy cases using polymerase chain reaction. Intervirology 56:1–5 Mesri EA, Cesarman E, Boshoff C (2010) Kaposi’s sarcoma and its associated herpesvirus. Nat Rev Cancer 10:707–19

Ju¨rgen C Becker works as consultant for Novartis and Leo Pharma. The remaining authors state no conflict of interest.

Pastrana DV, Tolstov YL, Becker JC et al. (2009) Quantitation of human seroresponsiveness to merkel cell polyomavirus. PLoS Pathog 5: e1000578

ACKNOWLEDGMENTS

Rodig SJ, Cheng J, Wardzala J et al. (2012) Improved detection suggests all Merkel cell carcinomas harbor Merkel polyomavirus. J Clin Invest 122:4645–53

We thank Christopher Buck and Patrick Moore for supplying antibodies. This study was supported by the Wilhem-Sander-Stiftung (2007.057.3) and the IZKF Wu¨rzburg (B-157).

REFERENCES Ali SH, DeCaprio JA (2001) Cellular transformation by SV40 large T antigen: interaction with host proteins. Semin Cancer Biol 11:15–23 Becker JC (2010) Merkel cell carcinoma. Ann Oncol 21:vii81–5 Becker JC, Houben R, Ugurel S et al. (2009a) MC polyomavirus is frequently present in Merkel cell carcinoma of European patients. J Invest Dermatol 129:248–50 Becker JC, Schrama D, Houben R (2009b) Merkel cell carcinoma. Cell Mol Life Sci 66:1–8 Boccardo E, Lepique AP, Villa LL (2010) The role of inflammation in HPV carcinogenesis. Carcinogenesis 31:1905–12 Feng H, Shuda M, Chang Y et al. (2008) Clonal integration of a polyomavirus in human Merkel cell carcinoma. Science 319:1096–100 Garneski KM, Warcola AH, Feng Q et al. (2009) Merkel cell polyomavirus is more frequently present in North American than Australian Merkel cell carcinoma tumors. J Invest Dermatol 129:246–8 Gillis S, Watson J (1980) Biochemical and biological characterization of lymphocyte regulatory molecules. V. Identification of an interleukin 2producing human leukemia T cell line. J Exp Med 152:1709–19 Hahn WC, Dessain SK, Brooks MW et al. (2002) Enumeration of the simian virus 40 early region elements necessary for human cell transformation. Mol Cell Biol 22:2111–23

2064 Journal of Investigative Dermatology (2013), Volume 133

Rosen ST, Gould VE, Salwen HR et al. (1987) Establishment and characterization of a neuroendocrine skin carcinoma cell line. Lab Invest 56: 302–12 Schmitt M, Wieland U, Kreuter A et al. (2012) C-terminal deletions of Merkel cell polyomavirus large T-antigen, a highly specific surrogate marker for virally induced malignancy. Int J Cancer 131:2863–8 Schowalter RM, Reinhold WC, Buck CB (2012) Entry tropism of BK and Merkel Cell Polyomaviruses in cell culture. PLoS One 7:e42181 Shuda M, Arora R, Kwun HJ et al. (2009) Human Merkel cell polyomavirus infection I. MCV T antigen expression in Merkel cell carcinoma, lymphoid tissues and lymphoid tumors. Int J Cancer 125:1243–9 Shuda M, Feng H, Kwun HJ et al. (2008) T antigen mutations are a human tumor-specific signature for Merkel cell polyomavirus. Proc Natl Acad Sci USA 105:16272–7 Shuda M, Kwun HJ, Feng H et al. (2011) Human Merkel cell polyomavirus small T antigen is an oncoprotein targeting the 4E-BP1 translation regulator. J Clin Invest 121:3623–34 Van Gele M, Leonard JH, Van RN et al. (2002) Combined karyotyping, CGH and M-FISH analysis allows detailed characterization of unidentified chromosomal rearrangements in Merkel cell carcinoma. Int J Cancer 101:137–45 Verhaegen M, Bauer JA, Martin de la Vega C et al. (2006) A novel BH3 mimetic reveals a mitogen-activated protein kinase-dependent mechanism of melanoma cell death controlled by p53 and reactive oxygen species. Cancer Res 66:11348–59

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.