Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Tumour angiogenesis in Epstein-Barr virus-associated post-transplant smooth muscle tumours

Tumour angiogenesis in Epstein-Barr virus-associated post-transplant smooth muscle tumours Epstein-Barr virus (EBV)-associated post-transplant smooth muscle tumours (PTSMT), are rare complications following organ/stem cell transplantation. Despite the mainly benign behaviour of PTSMT, alternative therapies are needed for those patients with progressive tumours. In tumours not approachable by surgery or reduction of immunosuppression, the angiogenic microenvironment might be a potential target of therapy, an approach that is well utilised in other soft tissue neoplasms. In a previous study, we evaluated the expression of EBV-related genes and the microRNA profile in PTSMT, but so far the characteristics of angiogenesis in PTSMT are not known. Therefore, the aim of this study was to evaluate the expression pattern of angiogenesis-related genes in PTSMT, in order to identify potential target molecules for anti-angiogenic therapy. PTSMT (n = 5 tumours) were compared with uterine leiomyomas (n = 7). Analyses included real-time PCR of 45 angiogenesis-associated genes, immunohistochemistry (CD31, prostaglandin endoperoxide synthase 1/PTGS1) and assessment of tumour vascularisation by conventional histopathology. PTSMT showed similar or fewer vessels than leiomyomas. Of the genes under investigation, 23 were down-deregulated (pro-angiogenic and some anti-angiogenic factors) and five were up-regulated (e.g. PTGS1 which is expressed at very low levels in leiomyomas but moderately higher levels in PTSMT). In summary, no particular target molecule could be identified, because tumour angiogenesis in PTSMT is characterised by low levels of major pro-angiogenic factors and there is no prominent increase in tumour vascularisation. EBV can induce angiogenesis via its viral late membrane protein 1 (LMP1) but PTSMT frequently do not express LMP1, which could be an explanation why, despite EBV infection, PTSMT show no exaggerated tumour angiogenesis. Keywords: PTSMT, Post-transplant smooth muscle tumours, EBV, Angiogenesis, Tumour Introduction previously analysed cell cycle factors, cytokines and gene Epstein-Barr virus (EBV)-associated post-transplant smooth promoter methylation in PTSMT and found an activated muscle tumours (PTSMT) are rare complications following phosphoinositide 3-kinase (PI3K)/mammalian target of solid graft and stem cell transplantation [1]. The molecular rapamycin (mTOR) cell cycle pathway as well as ex- pathobiology of this rare neoplastic entity is not fully under- pression of vascular endothelial growth factor (VEGF) stood and only few experimental analyses have addressed and Fms-related tyrosine kinase 1 (FLT1/VEGFR1) [1]. this issue [1,2]. Tumour cells are thought to be derived from In general, in addition to endogenous molecular de- aberrant myogenous venous/perivascular wall cells [3]. They fectswhich affect mitosisand apoptosisofthe tumour express smooth muscle proteins (actin and desmin), but not cells, angiogenesis is a major mechanism which con- CD117, CD34 or other endothelial marker proteins. Histo- tributes to tumour cell survival by supplying the me- morphology is characterised by mild atypia, low mitotic rate tabolism of aberrant cell proliferation. Currently, for and absence of prominent tumour necrosis. All in all, PTSMT, surgery and reduced immunosuppression are PTSMT show more histological features of benign leiomyo- the therapy of choice [1]. At this point, there is no mas rather than leiomyosarcomas [1,2] and our group has proof that patients benefit from conventional chemo- therapy or radiation alone [1]. In other soft tissue neo- plasms, numerous studies have addressed the angiogenic * Correspondence: Hussein.Kais@MH-Hannover.de Institute of Pathology, Hannover Medical School (MHH), Carl-Neuberg-Str. 1, microenvironment as a potential target of therapy. In D-30625 Hanover, Germany © 2014 Jonigk et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 2 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 PTSMT, angiogenesis might be of special importance, as values (normalised to a mean of endogenous control the original/progenitor tumour cell in these neoplasms is genes). Statistical analysis was performed with Prism 5.0 generally thought to be derived from an aberrant perivas- (GraphPad Software, San Diego, CA, USA) by applying cular/venous wall cell. This topic is also important in the non-parametric Kruskal-Wallis test followed by the PTSMT, as these can manifest in any anatomical localisa- Mann-Whitney test for two-group comparison. P values < tion and cerebral tumours are in particular associated with 0.05 were considered as statistically significant. a poor prognosis [1]. From other tumours, in particular renal cell cancer, we Immunohistochemistry for evaluation of selected genes know that hypoxia-inducible factor 1, alpha subunit (HIF1A) Deparaffinised and rehydrated FFPE tissue sections signalling mediates expression of VEGF, platelet-derived (1-2 μm) were stained after autoclave pre-treatment. For growth factor (PDGF) and angiopoietin via the PI3K/mTOR staining of platelet/endothelial cell adhesion molecule 1 pathway [4,5]. These cytokines activate pro-angiogenic re- (VCAM1/CD31), sections were processed in an auto- ceptors such as VEGFR and PDGF receptors (PDGFR). For mated staining system (Benchmark ULTRA, Ventana a variety of neoplasms, e.g. soft tissue sarcomas such Medical Systems, Inc., Tucson, AZ, USA). Prostaglandin as leiomyosarcomas, it has been shown that a VEGFR/ endoperoxide synthase 1 (prostaglandin G/H synthase PDGFR-mediated increase of angiogenesis can be inhib- and cyclooxygenase) (PTGS1) was stained manually ited by anti-angiogenic agents [6-9]. The aim of this (positive control: FFPE prostate cancer tissue). Mouse analysis was to evaluate the expression pattern of monoclonal antibodies were used. Vascularisation was angiogenesis-related genes in PTSMT, in order to quantified by counting CD31 vessels per 10 high identify potential target molecules for anti-angiogenic power fields (HPF) and then correlating them in seri- therapy, in particular for those patients who suffer ally cut haematoxylin-eosin-stained sections. Statistical from irresectable or progressive tumours. analysis was performed with Prism 5.0 as described above. Material and methods Tissue specimens Five EBV PTSMT samples from four patients, including Results two tumours from one patient (#4), and seven EBV be- Vascularisation of PTSMT nign uterine leiomyomas from solid graft recipients were As previously described, PTSMT tumour cells them- analysed. These cases had been characterised earlier selves were negative for CD31. In the cerebral PTSMT (Additional file 1: Table S1) [1]. Formalin-fixed and we could previously demonstrate aneuploidy of the MYC paraffin-embedded (FFPE) samples were retrieved from locus 8q24 by fluorescence in situ hybridisation (FISH) [1]. the archives of the Institute of Pathology (Hannover In this case, endothelial cells showed a normal MYC con- Medical School/MHH, Hanover, Germany). The retro- figuration. Thus, a clonal relation between PTSMT and spective evaluation has been approved by the local eth- endothelial cells could not be proven (Additional file 1: ics committee (MHH). Figure S1). PTSMT showed similar or fewer vessels than leiomyo- Expression analysis of angiogenesis-associated factors mas (mean 301/range 201-518 versus mean 511/range Tissue from FFPE blocks with >90% tumour cells were 306-789 CD31 vessels/10 HPF, p = 0.0480; Figure 1). cut and processed for further PCR analysis. In blocks Corresponding to the low significance level, there was a with <90% aberrant neoplastic cells, the PTSMT compart- broad overlap in vessel density between these two leio- ments of the specimens were laser microdissected using a myomatous tumour entities. Furthermore, gene expres- SmartCutPlus-System (MMI, Glattbrugg, Switzerland), as sion analysis of CD31 did not correlate with vessel previously described [1,10]. Cells were digested in protein- density. Higher rather than lower expression levels of ase K and RNA was extracted with phenol/chloroform CD31 were detectable in PTSMT (mean 20.10/range [1,10]. Synthesis of cDNA from mRNA, subsequent pre- 5.26-30.48 in PTSMT versus mean 6.76/range 2.44-11.40 amplification of cDNA and real-time quantitative PCR of in leiomyomas; p = 0.0303). Sinusoids without smooth 45 angiogenesis-associated genes and three endogenous muscle cell wall appeared generally smaller in PTSMT controls with a 7900HT Fast Real-Time PCR system were and more hyalinised but, in comparison to leiomyomas performed according to the manufacturers’ instructions the quantitative difference was not significant. PTSMT (Applied Biosystems, Carlsbad, CA, USA). Endogenous had significantly fewer arterioles, as defined by vessels controls were polymerase (RNA) II (DNA-directed) with a smooth muscle wall (mean 1 versus 15 vessels/10 polypeptide A, 220 kDa (POLR2), glucuronidase beta HPF, p = 0.0058). In summary, there was no clear evi- (GUSB) and glyceraldehyde-3-phosphate dehydrogenase dence that PTSMT are generally more vascularised than -ΔCT (GAPDH). Delta C values were converted into 2 leiomyomas. T Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 3 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Figure 1 Similar vascularisation of PTSMT and leiomyomas. A) Histological counts of vessels. B) CD31+ vessels in PTSMT (original magnification x200). C) CD31+ vascularisation in leiomyomas (x200). D) Weak PTGS1 protein expression in PTSMT (x400). E) Very weak protein expression in a leiomyoma (x400). Reduced expression of angiogenesis-associated genes in Except for CD31, significant differences of other up- PTSMT regulated factors were due to very low expression in Among 45 angiogenesis-associated mediators under in- leiomyomas rather than strong expression in PTSMT. vestigation, 28 were significantly deregulated in PTSMT: These factors were angiopoietin 2 (ANGPT2), PDGFRA, 23 were down-deregulated and 5 (including CD31, see PTGS1 and thymidine phosphorylase (TYMP). Because above) were up-regulated (Figure 2, Table 1). PTGS1 can be inhibited by widely used non-steroidal Prominently down-regulated factors included e.g. pro- anti-inflammatory drugs, immunohistochemistry was angiogenic HIF1A, fibroblast growth factor receptor 1 performed for evaluation if the tumour cells showed a (FGFR1/FLT2), kinase insert domain receptor (VEGFR2/ corresponding protein expression. A weak expression KDR) and VEGFA as well as anti-angiogenic serpin peptid- of PTGS1 proteins in PTSMT and leiomyomatous ase inhibitor, clade E (nexin, plasminogen activator inhibi- smooth muscle spindle cells was detectable (Figure 1). tor type 1), member 1 (SERPINE1), thrombospondin 1 Weak protein expression corresponded with relatively (THBS1) and TIMP metallopeptidase inhibitor 2 (TIMP2). low transcript expression levels in both tumour types Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 4 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Figure 2 Representative summary of up- and down-regulated angiogenesis-associated factors in PTSMT. (mean 0.41/range 0.09-0.94 in PTSMT versus mean the expression profiles of angiogenesis-related factors in 0.03/range 0.00-0.08 in leiomyomas; p = 0.0061). PTSMT. However, in contrast to this assumption we found almost the opposite: PTSMT showed similar or Discussion even reduced vascularisation, when compared to sporadic Patients suffering from PTSMT benefit from surgical leiomyomas. Furthermore, we could show that this mor- tumour resection and/or reduction of immunosuppres- phological feature was based on a previously unknown sion [1]. However, surgical respectability depends on molecular characteristic of PTSMT, namely expression of tumour site and, of note, PTSMT can manifest at any lo- low levels of pro-angiogenic factors and high levels of calisation, including the transplanted organ, in particular anti-angiogenic genes. In particular major factors of liver grafts [1]. Furthermore, multiple PTSMT, e.g. in hypoxia-inducible angiogenesis such as HIF1A, VEGFA, the lung, are not suitable for a surgical approach [11]. VEGFC, VEGFR1/FLT1, VEGFR2/KDR and FGFR1/FLT2 Due to the rarity of this tumour entity, prospective eval- were expressed at low levels. In contrast to PTSMT, leio- uations of therapeutic strategies will not be applicable in myosarcomas show generally higher expression of VEGFA a considerable number of patients. However, additional than leiomyomas [15-17]. In leiomyosarcoma-derived cell therapy options are mandatory for those patients who lines it could be demonstrated that hepatocyte growth fac- cannot be operated and/or whose transplant organ does tor (HGF) induces a decrease in anti-angiogeneic THBS1 not tolerate reduction of immunosuppression. In indi- and an increase in VEGFA [18]. In PTSMT, HGF, THBS1 vidual patients, it has been shown that inhibition of and VEGFA are all expressed at low levels, indicating mTOR signal pathways by sirolimus might be of thera- that HGF signalling does not contribute significantly peutic benefit [12-14]. The rationale for administration to tumour angiogenesis. In PTSMT, low levels were of an mTOR signalling inhibitor was based on the find- also detectable for other pro-angiogenic genes which ing that PTSMT and HIV-associated SMT, which share are involved in differentiation and proliferation of endo- morphological similarities with PTSMT, express mTOR thelial cells, e.g. vascular development-related EPH recep- [2]. However, sirolimus cannot be administered to all tor B4 (EPHB4) and sphingosine-1-phosphate receptor 1 transplanted patients, e.g. after renal transplantation, (S1PR1), the endothelium-specific receptor tyrosine kinase because the drug is potentially nephrotoxic. Another TEK and the growth factor midkine (MDK). Immune response-associated caveolae are plasma membrane invag- class of drugs which is widely used for systemic ther- apy of soft tissue neoplasms/sarcomas are anti-angiogenic inations of 60-80 nm in diameter in endothelial cells, agents, e.g. leiomyosarcoma [8]. Basic analysis of tumour- smooth muscle cells and other cell types [19] and caveolae components CAV2 (p = 0.07) and PTRF (p < 0.01) were associated angiogenesis is important for assessing the vulnerability of a given tumour type to these drugs. both decreased in PTSMT. In addition to several blood Prominent proliferation of vessels, high expression levels vessel-associated factors, lymphatic vessel protein podo- planin was decreased in PTSMT. Again, in leiomyosarco- of pro-angiogenic and low levels of anti-angiogenic genes would make it likely that PTSMT patients could benefit mas, podoplanin-positive vessels are especially found in from anti-angiogenic drug therapy. Therefore, we evaluated tumours with lymph node metastases [20]. In our cohort, Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 5 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Table 1 Expression profile of angiogenesis-related genes Gene (abbreviation) Gene name PTSMT (mean) Leiomyomas (mean) Significance ANGPT1 Angiopoietin 1 0.03 0.51 n.s. ANGPT2 Angiopoietin 2 0.91 0.31 p = 0.0303 ANGPTL1 Angiopoietin-like 1 0.05 0.05 n.s. ANGPTL4 Angiopoietin-like 4 0.03 0.05 n.s. BAI1 Brain-specific angiogenesis inhibitor 1 0.06 0.18 n.s. CAV2 Caveolin 2 0.84 2.35 n.s. CDK1 Cyclin-dependent kinase 1 0.01 0.01 n.s. CXCL1 Chemokine (C-X-C motif) ligand 1 (melanoma 4.25 7.55 n.s. growth stimulating activity, alpha) EPHB4 EPH receptor B4 0.06 2.50 p = 0.0025 FGF2 Fibroblast growth factor 2 (basic) 0.01 0.16 p = 0.0121 FGFR1/FLT2 Fibroblast growth factor receptor 1 3.36 21.33 p = 0.0051 GREM1 Gremlin 1, DAN family BMP antagonist 0.14 0.20 n.s. HGF Hepatocyte growth factor 0.02 0.07 p = 0.0242 HIF1A Hypoxia inducible factor 1, alpha subunit 6.48 13.81 p = 0.0051 IL8 Interleukin 8 0.60 0.03 n.s. LECT1 Leukocyte cell derived chemotaxin 1 0.11 Not detectable – MDK Midkine (neurite growth-promoting factor 2) 0.14 1.24 p = 0.0061 MMP2 Matrix metallopeptidase 2 (gelatinase A, 72 kDa 1.19 34.07 p = 0.0025 gelatinase, 72 kDa type IV collagenase) NOS3 Nitric oxide synthase 3 (endothelial cell) 0.05 0.31 p = 0.0025 NOTCH4 Notch 4 (receptor) 0.28 2.68 p = 0.0025 NOX4 NADPH oxidase 4 0.05 1.32 p = 0.0061 NRP1 Neuropilin 1 0.17 0.82 n.s. OLR1 Oxidized low density lipoprotein (lectin-like) receptor 1 0.04 0.02 n.s. PDGFA Platelet-derived growth factor alpha polypeptide 0.36 0.12 p = 0.0480 PDGFRA Platelet-derived growth factor receptor, alpha 1.66 7.31 n.s. polypeptide PDPN Podoplanin 0.09 3.36 p = 0.0061 PECAM1/CD31 Platelet/endothelial cell adhesion molecule 20.10 6.76 p = 0.0303 PTGS1 Prostaglandin-endoperoxide synthase 1 (prostaglandin 0.41 0.03 p = 0.0061 G/H synthase and cyclooxygenase) PTRF Polymerase I and transcript release factor 7.68 56.80 p = 0.0025 S1PR1 Sphingosine-1-phosphate receptor 1 0.15 0.76 p = 0.0172 SERPINF1 Serpin peptidase inhibitor, clade F (alpha-2 antiplasmin, 0.19 6.45 p = 0.0061 pigment epithelium derived factor), member 1 STAB1 Stabilin 1 0.06 0.16 n.s. TEK TEK tyrosine kinase, endothelial 0.11 1.17 p = 0.0043 TGFBR1 Transforming growth factor, beta receptor 1 0.05 0.38 p = 0.0025 THBS1 Thrombospondin 1 13.56 134.17 p = 0.0025 TIMP2 TIMP metallopeptidase inhibitor 2 9.32 54.78 p = 0.0025 TNFAIP2 Tumor necrosis factor, alpha-induced protein 2 0.08 0.01 n.s. TYMP Thymidine phosphorylase 0.17 0.08 p = 0.0480 VCAM1 Vascular cell adhesion molecule 1 0.18 0.95 p = 0.0303 VEGFA Vascular endothelial growth factor A 3.38 36.05 p = 0.0025 VEGFC Vascular endothelial growth factor C 0.47 1.66 p = 0.0043 Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 6 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Table 1 Expression profile of angiogenesis-related genes (Continued) VEGFD/FIGF C-fos induced growth factor (vascular endothelial 0.01 0.23 n.s. growth factor D) VEGFR1/FLT1 Fms-related tyrosine kinase 1 (vascular endothelial 0.15 0.73 p = 0.0061 growth factor/vascular permeability factor receptor) VEGFR2/KDR Kinase insert domain receptor (a type III receptor 0.26 3.48 p = 0.0061 tyrosine kinase) VEGFR3/FLT4 Fms-related tyrosine kinase 4 0.08 0.10 n.s. none of the PTSMT manifested in lymph nodes and, in cells in vitro and in vivo [30], in PTSMT increased MYC general, involvement of lymph nodes is rare in this type of expression is associated with decreased THBS1 expression transplant-associated neoplasm [1]. MMP2, which de- but there is no indication for a specific microRNA regula- grades the collagen IV-rich basal membrane as a necessary tion. Furthermore, while in leiomyosarcomas low expres- requisite for metastasis [21], was reduced in PTSMT, sion of THBS1 and TIMP2 is accompanied by increased which indicates no major remodelling of extracellular expression of pro-angiogenic factors such as VEGFA, matrix during tumour cell and endothelial proliferation. PTSMT in general did not show such a global pro- Compared to leiomyomas, only a few pro-angiogenic angiogenic expression profile. factors such as TYMP, ANGPTL2 and PTGS1 were in- As reviewed by Paydas [31], in lymphomas and naso- creased in PTSMT. However, statistical significances were pharyngeal carcinomas, tumour cell infection with EBV the result of very low expression levels in leiomyomas ra- is related to increased angiogenesis, in particular because ther than a prominent up-regulation in PTSMT. The the viral late membrane protein 1 (LMP1) induces ex- mean relative expression levels of these three factors pression of VEGF and activation of PTGS2, interleukin 8 was <1, indicating no major role in mediating tumour (IL8), fibroblast growth factor 2 (FGF2) and other pro- angiogenesis. angiogenic factors. Although PTSMT are infected with In PTSMT, three important anti-angiogenetic factors EBV, these tumours do not usually express LMP1 pro- were decreased: TIMP2, SERPINF1 and THBS1. TIMP2 teins [1,2,32] and this could be an explanation why, and SERPINF1 are strong inhibitors of endothelial pro- despite viral infection, PTSMT show no exaggerated liferation [8,22] and THBS1 induces reduced migration tumour angiogenesis. It is not known how the EBV+ ability of endothelial cells [23]. Furthermore, THBS1 can tumour cells suppress the expression of LMP1 while inhibit the binding of activating cytokines at receptors of expressing other viral proteins and the tumour bio- endothelial cells and can also bind to the thrombospon- logical benefit of this selective lack of LMP1 for the din receptor CD36 which induces endothelial apoptosis PTSMT proliferation is also not known. [23]. Other groups found that leiomyomas express Although clinical testing has not yet been performed, THBS1 more frequently than leiomyosarcomas [24]. In on the one hand it is questionable whether patients who addition, TIMP2 is also expressed at relatively low levels suffer from this type of soft tissue tumour might benefit in leiomyosarcomas [22]. from systematic anti-angiogenic drug therapy. On the It has been shown that the transcription factor MYC other hand, it could be assumed that PTSMT found their leads to expression of the chromosome segment own equilibrium of tumour vascularisation that allows 13q31.3-encoded microRNA 17~92 cluster which in- survival and growth without increasing the expression of cludes the two paralogues miR-19a and miR-19b-1 pro-angiogenic factors (e.g. due to aberrant signalling [25-27]. MicroRNA are non-coding molecules of 20- downstream of pro-angiogenic receptors). This might principally indicate a limited ability to circumvent therapy 25 nucleotides which bind to mRNA and negatively regulate protein translation [28]. THBS1-mRNA has and therefore anti-angiogenic drugs might not necessarily a miR-19 binding site and therefore MYC-related be ineffective since this would disrupt the equilibrium of PTSMT vascularisation. Anti-angiogenic drugs could still miR-19 expression down-regulates THBS1 [25-27]. PTSMT have an increased MYC expression [1] and be administered to PTSMT patients with no other treat- low levels of THBS1 but no up-regulation of the miR ment options available but, in these present analyses, we could not identify a specific target molecule. 17~92 cluster, including miR-19a (in PTSMT mean relative expression level 0.02 versus 0.03 in leiomyo- In summary, our analyses of the tumour angiogenesis mas) and miR-19b (mean 1.63 in PTSMT versus 2.23 in PTSMT revealed no particular target molecule, be- cause PTSMT are characterised by low levels of major in leiomyomas) [29]. The microRNA profile in PTSMT is overall associated with leiomyomatous differentiation of pro-angiogenic factors and there is no prominent in- the tumour cells [29]. Therefore, similar to mesenchymal crease in tumour vascularisation. Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 7 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Additional file of overexpression in leiomyosarcoma. J Cancer Res Clin Oncol 2004, 130:52–56. 17. Sanci M, Dikis C, Inan S, Turkoz E, Dicle N, Ispahi C: Immunolocalization of Additional file 1: Table S1. Patient cohort. Figure S1. Fluorescence in VEGF, VEGF receptors, EGF-R and Ki-67 in leiomyoma, cellular leiomyoma situ hybridisation of the MYC gene shows no aneuploidy in endothelial and leiomyosarcoma. Acta Histochem 2011, 113:317–325. cells (white arrows) while a subfraction of PTSMT cells had an aneuploidy. 18. Zhang YW, Su Y, Volpert OV, Vande Woude GF: Hepatocyte growth factor/ scatter factor mediates angiogenesis through positive VEGF and negative thrombospondin 1 regulation. Proc Natl Acad Sci U S A 2003, Competing interests 100:12718–12723. The authors of this manuscript have no conflicts of interests to disclose. 19. Feng H, Guo W, Han J, Li XA: Role of caveolin-1 and caveolae signaling in endotoxemia and sepsis. Life Sci 2013. doi:10.1016/j.lfs.2013.05.016. Authors’ contributions 20. Lamyman MJ, Giele HP, Critchley P, Whitwell D, Gibbons M, Athanasou NA: Histomorphology (DJ, KH, HK), molecular analysis (LM, NI, ES), data collection, Local recurrence and assessment of sentinel lymph node biopsy in deep analysis of data and manuscript preparation (DJ, LM, NI, ES, HK, KH). All soft tissue leiomyosarcoma of the extremities. Clin Sarcoma Res 2011, 1:7. authors read and approved the final manuscript. 21. Bhuvarahamurthy V, Kristiansen GO, Johannsen M, et al: In situ gene expression and localization of metalloproteinases MMP1, MMP2, MMP3, Acknowledgements MMP9, and their inhibitors TIMP1 and TIMP2 in human renal cell The authors thank Regina Engelhardt for excellent technical support and carcinoma. Oncol Rep 2006, 15:1379–1384. Gillian Teicke for editing the text. 22. Chuang TD, Panda H, Luo X, Chegini N: miR-200c is aberrantly expressed Funding (DJ): “Deutsche Forschungsgemeinschaft” (DFG, GZ: 30 743/2-1). in leiomyomas in an ethnic-dependent manner and targets ZEBs, VEGFA, TIMP2, and FBLN5. Endocr Relat Cancer 2012, 19:541–556. Received: 29 August 2013 Accepted: 1 November 2013 23. Fernando NT, Koch M, Rothrock C, et al: Tumor escape from endogenous, Published: 7 January 2014 extracellular matrix-associated angiogenesis inhibitors by up-regulation of multiple proangiogenic factors. Clin Cancer Res 2008, 14:1529–39. References 24. Bodner-Adler B, Nather A, Bodner K, et al: Expression of thrombospondin 1 1. Jonigk D, Laenger F, Maegel L, et al: Molecular and clinicopathological (TSP 1) in patients with uterine smooth muscle tumors: an analysis of Epstein-Barr virus-associated posttransplant smooth muscle immunohistochemical study. Gynecol Oncol 2006, 103:186–189. tumors. Am J Transplant 2012, 12:1908–1917. 25. Italiano A, Thomas R, Breen M, et al: The miR-17-92 cluster and its target 2. Ong KW, Teo M, Lee V, et al: Expression of EBV latent antigens, THBS1 are differentially expressed in angiosarcomas dependent on MYC mammalian target of rapamycin, and tumor suppression genes in amplification. Genes Chromosomes Cancer 2012, 51:569–578. EBV-positive smooth muscle tumors: clinical and therapeutic implications. 26. Dews M, Homayouni A, Yu D, et al: Augmentation of tumor angiogenesis Clin Cancer Res 2009, 15:5350–5358. by a myc-activated microRNA cluster. Nat Genet 2006, 38:1060–1065. 3. Chaves NJ, Kotsimbos TC, Warren MA, et al: Cranial leiomyosarcoma in an 27. Dews M, Fox JL, Hultine S, et al: The myc-miR-17 92 axis blunts TGF{beta} Epstein-Barr virus (EBV)-mismatched lung transplant recipient. J Heart signaling and production of multiple TGF{beta}-dependent antiangiogenic Lung Transplant 2007, 26:753–755. factors. Cancer Res 2010, 70:8233–8246. 4. Figlin RA, Kaufmann I, Brechbiel J: Targeting PI3K and mTORC2 in 28. Hussein K: Pathobiology of the microRNA system. Pathologe 2012, 33:70–78. metastatic renal cell carcinoma: new strategies for overcoming 29. Jonigk D, Izykowski N, Maegel L, et al: MicroRNA expression in Epstein-Barr resistance to VEGFR and mTORC1 inhibitors. Int J Cancer 2013. virus-associated post-transplant smooth muscle tumours is related to doi:10.1002/ijc.28023. leiomyomatous phenotype. Clin Sarcoma Res 2013, 3:9. 5. Iacovelli R, Palazzo A, Trenta P, et al: Management of metastatic renal cell 30. Ngo CV, Gee M, Akhtar N, et al: An in vivo function for the transforming carcinoma progressed after sunitinib or another antiangiogenic myc protein: elicitation of the angiogenic phenotype. Cell Growth Differ treatment. Am J Clin Oncol 2013. doi:10.1097/COC.0b013e31827de888. 2000, 11:201–210. 6. van der Graaf WT, Blay JY, Chawla SP, et al: Pazopanib for metastatic 31. Paydas S: Strong cross-talk between angiogenesis and EBV: do we need soft-tissue sarcoma (PALETTE): a randomised, double-blind, different treatment approaches in lymphoma cases with EBV and/or placebo-controlled phase 3 trial. Lancet 2012, 379:1879–1886. high angiogenic capacity. Med Oncol 2012, 29:2159–2165. 7. Hensley ML, Sill MW, Scribner DR Jr, et al: Sunitinib malate in the 32. Rougemont AL, Alfieri C, Fabre M, et al: Atypical Epstein-Barr virus (EBV) treatment of recurrent or persistent uterine leiomyosarcoma: a latent protein expression in EBV-associated smooth muscle tumours gynecologic oncology group phase II study. Gynecol Oncol 2009, 115:460–465. occurring in paediatric transplant recipients. Histopathology 2008, 8. Ren JG, Jie C, Talbot C: How PEDF prevents angiogenesis: a hypothesized 53:363–367. pathway. Med Hypotheses 2005, 64:74–78. 9. Ganjoo KN: New developments in targeted therapy for soft tissue doi:10.1186/2045-3329-4-1 sarcoma. Curr Oncol Rep 2010, 12:261–265. Cite this article as: Jonigk et al.: Tumour angiogenesis in Epstein-Barr virus- 10. Jonigk D, Merk M, Hussein K, et al: Obliterative airway remodeling: associated post-transplant smooth muscle tumours. Clinical Sarcoma Research molecular evidence for shared pathways in transplanted and native 2014 4:1. lungs. Am J Pathol 2011, 178:599–608. 11. Somers GR, Tesoriero AA, Hartland E, et al: Multiple leiomyosarcomas of both donor and recipient origin arising in a heart-lung transplant patient. Am J Surg Pathol 1998, 22:1423–1428. Submit your next manuscript to BioMed Central 12. Toh HC, Teo M, Ong KW, et al: Use of sirolimus for Epstein-Barr virus-positive smooth-muscle tumour. Lancet Oncol 2006, 7:955–957. and take full advantage of: 13. Belingheri M, Comoli P, Locatelli F, et al: Successful medical treatment of EBV smooth muscle tumor in a renal transplant recipient. Pediatr • Convenient online submission Transplant 2010, 14:E101–104. • Thorough peer review 14. TanCS, LohHL, FooMW, Choong LH,WongKS, KeeTY: Epstein-Barr virus-associated smooth muscle tumors after kidney transplantation: • No space constraints or color figure charges treatment and outcomes in a single center. Clin Transplant 2013. • Immediate publication on acceptance doi:10.1111/ctr.12139. • Inclusion in PubMed, CAS, Scopus and Google Scholar 15. Hong T, Shimada Y, Uchida S, et al: Expression of angiogenic factors and apoptotic factors in leiomyosarcoma and leiomyoma. Int J Mol Med 2001, • Research which is freely available for redistribution 8:141–148. 16. Potti A, Ganti AK, Tendulkar K, et al: Determination of vascular endothelial Submit your manuscript at growth factor (VEGF) overexpression in soft tissue sarcomas and the role www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Clinical Sarcoma Research Springer Journals

Tumour angiogenesis in Epstein-Barr virus-associated post-transplant smooth muscle tumours

Loading next page...
 
/lp/springer-journals/tumour-angiogenesis-in-epstein-barr-virus-associated-post-transplant-CJ4SitVqLq

References (70)

Publisher
Springer Journals
Copyright
Copyright © 2014 by Jonigk et al.; licensee BioMed Central Ltd.
Subject
Biomedicine; Cancer Research; Oncology; Surgical Oncology
eISSN
2045-3329
DOI
10.1186/2045-3329-4-1
pmid
24398114
Publisher site
See Article on Publisher Site

Abstract

Epstein-Barr virus (EBV)-associated post-transplant smooth muscle tumours (PTSMT), are rare complications following organ/stem cell transplantation. Despite the mainly benign behaviour of PTSMT, alternative therapies are needed for those patients with progressive tumours. In tumours not approachable by surgery or reduction of immunosuppression, the angiogenic microenvironment might be a potential target of therapy, an approach that is well utilised in other soft tissue neoplasms. In a previous study, we evaluated the expression of EBV-related genes and the microRNA profile in PTSMT, but so far the characteristics of angiogenesis in PTSMT are not known. Therefore, the aim of this study was to evaluate the expression pattern of angiogenesis-related genes in PTSMT, in order to identify potential target molecules for anti-angiogenic therapy. PTSMT (n = 5 tumours) were compared with uterine leiomyomas (n = 7). Analyses included real-time PCR of 45 angiogenesis-associated genes, immunohistochemistry (CD31, prostaglandin endoperoxide synthase 1/PTGS1) and assessment of tumour vascularisation by conventional histopathology. PTSMT showed similar or fewer vessels than leiomyomas. Of the genes under investigation, 23 were down-deregulated (pro-angiogenic and some anti-angiogenic factors) and five were up-regulated (e.g. PTGS1 which is expressed at very low levels in leiomyomas but moderately higher levels in PTSMT). In summary, no particular target molecule could be identified, because tumour angiogenesis in PTSMT is characterised by low levels of major pro-angiogenic factors and there is no prominent increase in tumour vascularisation. EBV can induce angiogenesis via its viral late membrane protein 1 (LMP1) but PTSMT frequently do not express LMP1, which could be an explanation why, despite EBV infection, PTSMT show no exaggerated tumour angiogenesis. Keywords: PTSMT, Post-transplant smooth muscle tumours, EBV, Angiogenesis, Tumour Introduction previously analysed cell cycle factors, cytokines and gene Epstein-Barr virus (EBV)-associated post-transplant smooth promoter methylation in PTSMT and found an activated muscle tumours (PTSMT) are rare complications following phosphoinositide 3-kinase (PI3K)/mammalian target of solid graft and stem cell transplantation [1]. The molecular rapamycin (mTOR) cell cycle pathway as well as ex- pathobiology of this rare neoplastic entity is not fully under- pression of vascular endothelial growth factor (VEGF) stood and only few experimental analyses have addressed and Fms-related tyrosine kinase 1 (FLT1/VEGFR1) [1]. this issue [1,2]. Tumour cells are thought to be derived from In general, in addition to endogenous molecular de- aberrant myogenous venous/perivascular wall cells [3]. They fectswhich affect mitosisand apoptosisofthe tumour express smooth muscle proteins (actin and desmin), but not cells, angiogenesis is a major mechanism which con- CD117, CD34 or other endothelial marker proteins. Histo- tributes to tumour cell survival by supplying the me- morphology is characterised by mild atypia, low mitotic rate tabolism of aberrant cell proliferation. Currently, for and absence of prominent tumour necrosis. All in all, PTSMT, surgery and reduced immunosuppression are PTSMT show more histological features of benign leiomyo- the therapy of choice [1]. At this point, there is no mas rather than leiomyosarcomas [1,2] and our group has proof that patients benefit from conventional chemo- therapy or radiation alone [1]. In other soft tissue neo- plasms, numerous studies have addressed the angiogenic * Correspondence: Hussein.Kais@MH-Hannover.de Institute of Pathology, Hannover Medical School (MHH), Carl-Neuberg-Str. 1, microenvironment as a potential target of therapy. In D-30625 Hanover, Germany © 2014 Jonigk et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 2 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 PTSMT, angiogenesis might be of special importance, as values (normalised to a mean of endogenous control the original/progenitor tumour cell in these neoplasms is genes). Statistical analysis was performed with Prism 5.0 generally thought to be derived from an aberrant perivas- (GraphPad Software, San Diego, CA, USA) by applying cular/venous wall cell. This topic is also important in the non-parametric Kruskal-Wallis test followed by the PTSMT, as these can manifest in any anatomical localisa- Mann-Whitney test for two-group comparison. P values < tion and cerebral tumours are in particular associated with 0.05 were considered as statistically significant. a poor prognosis [1]. From other tumours, in particular renal cell cancer, we Immunohistochemistry for evaluation of selected genes know that hypoxia-inducible factor 1, alpha subunit (HIF1A) Deparaffinised and rehydrated FFPE tissue sections signalling mediates expression of VEGF, platelet-derived (1-2 μm) were stained after autoclave pre-treatment. For growth factor (PDGF) and angiopoietin via the PI3K/mTOR staining of platelet/endothelial cell adhesion molecule 1 pathway [4,5]. These cytokines activate pro-angiogenic re- (VCAM1/CD31), sections were processed in an auto- ceptors such as VEGFR and PDGF receptors (PDGFR). For mated staining system (Benchmark ULTRA, Ventana a variety of neoplasms, e.g. soft tissue sarcomas such Medical Systems, Inc., Tucson, AZ, USA). Prostaglandin as leiomyosarcomas, it has been shown that a VEGFR/ endoperoxide synthase 1 (prostaglandin G/H synthase PDGFR-mediated increase of angiogenesis can be inhib- and cyclooxygenase) (PTGS1) was stained manually ited by anti-angiogenic agents [6-9]. The aim of this (positive control: FFPE prostate cancer tissue). Mouse analysis was to evaluate the expression pattern of monoclonal antibodies were used. Vascularisation was angiogenesis-related genes in PTSMT, in order to quantified by counting CD31 vessels per 10 high identify potential target molecules for anti-angiogenic power fields (HPF) and then correlating them in seri- therapy, in particular for those patients who suffer ally cut haematoxylin-eosin-stained sections. Statistical from irresectable or progressive tumours. analysis was performed with Prism 5.0 as described above. Material and methods Tissue specimens Five EBV PTSMT samples from four patients, including Results two tumours from one patient (#4), and seven EBV be- Vascularisation of PTSMT nign uterine leiomyomas from solid graft recipients were As previously described, PTSMT tumour cells them- analysed. These cases had been characterised earlier selves were negative for CD31. In the cerebral PTSMT (Additional file 1: Table S1) [1]. Formalin-fixed and we could previously demonstrate aneuploidy of the MYC paraffin-embedded (FFPE) samples were retrieved from locus 8q24 by fluorescence in situ hybridisation (FISH) [1]. the archives of the Institute of Pathology (Hannover In this case, endothelial cells showed a normal MYC con- Medical School/MHH, Hanover, Germany). The retro- figuration. Thus, a clonal relation between PTSMT and spective evaluation has been approved by the local eth- endothelial cells could not be proven (Additional file 1: ics committee (MHH). Figure S1). PTSMT showed similar or fewer vessels than leiomyo- Expression analysis of angiogenesis-associated factors mas (mean 301/range 201-518 versus mean 511/range Tissue from FFPE blocks with >90% tumour cells were 306-789 CD31 vessels/10 HPF, p = 0.0480; Figure 1). cut and processed for further PCR analysis. In blocks Corresponding to the low significance level, there was a with <90% aberrant neoplastic cells, the PTSMT compart- broad overlap in vessel density between these two leio- ments of the specimens were laser microdissected using a myomatous tumour entities. Furthermore, gene expres- SmartCutPlus-System (MMI, Glattbrugg, Switzerland), as sion analysis of CD31 did not correlate with vessel previously described [1,10]. Cells were digested in protein- density. Higher rather than lower expression levels of ase K and RNA was extracted with phenol/chloroform CD31 were detectable in PTSMT (mean 20.10/range [1,10]. Synthesis of cDNA from mRNA, subsequent pre- 5.26-30.48 in PTSMT versus mean 6.76/range 2.44-11.40 amplification of cDNA and real-time quantitative PCR of in leiomyomas; p = 0.0303). Sinusoids without smooth 45 angiogenesis-associated genes and three endogenous muscle cell wall appeared generally smaller in PTSMT controls with a 7900HT Fast Real-Time PCR system were and more hyalinised but, in comparison to leiomyomas performed according to the manufacturers’ instructions the quantitative difference was not significant. PTSMT (Applied Biosystems, Carlsbad, CA, USA). Endogenous had significantly fewer arterioles, as defined by vessels controls were polymerase (RNA) II (DNA-directed) with a smooth muscle wall (mean 1 versus 15 vessels/10 polypeptide A, 220 kDa (POLR2), glucuronidase beta HPF, p = 0.0058). In summary, there was no clear evi- (GUSB) and glyceraldehyde-3-phosphate dehydrogenase dence that PTSMT are generally more vascularised than -ΔCT (GAPDH). Delta C values were converted into 2 leiomyomas. T Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 3 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Figure 1 Similar vascularisation of PTSMT and leiomyomas. A) Histological counts of vessels. B) CD31+ vessels in PTSMT (original magnification x200). C) CD31+ vascularisation in leiomyomas (x200). D) Weak PTGS1 protein expression in PTSMT (x400). E) Very weak protein expression in a leiomyoma (x400). Reduced expression of angiogenesis-associated genes in Except for CD31, significant differences of other up- PTSMT regulated factors were due to very low expression in Among 45 angiogenesis-associated mediators under in- leiomyomas rather than strong expression in PTSMT. vestigation, 28 were significantly deregulated in PTSMT: These factors were angiopoietin 2 (ANGPT2), PDGFRA, 23 were down-deregulated and 5 (including CD31, see PTGS1 and thymidine phosphorylase (TYMP). Because above) were up-regulated (Figure 2, Table 1). PTGS1 can be inhibited by widely used non-steroidal Prominently down-regulated factors included e.g. pro- anti-inflammatory drugs, immunohistochemistry was angiogenic HIF1A, fibroblast growth factor receptor 1 performed for evaluation if the tumour cells showed a (FGFR1/FLT2), kinase insert domain receptor (VEGFR2/ corresponding protein expression. A weak expression KDR) and VEGFA as well as anti-angiogenic serpin peptid- of PTGS1 proteins in PTSMT and leiomyomatous ase inhibitor, clade E (nexin, plasminogen activator inhibi- smooth muscle spindle cells was detectable (Figure 1). tor type 1), member 1 (SERPINE1), thrombospondin 1 Weak protein expression corresponded with relatively (THBS1) and TIMP metallopeptidase inhibitor 2 (TIMP2). low transcript expression levels in both tumour types Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 4 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Figure 2 Representative summary of up- and down-regulated angiogenesis-associated factors in PTSMT. (mean 0.41/range 0.09-0.94 in PTSMT versus mean the expression profiles of angiogenesis-related factors in 0.03/range 0.00-0.08 in leiomyomas; p = 0.0061). PTSMT. However, in contrast to this assumption we found almost the opposite: PTSMT showed similar or Discussion even reduced vascularisation, when compared to sporadic Patients suffering from PTSMT benefit from surgical leiomyomas. Furthermore, we could show that this mor- tumour resection and/or reduction of immunosuppres- phological feature was based on a previously unknown sion [1]. However, surgical respectability depends on molecular characteristic of PTSMT, namely expression of tumour site and, of note, PTSMT can manifest at any lo- low levels of pro-angiogenic factors and high levels of calisation, including the transplanted organ, in particular anti-angiogenic genes. In particular major factors of liver grafts [1]. Furthermore, multiple PTSMT, e.g. in hypoxia-inducible angiogenesis such as HIF1A, VEGFA, the lung, are not suitable for a surgical approach [11]. VEGFC, VEGFR1/FLT1, VEGFR2/KDR and FGFR1/FLT2 Due to the rarity of this tumour entity, prospective eval- were expressed at low levels. In contrast to PTSMT, leio- uations of therapeutic strategies will not be applicable in myosarcomas show generally higher expression of VEGFA a considerable number of patients. However, additional than leiomyomas [15-17]. In leiomyosarcoma-derived cell therapy options are mandatory for those patients who lines it could be demonstrated that hepatocyte growth fac- cannot be operated and/or whose transplant organ does tor (HGF) induces a decrease in anti-angiogeneic THBS1 not tolerate reduction of immunosuppression. In indi- and an increase in VEGFA [18]. In PTSMT, HGF, THBS1 vidual patients, it has been shown that inhibition of and VEGFA are all expressed at low levels, indicating mTOR signal pathways by sirolimus might be of thera- that HGF signalling does not contribute significantly peutic benefit [12-14]. The rationale for administration to tumour angiogenesis. In PTSMT, low levels were of an mTOR signalling inhibitor was based on the find- also detectable for other pro-angiogenic genes which ing that PTSMT and HIV-associated SMT, which share are involved in differentiation and proliferation of endo- morphological similarities with PTSMT, express mTOR thelial cells, e.g. vascular development-related EPH recep- [2]. However, sirolimus cannot be administered to all tor B4 (EPHB4) and sphingosine-1-phosphate receptor 1 transplanted patients, e.g. after renal transplantation, (S1PR1), the endothelium-specific receptor tyrosine kinase because the drug is potentially nephrotoxic. Another TEK and the growth factor midkine (MDK). Immune response-associated caveolae are plasma membrane invag- class of drugs which is widely used for systemic ther- apy of soft tissue neoplasms/sarcomas are anti-angiogenic inations of 60-80 nm in diameter in endothelial cells, agents, e.g. leiomyosarcoma [8]. Basic analysis of tumour- smooth muscle cells and other cell types [19] and caveolae components CAV2 (p = 0.07) and PTRF (p < 0.01) were associated angiogenesis is important for assessing the vulnerability of a given tumour type to these drugs. both decreased in PTSMT. In addition to several blood Prominent proliferation of vessels, high expression levels vessel-associated factors, lymphatic vessel protein podo- planin was decreased in PTSMT. Again, in leiomyosarco- of pro-angiogenic and low levels of anti-angiogenic genes would make it likely that PTSMT patients could benefit mas, podoplanin-positive vessels are especially found in from anti-angiogenic drug therapy. Therefore, we evaluated tumours with lymph node metastases [20]. In our cohort, Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 5 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Table 1 Expression profile of angiogenesis-related genes Gene (abbreviation) Gene name PTSMT (mean) Leiomyomas (mean) Significance ANGPT1 Angiopoietin 1 0.03 0.51 n.s. ANGPT2 Angiopoietin 2 0.91 0.31 p = 0.0303 ANGPTL1 Angiopoietin-like 1 0.05 0.05 n.s. ANGPTL4 Angiopoietin-like 4 0.03 0.05 n.s. BAI1 Brain-specific angiogenesis inhibitor 1 0.06 0.18 n.s. CAV2 Caveolin 2 0.84 2.35 n.s. CDK1 Cyclin-dependent kinase 1 0.01 0.01 n.s. CXCL1 Chemokine (C-X-C motif) ligand 1 (melanoma 4.25 7.55 n.s. growth stimulating activity, alpha) EPHB4 EPH receptor B4 0.06 2.50 p = 0.0025 FGF2 Fibroblast growth factor 2 (basic) 0.01 0.16 p = 0.0121 FGFR1/FLT2 Fibroblast growth factor receptor 1 3.36 21.33 p = 0.0051 GREM1 Gremlin 1, DAN family BMP antagonist 0.14 0.20 n.s. HGF Hepatocyte growth factor 0.02 0.07 p = 0.0242 HIF1A Hypoxia inducible factor 1, alpha subunit 6.48 13.81 p = 0.0051 IL8 Interleukin 8 0.60 0.03 n.s. LECT1 Leukocyte cell derived chemotaxin 1 0.11 Not detectable – MDK Midkine (neurite growth-promoting factor 2) 0.14 1.24 p = 0.0061 MMP2 Matrix metallopeptidase 2 (gelatinase A, 72 kDa 1.19 34.07 p = 0.0025 gelatinase, 72 kDa type IV collagenase) NOS3 Nitric oxide synthase 3 (endothelial cell) 0.05 0.31 p = 0.0025 NOTCH4 Notch 4 (receptor) 0.28 2.68 p = 0.0025 NOX4 NADPH oxidase 4 0.05 1.32 p = 0.0061 NRP1 Neuropilin 1 0.17 0.82 n.s. OLR1 Oxidized low density lipoprotein (lectin-like) receptor 1 0.04 0.02 n.s. PDGFA Platelet-derived growth factor alpha polypeptide 0.36 0.12 p = 0.0480 PDGFRA Platelet-derived growth factor receptor, alpha 1.66 7.31 n.s. polypeptide PDPN Podoplanin 0.09 3.36 p = 0.0061 PECAM1/CD31 Platelet/endothelial cell adhesion molecule 20.10 6.76 p = 0.0303 PTGS1 Prostaglandin-endoperoxide synthase 1 (prostaglandin 0.41 0.03 p = 0.0061 G/H synthase and cyclooxygenase) PTRF Polymerase I and transcript release factor 7.68 56.80 p = 0.0025 S1PR1 Sphingosine-1-phosphate receptor 1 0.15 0.76 p = 0.0172 SERPINF1 Serpin peptidase inhibitor, clade F (alpha-2 antiplasmin, 0.19 6.45 p = 0.0061 pigment epithelium derived factor), member 1 STAB1 Stabilin 1 0.06 0.16 n.s. TEK TEK tyrosine kinase, endothelial 0.11 1.17 p = 0.0043 TGFBR1 Transforming growth factor, beta receptor 1 0.05 0.38 p = 0.0025 THBS1 Thrombospondin 1 13.56 134.17 p = 0.0025 TIMP2 TIMP metallopeptidase inhibitor 2 9.32 54.78 p = 0.0025 TNFAIP2 Tumor necrosis factor, alpha-induced protein 2 0.08 0.01 n.s. TYMP Thymidine phosphorylase 0.17 0.08 p = 0.0480 VCAM1 Vascular cell adhesion molecule 1 0.18 0.95 p = 0.0303 VEGFA Vascular endothelial growth factor A 3.38 36.05 p = 0.0025 VEGFC Vascular endothelial growth factor C 0.47 1.66 p = 0.0043 Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 6 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Table 1 Expression profile of angiogenesis-related genes (Continued) VEGFD/FIGF C-fos induced growth factor (vascular endothelial 0.01 0.23 n.s. growth factor D) VEGFR1/FLT1 Fms-related tyrosine kinase 1 (vascular endothelial 0.15 0.73 p = 0.0061 growth factor/vascular permeability factor receptor) VEGFR2/KDR Kinase insert domain receptor (a type III receptor 0.26 3.48 p = 0.0061 tyrosine kinase) VEGFR3/FLT4 Fms-related tyrosine kinase 4 0.08 0.10 n.s. none of the PTSMT manifested in lymph nodes and, in cells in vitro and in vivo [30], in PTSMT increased MYC general, involvement of lymph nodes is rare in this type of expression is associated with decreased THBS1 expression transplant-associated neoplasm [1]. MMP2, which de- but there is no indication for a specific microRNA regula- grades the collagen IV-rich basal membrane as a necessary tion. Furthermore, while in leiomyosarcomas low expres- requisite for metastasis [21], was reduced in PTSMT, sion of THBS1 and TIMP2 is accompanied by increased which indicates no major remodelling of extracellular expression of pro-angiogenic factors such as VEGFA, matrix during tumour cell and endothelial proliferation. PTSMT in general did not show such a global pro- Compared to leiomyomas, only a few pro-angiogenic angiogenic expression profile. factors such as TYMP, ANGPTL2 and PTGS1 were in- As reviewed by Paydas [31], in lymphomas and naso- creased in PTSMT. However, statistical significances were pharyngeal carcinomas, tumour cell infection with EBV the result of very low expression levels in leiomyomas ra- is related to increased angiogenesis, in particular because ther than a prominent up-regulation in PTSMT. The the viral late membrane protein 1 (LMP1) induces ex- mean relative expression levels of these three factors pression of VEGF and activation of PTGS2, interleukin 8 was <1, indicating no major role in mediating tumour (IL8), fibroblast growth factor 2 (FGF2) and other pro- angiogenesis. angiogenic factors. Although PTSMT are infected with In PTSMT, three important anti-angiogenetic factors EBV, these tumours do not usually express LMP1 pro- were decreased: TIMP2, SERPINF1 and THBS1. TIMP2 teins [1,2,32] and this could be an explanation why, and SERPINF1 are strong inhibitors of endothelial pro- despite viral infection, PTSMT show no exaggerated liferation [8,22] and THBS1 induces reduced migration tumour angiogenesis. It is not known how the EBV+ ability of endothelial cells [23]. Furthermore, THBS1 can tumour cells suppress the expression of LMP1 while inhibit the binding of activating cytokines at receptors of expressing other viral proteins and the tumour bio- endothelial cells and can also bind to the thrombospon- logical benefit of this selective lack of LMP1 for the din receptor CD36 which induces endothelial apoptosis PTSMT proliferation is also not known. [23]. Other groups found that leiomyomas express Although clinical testing has not yet been performed, THBS1 more frequently than leiomyosarcomas [24]. In on the one hand it is questionable whether patients who addition, TIMP2 is also expressed at relatively low levels suffer from this type of soft tissue tumour might benefit in leiomyosarcomas [22]. from systematic anti-angiogenic drug therapy. On the It has been shown that the transcription factor MYC other hand, it could be assumed that PTSMT found their leads to expression of the chromosome segment own equilibrium of tumour vascularisation that allows 13q31.3-encoded microRNA 17~92 cluster which in- survival and growth without increasing the expression of cludes the two paralogues miR-19a and miR-19b-1 pro-angiogenic factors (e.g. due to aberrant signalling [25-27]. MicroRNA are non-coding molecules of 20- downstream of pro-angiogenic receptors). This might principally indicate a limited ability to circumvent therapy 25 nucleotides which bind to mRNA and negatively regulate protein translation [28]. THBS1-mRNA has and therefore anti-angiogenic drugs might not necessarily a miR-19 binding site and therefore MYC-related be ineffective since this would disrupt the equilibrium of PTSMT vascularisation. Anti-angiogenic drugs could still miR-19 expression down-regulates THBS1 [25-27]. PTSMT have an increased MYC expression [1] and be administered to PTSMT patients with no other treat- low levels of THBS1 but no up-regulation of the miR ment options available but, in these present analyses, we could not identify a specific target molecule. 17~92 cluster, including miR-19a (in PTSMT mean relative expression level 0.02 versus 0.03 in leiomyo- In summary, our analyses of the tumour angiogenesis mas) and miR-19b (mean 1.63 in PTSMT versus 2.23 in PTSMT revealed no particular target molecule, be- cause PTSMT are characterised by low levels of major in leiomyomas) [29]. The microRNA profile in PTSMT is overall associated with leiomyomatous differentiation of pro-angiogenic factors and there is no prominent in- the tumour cells [29]. Therefore, similar to mesenchymal crease in tumour vascularisation. Jonigk et al. Clinical Sarcoma Research 2014, 4:1 Page 7 of 7 http://www.clinicalsarcomaresearch.com/content/4/1/1 Additional file of overexpression in leiomyosarcoma. J Cancer Res Clin Oncol 2004, 130:52–56. 17. Sanci M, Dikis C, Inan S, Turkoz E, Dicle N, Ispahi C: Immunolocalization of Additional file 1: Table S1. Patient cohort. Figure S1. Fluorescence in VEGF, VEGF receptors, EGF-R and Ki-67 in leiomyoma, cellular leiomyoma situ hybridisation of the MYC gene shows no aneuploidy in endothelial and leiomyosarcoma. Acta Histochem 2011, 113:317–325. cells (white arrows) while a subfraction of PTSMT cells had an aneuploidy. 18. Zhang YW, Su Y, Volpert OV, Vande Woude GF: Hepatocyte growth factor/ scatter factor mediates angiogenesis through positive VEGF and negative thrombospondin 1 regulation. Proc Natl Acad Sci U S A 2003, Competing interests 100:12718–12723. The authors of this manuscript have no conflicts of interests to disclose. 19. Feng H, Guo W, Han J, Li XA: Role of caveolin-1 and caveolae signaling in endotoxemia and sepsis. Life Sci 2013. doi:10.1016/j.lfs.2013.05.016. Authors’ contributions 20. Lamyman MJ, Giele HP, Critchley P, Whitwell D, Gibbons M, Athanasou NA: Histomorphology (DJ, KH, HK), molecular analysis (LM, NI, ES), data collection, Local recurrence and assessment of sentinel lymph node biopsy in deep analysis of data and manuscript preparation (DJ, LM, NI, ES, HK, KH). All soft tissue leiomyosarcoma of the extremities. Clin Sarcoma Res 2011, 1:7. authors read and approved the final manuscript. 21. Bhuvarahamurthy V, Kristiansen GO, Johannsen M, et al: In situ gene expression and localization of metalloproteinases MMP1, MMP2, MMP3, Acknowledgements MMP9, and their inhibitors TIMP1 and TIMP2 in human renal cell The authors thank Regina Engelhardt for excellent technical support and carcinoma. Oncol Rep 2006, 15:1379–1384. Gillian Teicke for editing the text. 22. Chuang TD, Panda H, Luo X, Chegini N: miR-200c is aberrantly expressed Funding (DJ): “Deutsche Forschungsgemeinschaft” (DFG, GZ: 30 743/2-1). in leiomyomas in an ethnic-dependent manner and targets ZEBs, VEGFA, TIMP2, and FBLN5. Endocr Relat Cancer 2012, 19:541–556. Received: 29 August 2013 Accepted: 1 November 2013 23. Fernando NT, Koch M, Rothrock C, et al: Tumor escape from endogenous, Published: 7 January 2014 extracellular matrix-associated angiogenesis inhibitors by up-regulation of multiple proangiogenic factors. Clin Cancer Res 2008, 14:1529–39. References 24. Bodner-Adler B, Nather A, Bodner K, et al: Expression of thrombospondin 1 1. Jonigk D, Laenger F, Maegel L, et al: Molecular and clinicopathological (TSP 1) in patients with uterine smooth muscle tumors: an analysis of Epstein-Barr virus-associated posttransplant smooth muscle immunohistochemical study. Gynecol Oncol 2006, 103:186–189. tumors. Am J Transplant 2012, 12:1908–1917. 25. Italiano A, Thomas R, Breen M, et al: The miR-17-92 cluster and its target 2. Ong KW, Teo M, Lee V, et al: Expression of EBV latent antigens, THBS1 are differentially expressed in angiosarcomas dependent on MYC mammalian target of rapamycin, and tumor suppression genes in amplification. Genes Chromosomes Cancer 2012, 51:569–578. EBV-positive smooth muscle tumors: clinical and therapeutic implications. 26. Dews M, Homayouni A, Yu D, et al: Augmentation of tumor angiogenesis Clin Cancer Res 2009, 15:5350–5358. by a myc-activated microRNA cluster. Nat Genet 2006, 38:1060–1065. 3. Chaves NJ, Kotsimbos TC, Warren MA, et al: Cranial leiomyosarcoma in an 27. Dews M, Fox JL, Hultine S, et al: The myc-miR-17 92 axis blunts TGF{beta} Epstein-Barr virus (EBV)-mismatched lung transplant recipient. J Heart signaling and production of multiple TGF{beta}-dependent antiangiogenic Lung Transplant 2007, 26:753–755. factors. Cancer Res 2010, 70:8233–8246. 4. Figlin RA, Kaufmann I, Brechbiel J: Targeting PI3K and mTORC2 in 28. Hussein K: Pathobiology of the microRNA system. Pathologe 2012, 33:70–78. metastatic renal cell carcinoma: new strategies for overcoming 29. Jonigk D, Izykowski N, Maegel L, et al: MicroRNA expression in Epstein-Barr resistance to VEGFR and mTORC1 inhibitors. Int J Cancer 2013. virus-associated post-transplant smooth muscle tumours is related to doi:10.1002/ijc.28023. leiomyomatous phenotype. Clin Sarcoma Res 2013, 3:9. 5. Iacovelli R, Palazzo A, Trenta P, et al: Management of metastatic renal cell 30. Ngo CV, Gee M, Akhtar N, et al: An in vivo function for the transforming carcinoma progressed after sunitinib or another antiangiogenic myc protein: elicitation of the angiogenic phenotype. Cell Growth Differ treatment. Am J Clin Oncol 2013. doi:10.1097/COC.0b013e31827de888. 2000, 11:201–210. 6. van der Graaf WT, Blay JY, Chawla SP, et al: Pazopanib for metastatic 31. Paydas S: Strong cross-talk between angiogenesis and EBV: do we need soft-tissue sarcoma (PALETTE): a randomised, double-blind, different treatment approaches in lymphoma cases with EBV and/or placebo-controlled phase 3 trial. Lancet 2012, 379:1879–1886. high angiogenic capacity. Med Oncol 2012, 29:2159–2165. 7. Hensley ML, Sill MW, Scribner DR Jr, et al: Sunitinib malate in the 32. Rougemont AL, Alfieri C, Fabre M, et al: Atypical Epstein-Barr virus (EBV) treatment of recurrent or persistent uterine leiomyosarcoma: a latent protein expression in EBV-associated smooth muscle tumours gynecologic oncology group phase II study. Gynecol Oncol 2009, 115:460–465. occurring in paediatric transplant recipients. Histopathology 2008, 8. Ren JG, Jie C, Talbot C: How PEDF prevents angiogenesis: a hypothesized 53:363–367. pathway. Med Hypotheses 2005, 64:74–78. 9. Ganjoo KN: New developments in targeted therapy for soft tissue doi:10.1186/2045-3329-4-1 sarcoma. Curr Oncol Rep 2010, 12:261–265. Cite this article as: Jonigk et al.: Tumour angiogenesis in Epstein-Barr virus- 10. Jonigk D, Merk M, Hussein K, et al: Obliterative airway remodeling: associated post-transplant smooth muscle tumours. Clinical Sarcoma Research molecular evidence for shared pathways in transplanted and native 2014 4:1. lungs. Am J Pathol 2011, 178:599–608. 11. Somers GR, Tesoriero AA, Hartland E, et al: Multiple leiomyosarcomas of both donor and recipient origin arising in a heart-lung transplant patient. Am J Surg Pathol 1998, 22:1423–1428. Submit your next manuscript to BioMed Central 12. Toh HC, Teo M, Ong KW, et al: Use of sirolimus for Epstein-Barr virus-positive smooth-muscle tumour. Lancet Oncol 2006, 7:955–957. and take full advantage of: 13. Belingheri M, Comoli P, Locatelli F, et al: Successful medical treatment of EBV smooth muscle tumor in a renal transplant recipient. Pediatr • Convenient online submission Transplant 2010, 14:E101–104. • Thorough peer review 14. TanCS, LohHL, FooMW, Choong LH,WongKS, KeeTY: Epstein-Barr virus-associated smooth muscle tumors after kidney transplantation: • No space constraints or color figure charges treatment and outcomes in a single center. Clin Transplant 2013. • Immediate publication on acceptance doi:10.1111/ctr.12139. • Inclusion in PubMed, CAS, Scopus and Google Scholar 15. Hong T, Shimada Y, Uchida S, et al: Expression of angiogenic factors and apoptotic factors in leiomyosarcoma and leiomyoma. Int J Mol Med 2001, • Research which is freely available for redistribution 8:141–148. 16. Potti A, Ganti AK, Tendulkar K, et al: Determination of vascular endothelial Submit your manuscript at growth factor (VEGF) overexpression in soft tissue sarcomas and the role www.biomedcentral.com/submit

Journal

Clinical Sarcoma ResearchSpringer Journals

Published: Jan 7, 2014

There are no references for this article.