Screening for adenoviruses in haematological neoplasia: High prevalence in mantle cell lymphoma

European Journal of Cancer (2013) xxx, xxx– xxx

A v a i l a b l e a t w w w . s c i e n c e d i r e c t . c o m

ScienceDirect

journa l homepag e : www.e j cancer . com

Screening for adenoviruses in haematological neoplasia:High prevalence in mantle cell lymphoma

0959-8049/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.ejca.2013.10.013

⇑ Corresponding author: Address: St. Anna Kinderkrebsforschung, Children’s Cancer Research Institute, Zimmermannplatz 10, A-1090Austria. Tel.: +43 0 1 40077 4890; fax: +43 0 1 40077 64890.

E-mail address: [email protected] (T. Lion).

Please cite this article in press as: Kosulin K. et al., Screening for adenoviruses in haematological neoplasia: High prevalence in mantle cephoma, Eur J Cancer (2013), http://dx.doi.org/10.1016/j.ejca.2013.10.013

Karin Kosulin a, Margit Rauch a, Peter F. Ambros a,f, Ulrike Potschger a,Andreas Chott b, Ulrich Jager c, Johannes Drach d, Alexander Nader e,Thomas Lion a,f,⇑

a Children’s Cancer Research Institute, Vienna, Austriab Institute of Pathology, Medical University of Vienna, Austriac Department of Hematology, Medical University Vienna, Vienna, Austriad Department of Oncology, Medical University Vienna, Vienna, Austriae Institute of Pathology and Microbiology, Hanusch-Krankenhaus, Vienna, Austriaf Department of Pediatrics, Medical University Vienna, Vienna, Austria

KEYWORDS

Human adenovirusOncogenic virusMantle cell lymphomaLeukaemiaLymphoma

Abstract Human adenoviruses possess oncogenic capacity which is well documented in mam-malian animal models, but their possible implication in human malignancy has remained enig-matic. Following primary infection, adenoviruses can persist in a latent state in lymphocyteswhere the virus is apparently able to evade immune surveillance. In the present study, we haveemployed a broad-spectrum adenovirus polymerase chain reaction (PCR) assay to systemat-ically screen more than 200 diagnostic specimens of different lymphoid malignancies includingacute lymphocytic leukaemia (n = 50), chronic lymphocytic leukaemia (n = 50), various typesof malignant lymphoma (n = 100) and multiple myeloma (n = 11) for the presence of adeno-viral sequences. While most entities analysed revealed negative findings in virtually all speci-mens tested, adenoviral DNA was detected in 15/36 (42%) mantle cell lymphomasinvestigated. The most prevalent adenoviral species detected was C, and less commonly B.Adenovirus-positive findings in patients with mantle cell lymphoma were made at differentsites including bone marrow (n = 7), intestine (n = 5), lymph nodes (n = 2) and tonsillar tissue(n = 1). The presence of adenoviral sequences identified by PCR was confirmed in individualcells by fluorescence in-situ hybridisation (FISH). The frequent observation of adenoviruses inmantle cell lymphoma is intriguings, and raises questions about their possible involvement inthe pathogenesis of this lymphoid malignancy.� 2013 Elsevier Ltd. All rights reserved.

Vienna,

ll lym-


mailto:[email protected]




www.sciencedirect.com


2 K. Kosulin et al. / European Journal of Cancer xxx (2013) xxx–xxx

1. Introduction

Several decades ago, it was demonstrated that thehuman adenovirus (HAdV) type 12 can induce varioustumours in newborn hamsters, and similar observationswere later made in different experimental rodent modelswith other HAdV types belonging particularly to thespecies A, B and D [1–3]. The mechanism of malignanttransformation has been attributed to the early adenovi-ral proteins E1A and E1B which act as transcription fac-tors interfering with important tumour suppressorproteins such as Rb (retinoblastoma) or p53, as well asproteins of the DNA repair machinery and chromatinremodelling. Moreover, the adenoviral E3 and E4 pro-teins were shown to block the host immune responseand the onset of apoptosis [4–6]. These observationsindicate a rather complex pattern of interactionsbetween early adenoviral proteins and important cellu-lar control mechanisms which may contribute to malig-nant transformation.

Despite the well-established tumourigenic propertiesof different HAdVs in experimental mammalian animalmodels [1–3], no clear evidence for the association ofHAdV with any human malignancy has been providedto date. Few studies reported the presence of HAdV-DNA in neurogenic and small cell lung tumours [7,8],but other investigations revealed no evidence of HAdVin a variety of different solid human tumours [9,10].Two earlier studies focusing on haematological malig-nancies revealed the presence of HAdV in a small num-ber of paediatric and adult lymphoma patients, andabsence of the virus in various types of leukaemia[11,12].

Most of the studies in malignant neoplasia performedto date have focused on the screening for the most com-monly occurring types of HAdV species C, and specifi-cally for type 5 [8,11]. It is important to consider,however, that human adenoviruses constitute a largefamily which includes seven species, termed A–G, com-prising a steadily increasing number of currently morethan 60 different types 13–15. Primary infections withadenoviruses, which are clinically often inapparent, usu-ally occur in early childhood, and can lead to viral per-sistence in susceptible cells. A latent form of adenovirusinfection was shown to persist in tonsillar lymphocytesin nearly 80% of children investigated, and seems tooccur also in intestinal T-lymphocytes [16,17]. Ourrecent study revealed the presence of HAdV in T-lym-phocytes infiltrating liposarcoma [18]. These observa-tions indicate a particular prevalence of persistingHAdV in lymphocytes both within non-malignant andmalignant tissues, and it is an intriguing questionwhether intracellular persistence of viruses displayinghigh oncogenic potential might contribute to malignanttransformation of the lymphoid host cells. In the presentstudy, we have therefore screened over 200 diagnostic

Please cite this article in press as: Kosulin K. et al., Screening for adenovirphoma, Eur J Cancer (2013), http://dx.doi.org/10.1016/j.ejca.2013.10.013

specimens from adult patients with acute lymphocyticleukaemia (ALL), chronic lymphocytic leukaemia(CLL), multiple myeloma (MM) and different types ofmalignant lymphoma for the presence of HAdV-sequences by a highly sensitive broad-spectrum poly-merase chain reaction (PCR) assay established at ourcentre [19].

2. Materials and methods

2.1. Patients and tumour specimens

All patient samples studied were derived fromarchived materials collected at the time of diagnosis withinformed consent. Lymphoid leukaemia samples werecollected from 50 acute lymphocytic leukaemia (ALL)patients [peripheral blood (PB): n = 41; bone marrow(BM): n = 9] and 50 chronic lymphocytic leukaemia(CLL) patients (PB: n = 46; BM: n = 4). Furthermore,100 specimens from 95 patients with different types oflymphoma were investigated including follicular lym-phoma (FL; n = 12), diffuse large B-cell lymphoma(DLBL; n = 10), mucosa-associated lymphoid tissuelymphoma (MALT; n = 12), Hodgkin’s disease (HD;n = 14), nodular lymphocyte predominance Hodgkin’sdisease (NLPHD; n = 5), angioimmunoblastic T-celllymphoma (AILT; n = 7), anaplastic large cell lym-phoma (ALCL; n = 4) and mantle cell lymphoma spec-imens (MCL; n = 36). Thirty six MCL specimens from31 patients were derived from different sites includingthe bone marrow (n = 17), lymph nodes (n = 12), intes-tine (n = 6) and tonsil (n = 1).

All lymphoma specimens analysed were derived fromparaffin-embedded tumour tissue. Moreover, enriched(CD-138 positive) plasma cell specimens were obtainedfrom patients with multiple myeloma (MM; n = 11).The samples from leukaemia and myeloma patients wereobtained from the Division of Haematology, MedicalUniversity of Vienna, Austria. Most lymphoma speci-mens were obtained from the Institute of Pathology,Medical University of Vienna (n = 73), and additionalMCL samples were obtained from the Institute ofPathology and Microbiology, Hanusch-Krankenhaus,Vienna, Austria (n = 27).

2.2. DNA extraction

DNA extraction from patient samples was carriedout by the QIAmp Tissue DNA Mini Kit (QIAGEN,Germany) according to the manufacturer’s recommen-dations with proteinase K lysis overnight at 56�C.

2.3. Real-time PCR analysis

A real time quantitative polymerase chain reaction(RQ-PCR) assay detecting a broad spectrum of human

uses in haematological neoplasia: High prevalence in mantle cell lym-


K. Kosulin et al. / European Journal of Cancer xxx (2013) xxx–xxx 3

HAdV-types was used for screening of the tumour spec-imens [19]. The PCR tests were carried out using the7700 ABI Genetic Analyzer (Applied Biosystems, CA,United States of America (USA)) and 50–100 ng DNAwere used as template. To control the DNA templatequality, a single-copy human house-keeping gene(beta-2-microglobulin) was tested in parallel by RQ-PCR, and the presence of 5 � 103 copies of the controlgene was the minimum requirement for adequate samplequality. Samples that tested HAdV-positive by thescreening assay were subjected to species identificationby standardised RQ-PCR tests described previously[20,21] along with several negative controls. All PCRanalyses were carried out in a referral laboratory forvirus analysis under safety conditions facilitating pre-vention of contamination [22].

2.4. Fluorescence in situ hybridisation (FISH)

To control the RQ-PCR findings by an independentmethod, three paraffin-embedded tissue sections fromfive randomly selected MCL patients who testedHAdV-positive by PCR were subjected to FISH analy-sis, and images obtained from two patients displayingbright intracellular HAdV hybridisation signals areshown (Fig. 1B and C). The adenovirus probe forhybridisation was generated from a bacmid containingthe entire genome of the adenovirus type 5, kindlyprovided by T. Dobner, Heinrich-Pette-Institute,Leibniz Institute for experimental Virology, Hamburg,Germany. The adenoviral DNA was labelled by Nick-translation with bio-dUTP. Paraffin-embedded tumourtissue sections from MCL patients displaying a thicknessof 5 lm and HEK293 cells were hybridised with thelabelled human adenovirus DNA probe as describedpreviously [23,24]. FISH-analyses including pre-treat-ment, proteolytic treatment, hybridisation and detectionwere performed as described earlier [25]. The hybridisedbiotin-labelled probes were visualised by two antibodydetection steps using mouse anti-biotin and rabbit

Fig. 1. Fluorescence in-situ hybridisation (FISH) analysis of human adenlymphoma (MCL). FISH with a specific HAdV probe (red signals) on interpparaffin-embedded MCL specimens derived from bone marrow (panels B, Ccells. Panel D shows a lymph node section of another patient with MCL (Nnumber.


anti-mouse tetramethylrhodamine B isothiocyanate-conjugated secondary antibodies (DAKO, Denmark).

2.5. Statistics

The proportion of negative samples for leukaemiaand MCL samples is given together with 95% confidenceintervals (CI). Fisher’s exact test was used for statisticalcomparison. In myeloma and lymphoma entities otherthan MCL, the number of specimens was too small topermit statistical analysis.

3. Results and discussion

In the present study, over 200 diagnostic specimens ofpatients with various haematological neoplasias includ-ing different types of lymphoma, acute and chronic lym-phocytic leukaemia and multiple myeloma wereinvestigated for the presence of adenoviral sequencesusing a broad-range quantitative PCR assay [19]. Theyield and quality of DNA extracted from individualspecimens was adequate for analysis in all instances, asdocumented by parallel PCR-amplification of a controlhouse-keeping gene.

The lymphoid leukaemia specimens, including 50ALL and 50 CLL samples, tested negative for HAdVDNA, with the exception of a single ALL case. Hence,the calculated confidence interval for the lack of occur-rence of HAdV in ALL was 95% CI: 0.1–11%, and inCLL 95% CI: 0–7%. All eleven MM samples analysedalso revealed negative test results, but 18/100 specimensderived from patients with various types of lymphomatested positive for HAdV DNA (Table 1). Among theeight different types of lymphoma investigated, HAdV-positive specimens were observed in three entities includ-ing MCL, MALT and AILT (Table 1). In the latter twolymphoma types, positive specimens were observed in1/7 and 2/12 patients, respectively. In MCL, however,initial PCR screening revealed HAdV-positive resultsin five of the first ten patient samples investigated, thusprompting us to test a larger number of MCL

ovirus (HAdV) DNA in HEK293 cells and patients with mantle cellhase nuclei of HEK293 cells (panel A), and two independent sections of; patient No. 7-Table 2) with 15–20 HAdV signals in the virus-positive

o. 11-Table 2) displaying a rare event of a cell with very high virus copy



Table 1Lymphoid tumour specimens analysed by human adenovirus (HAdV)-specific real time quantitative polymerase chain reaction (RQ-PCR)assay.

Tumour entity No. of investigatedsamples

No. of HAdV-positivesamples

LymphomaFL 12 –DLBL 10 –MCL 36 15MALT 12 2ALCL 4 –AILT 7 1HD 14 –NLPHD 5 –

Lymphocytic leukaemiaALL 50 1CLL 50 –

Multiple myeloma(MM)

11 –

Total 211 19

FL, follicular lymphoma; DLBL, diffuse large B cell lymphoma; MCL,mantle cell lymphoma; MALT, mucosa-associated lymphoid tissuelymphoma; ALCL, anaplastic large cell lymphoma; AILT, angi-oimmunoblastic T-cell lymphoma; HD, Hodgkin’s disease; NLPHD,nodular lymphocyte predominance Hodgkin’s disease; ALL, acutelymphoid lymphoma; CLL, chronic lymphoid lymphoma.

Table 2Human adenovirus (HAdV)-positive lymphoma specimens: HAdVspecies and localisation.

Lymphoma (Patient No.) HAdV species Localisation

MCL (15/36)1 B Lymph node2 C Colon3 C Colon4 C Colon5 C Colon6 B + C Colon6 B + C Bone marrow7 C Tonsil7 C Bone marrow8 C Bone marrow9 C Bone marrow

10 C Bone marrow11 C Lymph node12 C Bone marrow13 C Bone marrow

MALT-L. (2/12)1 C Stomach2 C Orbita

AILT (1/7)1 B Lymph node

MCL, mantle cell lymphoma; MALT, mucosa-associated lymphoidtissue lymphomal; AILT, angioimmunoblastic T-cell lymphoma.


specimens. The extended analysis in MCL unraveled thepresence of HAdV DNA in 15/36 (42%) of the samplesanalysed (95% CI: 26–59%) and, in comparison withother testable entities, the HAdV positivity in MCLwas highly significant (p < 0.001). The number of HAdVparticles in MCL specimens estimated by RQ-PCR wasin a range between 40 and 800 virus copies per 103 cells.In the MCL patients investigated, HAdV-positive spec-imens were observed in 83% of tumour specimens local-ised in the intestine, and in 41% of samples derived fromthe bone marrow. Adenovirus typing by species-specificPCR [20] revealed the presence of HAdV C in mostinstances, but HAdV B was detectable in some lym-phoma specimens with or without concomitant presenceof HAdV C (Table 2). In the HAdV species identified,molecular subtyping down to the genotype level or per-formance of viral gene expression studies was not feasi-ble due to the low amount of adenovirus-positive cellswithin the tumour tissue. The lymphoma specimens dis-playing HAdV B revealed the highest virus copy num-bers among the samples tested with 300, 700 and 800copies per 103 cells in tumour sections derived fromthe intestine, lymph node and bone marrow, respec-tively, as calculated by RQ-PCR analysis. The MCLspecimens investigated were derived predominantlyfrom the time of primary diagnosis. It is not conceivabletherefore that the frequent observation of HAdV inMCL specimens could be related to impaired immuneresponse resulting from preceding therapeutic proce-dures such as autologous stem cell transplantation.


To address the possibility of false HAdV-positivePCR findings attributable particularly to contamina-tion, randomly selected HAdV-positive MCL specimenswere subjected to FISH-analysis with a virus-specificDNA probe. The HEK293 cell line, which displays inte-grated adenoviral sequences, was used as an externalcontrol for successful hybridisation. The cell line dis-played two or three distinct small-sized signals in eachcell (Fig. 1A). By contrast, tissue sections from MCLpatients showed signals reflecting the presence of HAdVonly in a small proportion of cells within the tumour,generally not exceeding 1 in 10.000. The loosely scat-tered HAdV-positive cells within individual tumour sec-tions mostly displayed 15–20 hybridisation signalsindicating the presence of multiple virus copies, withreproducible patterns in repeated experiments usingthree different sections of the tumours analysed(Fig. 1B and C). Only exceptionally, a higher densityof virus copies was found in individual cells (Fig. 1D).

The best-documented site of HAdV persistence afterprimary infection is in T-lymphocytes from adenoidsand tonsils [9,16,26,27]. In vitro studies showed, how-ever, that B-lymphocytes are also susceptible to adeno-viral infection, even if the main receptor mediatingcellular entry of the virus, CAR (Coxsackie-AdenovirusReceptor), is expressed at low levels [28]. In theseinstances, infection may occur by CAR-independentvirus uptake via integrins [29]. Adenoviruses have beendetected in T-lymphocytes infiltrating malignant tissue[18], and persistence of HAdV genomes in episomal



K. Kosulin et al. / European Journal of Cancer xxx (2013) xxx–xxx 5

form has been documented both in T- and B-lympho-cytic cell lines [30]. These observations and theoncogenic properties of different HAdV types in experi-mental animal models have fuelled the hypothesis thatpersistence of these viruses in lymphoid cells could con-tribute to malignant transformation. In the presentstudy, we have therefore investigated more than 200diagnostic specimens from patients with various lym-phoid malignancies, and observed absence or only spo-radic evidence of HAdV sequences in most of theneoplastic entities investigated. These observations arein line with earlier studies in leukaemia and lymphoma[11,31,32]. By contrast, we have demonstrated a highprevalence of HAdV in mantle cell lymphoma, whereviral DNA sequences were detected in 42% of thepatients.

Mantle cell lymphomas represent a rare type of non-Hodgkin’s lymphoma (NHL) with poor prognosis [33].They constitute a subtype of B-cell lymphoma display-ing CD5-positive antigen-naive pregerminal centre cellsin the mantle zone surrounding normal germinal centrefollicles. The major genetic deregulations in MCLinclude the translocation t(11;14)(q13;32) leading tooverexpression of cyclin D1. Moreover, attenuation ofthe DNA damage response by mutations in p53 andATM as well as alterations in PI3K and TGFbeta havebeen documented [34]. A possible involvement of HAdVin the pathogenesis of MCL could be related to theknown interaction of adenoviral oncogenes includingE1A, E1B 55 kDa, E4ORF3 and E4ORF6 with impor-tant regulatory elements of cell cycle control, DNArepair, apoptosis and transcription such as p53, Daxxand PML [35–37].

The viral species detected in MCL included particu-larly HAdV C, and less commonly B. The presence ofHAdV C in intestinal locations of MCL may be relatedto the notion that the gastrointestinal tract has been asuspected site of adenoviral persistence for many years[17,38]. Different studies have demonstrated a high prev-alence of HAdV C types in stool samples of immuno-suppressed patients [20,39,40] which is believed to arisefrom reactivation of the virus from persistent infectionsin the intestine [26].

Although false positivity resulting from contamina-tion is a general problem associated with PCR analysis,it can be virtually excluded in the study performed dueto the specific precautions and controls employed [22].Nevertheless, the PCR findings were verified by an inde-pendent method, fluorescence in-situ hybridisation(FISH), which is not prone to false-positive results dueto contamination. The relatively high number of viruscopies in very rare individual HAdV-positive cells ofMCL specimens observed by FISH-analysis (Fig. 1)could indicate the presence of replicating virus, but sim-ilar observations have also been made in latentlyinfected lymphocytes [26]. It is conceivable that MCL


cells provide an environment more conducive to theexpression of viral oncogenic proteins than other typesof B- or T-lymphocytes, and could therefore be moreprone to deregulation of host-cellular mechanisms lead-ing to a transformed phenotype. However, the putativerole of HAdV in the pathogenesis of this disease entityremains to be elucidated.

Conflict of interest statement

None declared.

Acknowledgements

This study was supported by the Jubilaumsfonds ofthe National Bank of Austria (Grant No. 13078).

References

[1] Graham FL, Rowe DT, McKinnon R, Bacchetti S, Ruben M,Branton PE. Transformation by human adenoviruses. J CellPhysiol Suppl 1984;3:151–63.

[2] Hohlweg U, Dorn A, Hosel M, Webb D, Buettner R, Doerfler W.Tumorigenesis by adenovirus type 12 in newborn Syrian ham-sters. Curr Top Microbiol Immunol 2004;273:215–44.

[3] Javier R, Raska Jr K, Macdonald GJ, Shenk T. Humanadenovirus type 9-induced rat mammary tumors. J Virol 1991;65:3192–202.

[4] Nevels M, Rubenwolf S, Spruss T, Wolf H, Dobner T. Theadenovirus E4orf6 protein can promote E1A/E1B-induced focusformation by interfering with p53 tumor suppressor function.Proc Natl Acad Sci U S A 1997;94:1206–11.

[5] Endter C, Dobner T. Cell transformation by human adenoviruses.Curr Top Microbiol Immunol 2004;273:163–214.

[6] Schreiner S, Burck C, Glass M, et al. Control of humanadenovirus type 5 gene expression by cellular Daxx/ATRXchromatin-associated complexes. Nucleic Acids Res 2013;41:3532–50.

[7] Ibelgaufts H, Jones KW, Maitland N, Shaw JF. Adenovirus-related RNA sequences in human neurogenic tumours. ActaNeuropathologica 1982;56:113–7.

[8] Kuwano K, Kawasaki M, Kunitake R, et al. Detection of groupC adenovirus DNA in small-cell lung cancer with the nestedpolymerase chain reaction. J Cancer Res Clin Oncol 1997;123:377–82.

[9] Green M, Wold WS, Mackey JK, Rigden P. Analysis of humantonsil and cancer DNAs and RNAs for DNA sequences of groupC (serotypes 1, 2, 5 and 6) human adenoviruses. Proc Natl AcadSci U S A 1979;76:6606–10.

[10] Mende Y, Schneider PM, Baldus SE, Doerfler W. PCR-screeningof human esophageal and bronchial cancers reveals absence ofadenoviral DNA sequences. Virus Res 2004;104:81–5.

[11] Fernandez-Soria V, Bornstein R, Forteza J, Parada C, Sanchez-Prieto R, Cajal S. Inconclusive presence of adenovirus sequencesin human leukemias and lymphomas. Oncol Rep 2002;9:897–902.

[12] Kosulin K, Haberler C, Hainfellner JA, Amann G, Lang S, LionT. Investigation of adenovirus occurrence in pediatric tumorentities. J Virol 2007;81:7629–35.

[13] Berk AJ. Adenoviridae: the viruses and their replication. In:Knipe DM, Howley PM, et al., editors. Fields virology, vol. 2, 5thed. Philadelphia: Lippincott Williams & Wilkins, 2007. p. 2355–94.

[14] Zhou X, Robinson C, Rajaiya J, et al. Analysis of humanadenovirus type 19 associated with epidemic keratoconjunctivitis


http://refhub.elsevier.com/S0959-8049(13)00951-9/h0005











































and its reclassification as adenovirus type 64. Invest OphthalmolVis Sci 2012;53:2804–11.

[15] Kaneko H, Aoki K, Ishida S, et al. Recombination analysis ofintermediate human adenovirus type 53 in Japan by completegenome sequence. J Gen Virol 2011;92:1251–9.

[16] Garnett CT, Erdman D, Xu W, Gooding LR. Prevalence andquantitation of species C adenovirus DNA in human mucosallymphocytes. J Virol 2002;76:10608–16.

[17] Roy S, Calcedo R, Medina-Jaszek A, Keough M, Peng H, WilsonJM. Adenoviruses in lymphocytes of the human gastro-intestinaltract. PloS One 2011;6:e24859.

[18] Kosulin K, Hoffmann F, Clauditz T, Wilczak W, Dobner T.Presence of adenovirus species C in infiltrating lymphocytes ofhuman sarcoma. PloS One 2013;8:e63646.

[19] Ebner K, Suda M, Watzinger F, Lion T. Molecular detection ofthe entire spectrum of human adenoviruses by a two-reaction RQ-PCR assay. J Clin Microbiol 2005;43:3049–53.

[20] Lion T, Baumgartinger R, Watzinger F, et al. Molecularmonitoring of adenovirus in peripheral blood after allogeneicbone marrow transplantation permits early diagnosis of dissem-inated disease. Blood 2003;102:1114–20.

[21] Ebner K, Rauch M, Preuner S, Lion T. Typing of humanadenoviruses in specimens from immunosuppressed patients byPCR-fragment length analysis and real-time quantitative PCR. JClin Microbiol 2006;44:2808–15.

[22] Watzinger F, Suda M, Preuner S, et al. Real-time quantitativePCR assays for detection and monitoring of pathogenic humanviruses in immunosuppressed pediatric patients. J Clin Microbiol2004;42:5189–98.

[23] Arbuckle JH, Medveczky MM, Luka J, et al. The latent humanherpesvirus-6A genome specifically integrates in telomeres ofhuman chromosomes in vivo and in vitro. Proc Natl Acad Sci U SA 2010;107:5563–8.

[24] Reisinger J, Rumpler S, Lion T, Ambros PF. Visualization ofepisomal and integrated Epstein-Barr virus DNA by fiberfluorescence in situ hybridization. Int J Cancer 2006;118:1603–8.

[25] Stock C, Ambros IM, Lion T, et al. Detection of numerical andstructural chromosome abnormalities in pediatric germ celltumors by means of interphase cytogenetics. Genes ChromosomesCancer 1994;11:40–50.

[26] Lion T, Kosulin K, Landlinger C, et al. Monitoring of adenovirusload in stool by real-time PCR permits early detection ofimpending invasive infection in patients after allogeneic stem celltransplantation. Leukemia 2010;24:706–14.


[27] Garnett C, Talekar G, Mahr J, et al. Latent species C adenovi-ruses in human tonsil tissues. J Virol 2009;83:2417–28.

[28] Segerman A, Lindman K, Mei YF, Allard A, Wadell G.Adenovirus types 11p and 35 attach to and infect primarylymphocytes and monocytes, but hexon expression in T-cellsrequires prior activation. Virology 2006;349:96–111.

[29] Hall K, Blair Zajdel ME, Blair GE. Unity and diversity in thehuman adenoviruses: exploiting alternative entry pathways forgene therapy. Biochem J 2010;431:321–36.

[30] Zhang Y, Huang W, Ornelles DA, Gooding LR. Modelingadenovirus latency in human lymphocyte cell lines. J Virol 2010;84(17):8799–810.

[31] Gustafsson B, Huang W, Bogdanovic G, et al. Adenovirus DNAis detected at increased frequency in Guthrie cards from childrenwho develop acute lymphoblastic leukaemia. Br J Cancer 2007;97:992–4.

[32] Vasconcelos GM, Kang M, Pombo-de-Oliveira MS, et al. Ade-novirus detection in Guthrie cards from paediatric leukaemiacases and controls. Br J Cancer 2008;99:1668–72.

[33] Witzig TE. Current treatment approaches for mantle-cell lym-phoma. J Clin Oncol 2005;23(26):6409–14.

[34] Parekh S, Weniger MA, Wiestner A. New molecular targets inmantle cell lymphoma. Semin Cancer Biol 2011;21:335–46.

[35] Schreiner S, Wimmer P, Groitl P, et al. Adenovirus type 5 earlyregion 1B 55K oncoprotein-dependent degradation of cellularfactor Daxx is required for efficient transformation of primaryrodent cells. J Virol 2011;85:8752–65.

[36] Hartl B, Zeller T, Blanchette P, Kremmer E, Dobner T.Adenovirus type 5 early region 1B 55-kDa oncoprotein canpromote cell transformation by a mechanism independent fromblocking p53-activated transcription. Oncogene 2008;27:3673–84.

[37] Wimmer P, Schreiner S, Everett RD, Sirma H, Groitl P, DobnerT. SUMO modification of E1B–55K oncoprotein regulatesisoform-specific binding to the tumour suppressor protein PML.Oncogene 2010;29:5511–22.

[38] Wold WSM. Adenoviruses. In: Knipe DM, Howley PM, et al.,editors. Fields virology, vol. 2, 5th ed. Philadelphia: LippincottWilliams & Wilkins, 2007. p. 2395–2436.

[39] Heim A, Ebnet C, Harste G, Pring-Akerblom P. Rapid andquantitative detection of human adenovirus DNA by real-timePCR. J Med Virol 2003;70:228–39.

[40] Kroes AC, de Klerk EP, Lankester AC, et al. Sequentialemergence of multiple adenovirus serotypes after pediatric stemcell transplantation. J Clin Virol 2007;38:341–7.
























































































Documents

Screening for adenoviruses in haematological neoplasia: High prevalence in mantle cell lymphoma