Surgical Oncology 1992 ; 1 : 79-86

INTRODUCTION

The utilization of relatively selective cell-mediatedimmunosuppressive agents has allowed improvedorgan transplant survival rates with fewer infectiouscomplications [1] . Immunosuppressive therapy isadministered to transplant recipients for the life oftheir graft to prevent or treat rejection [2-5] . Com-plications of long term immunosuppression includean increase in the incidence of lymphoproliferativedisease (LPD) and malignancies including B-celllymphoma (BCL) [6] .The incidence of LPD in patients receiving pre-

dnisone and cyclosporine immunosuppressivetherapy following solid organ transplantation hasbeen reported to range from 2% to 5% [7] . Malatacket al. estimated the incidence of LPD in paediatricpatients after liver transplantation to be 2 .8% per

*Correspondence : H . Kim Lyerly, MD, Box 3551, Duke Uni-versity Medical Center, Durham, North Carolina 27710, USA.

Effects of cyclosporine on human B-cell lymphomadevelopment in vivo

T. J . BOYLE, R . E. COLES, A. M. KIZILBASH AND H . K. LYERLY"The Departments of Surgery and Pathology', Duke University Medical Center, Durham, North Carolina 27770, USA

Cyclosporine (CsA) is a potent immunosuppressive agent primarily affecting T-lympho-cyte function . Patients receive CsA following organ transplantation to prevent rejection .These patients are at high risk for developing Epstein-Barr virus (EBV)-inducedlymphoproliferative disease (LPD) or B-cell lymphoma (BCL) . Severe Combined Immuno-deficient (SCID) mice reconstituted with human peripheral blood leukocytes (PBL)develop fatal B-cell lymphomas of human origin following latent or active infectionwith EBV. This model was utilized to determine the role of CsA in the development ofhuman BCL. SCID mice were reconstituted with PBL, latently or actively infected withEBV, and treated with CsA. Following active EBV infection, mice developed human BCLwith or without CsA treatment . In contrast, treatment with CsA prevented the develop-ment of BCL in mice latently infected with EBV . This suggests a T-cell interaction withlatently infected B-cells which is perturbed by CsA . Further understanding of this inter-action and the occurrence of human BCL may allow the development of strategies toprevent, detect, or treat malignancies associated with immunosuppression . SurgicalOncology 1992; 1 : 79-86-

Keywords: B-cell lymphoma, cyclosporine, Epstein-Barr virus ; SCID mouse,transplantation .

79

year [6] . The incidence of B-cell lymphomas in post-transplant patients is 28- to 49-fold above thatencountered in age matched controls [8] . Whereaslymphomas make up 3-4% of all cancers in thispopulation of patients, they account for 21% ofmalignancies if non-melanoma skin cancers and insitu carcinomas of the uterine cervix are excluded[2-5] .

Epstein-Barr virus (EBV) is implicated in theaetiology of B-cell lymphoma development inimmunosuppressed patients . The EBV genome hasbeen identified in lymphomas occurring in thesepatients [9], and both enhanced salivary excretion[10] and increased numbers of latently infectedB-lymphocytes [11] have been noted in transplantrecipients. EBV is also associated with malignanciessuch as Burkitt's lymphoma and nasopharyngealcarcinoma [12, 131 . Patients with these diseasesoften have high levels of antibodies to EBV antigens[14, 15] and tumours contain EBV genomes andexpress certain EBV gene products [16, 17] . B-

80

T. J Boy/e et al .

lymphocytes infected with EBV in vitro becomeimmortalized B-lymphoblastoid cell lines (BLCL) andprovide a model system of EBV latency [18] .Homozygous C.B-17 scid/scid mice with severe

combined immune deficiency (SCID) have no T- orB-lymphocyte function [19] . Functioning humanhaematolymphoid cells can be engrafted into thesemice, however lethal lymphoproliferative disease,characterized by EBV-positive human B-cell lympho-mas (SCID/BCL) ensues if the peripheral bloodleukocytes (PBL) used for reconstitution are derivedfrom an EBV seropositive donor [20, 21] . SCID/BCLare aggressive and found in the abdomen, liver,thymus, and spleen . SCID/BCL also develop in miceengrafted with PBL from an EBV seronegative donorand infected with exogenous EBV [22] .Although B-cells infected with EBV in vitro

become immortalized B-lymphoblastoid cell lines[18], spontaneous outgrowth of BLCL is inhibited byanti-EBV cytotoxic T lymphocytes (CTL) [23] . CsAcan be utilized in vitro to inhibit anti-EBV CTL activityand thus allow more efficient transformation ofBLCL [24] . CsA blocks T-cell activation by inhibitionof Interleukin 2 (IL-2) production and IL-2 receptorexpression by these cells [25] . CsA and other potentT-cell-specific immunosuppressive agents have beenimplicated in the pathogenesis of LPD and lympho-mas developing in transplant recipients. It is pro-posed that their immunosuppressive effects in vivopermit EBV-induced polyclonal B-cell proliferationand the development of frankly malignant B-celllymphoma [6] . The SCID mouse model was utilizedto examine the effects of cyclosporine on humanB-cell lymphoma development in vivo.

MATERIALS AND METHODS

SCID mice

SCID mice were obtained from a hysterectomy-rederived central inbred colony of defined flora-gnotobiotic stock maintained in the Medical CenterVivarium at Duke University. Following transfer to abiosafety level 3 isolation facility, mice were main-tained in filter-capped Micro-Isolator cages (labproducts inc, Maywood NJ) which were autoclavedprior to use. Cages were housed within a HEPAfiltered Blickman isolator system (Blickman, FairlawnNJ). Mice were fed sterilized rodent chow, andprophylaxis against Pneumocystis carinii was pro-

vided with Septra oral suspension (Burroughs Well-come, Durham NC) ad libidum in sterile acidifiedwater at a concentration of 0 .89 mg ml - 'sulfamethoxazole and 0.18 mg ml - ' trimethoprim .Prior to all manipulation Micro-Isolator cages weretransferred to a laminar flow Biogard hood (TheBaker Company, inc., Stanford ME). Animals werehandled aseptically by investigators wearing sterilegloves, masks, and gowns . Blood samples (0.2 ml)were obtained by retro-orbital sinus bleed prior toinduction of all mice into experiments, and serumwas screened for murine Immunoglobulin G (IgG) byenzyme-linked immunosorbant assay (ELISA) usinggoat anti-mouse IgG (Organon Technika, DurhamNC) . Following quantification of total murine IgGlevels a concentration of 0.01 mg ml - ' was taken asa cut-off value, and mice with detectable levelsgreater than this, indicative of leakiness of the SCIDdefect, were excluded from study. Animals wereinspected daily and monitored carefully twiceweekly for signs of illness including weight loss,ruffled fur, rapid respiratory rate, inactivity, andpalpable abdominal masses . Animals were sacrifiedwhen abdominal tumours were evident by palpa-tion, or after a maximum of 26 weeks followingtransfer of donor cells. Sacrifice was by cardiacpuncture, to obtain terminal blood samples, follow-ing induction of halothane anaesthesia (Fluothane ® ,Ayerst laboratories Inc, New York, NY) . A fullautopsy was performed on all mice .

Human PBL engraftment

Peripheral blood leukocytes (PBL) were obtainedfrom adult donors following informed consent .Whole blood was obtained by venipuncture intoheparinized (preservative-free) syringes and sero-logically tested for EBV (ELISA : Du Pont, WilmingtonDE) and HIV-1 (ELISA : Du Pont, Wilmington DE) . PBLwere separated from whole blood using Lympho-cyte Separation Medium (LSM ® : Organon Technika,Durham NC) . Following immunotyping by fluores-cence-activated cell sorting (FACS) analysis, 50 x 10 6PBL in Dulbecco's phosphate buffered saline (D-PBS; GIBCO BRL, Grand Island NY) were transferredby intraperitoneal (IP) injection into SCID mice offrom five to eight weeks of age (hu-PBL-SCID) . Asingle EBV-seropositive or EBV-seronegative donorwas used for all reconstitutions .

Following PBL innoculation, blood samples wereobtained biweekly for measurement of human IgG

to confirm immune reconstitution . Human IgG inserum was quantified by ELISA using goat anti-human IgG (Organon Technika, Durham NC) .

The presence of human lymphocytes in the peri-pheral circulation, peritoneal washings, spleen, liver,

and lymph nodes of SCID mice engrafted withhuman cells was assessed selectively by FACSanalysis at ten weeks post-engraftment . Aliquots of50-70 pl of whole blood, obtained by cardiac punc-ture, were lysed with a commercial lysing reagent(Counter, Hialeah FL) to remove red cells . Peritonealwashings were obtained by the installation and sub-sequent removal after 5 min of 5 ml of D-PBS fromthe peritoneal cavity. Leukocytes in lavage speci-mens were analysed in 2X101 cell aliquots. Solidtissue (i .e., spleen, liver, kidney, and lymph nodes)was disaggregated mechanically in a Stomacher(Tekman Co., Cincinnati OH) . Cell suspensions werethen filtered through a 50 mm nylon mesh toremove aggregates and large debris . Cell aliquotsof 2 x 105 were analysed .

Direct immunofluorescent staining of all cellpopulations was performed using monoclonalmurine anti-human CD3, CD4, CD8, CD19, and CD21(Becton Dickinson, San Jose CA) . Cells were washedwith cold D-PBS containing 0 .5% BSA and 0 .1%NaN3 and incubated with labelled antibody for 30min at 4°C. Excess antibody was removed by wash-ing and cells were analysed using an EPICS Profile IIflow cytometer (Counter, Hialeah FL) at 488-nm forper cent positivity on a log fluorescent scale . Back-ground control cells were incubated with directlylabelled murine IgG .

EBV infection

Mice were latently infected with EBV by engraftmentwith PBL from EBV-seropositive donors . Mice wereactively infected with EBV on day 1 following PBLengraftment by i .p . innoculation of culture super-natants from B95-8 cells containing 10 6 immortaliz-ing units of EBV [26] .

Autopsy and evaluation for SCID/BCL

At the time of sacrifice, all animals were handled asdescribed above. Following cardiac puncture andperitoneal lavage, all mice underwent autopsy whichincluded inspection of the thoracic and abdominalcavities . All gross lesions suggestive of SCID/BCL

CsA effects on human lymphoma development in vivo

81

were harvested, sectioned into 0 .5 x 0 .5 x 0.3 cm3blocks and either snap-frozen in liquid nitrogen or

fixed in 10% neutral buffered formalin . In addition,

representative tissue samples from the spleen, liver,lungs, heart and lymph nodes were harvested andfixed in 10% neutral buffered formalin for histologicexamination . Tissues were dehydrated andembedded in paraffin, and 5 micron sections werestained with hematoxylin and eosin for morphologicevaluation. _PD lesions were studied to determine ifthey were monomorphic (malignant lymphomas) orpolymorphic (infectious mononucleosis-like), usingthe NIH Working Formulation classification [27] .

Cyclosporine

CsA (Sandimmune : Sandoz Pharmaceuticals Cor-poration, East Hanover NJ), diluted in 0.9% normalsaline, was administered by i .p . injection at an initialdose of 10 mg kg ' day - ' for 42 days. In subsequentexperiments CsA was administered at a dose of 10,25, or 50 mg kg - ' day - ' for 42 days . Mice were com-menced on CsA 24 h after latent or active infectionwith EBV, and dose was adjusted twice weeklyaccording to body weight. Control animals wereinjected with an equivalent volume of 0.9% normalsaline alone .

RESULTS

Engraftment

All animals studied were successfully engrafted withhuman immune components, as evidenced bydetection of human IgG in interim or terminal serumsamples, or by detection of human lymphocytes fol-lowing sacrifice, or both . No mice had detectablemurine IgG .

Development of SCID/BCL

Engraftment of SCID mice with PBL from EBV-seropositive donors led to the development of SCID/BCL in 70% of animals at 11 to 14 weeks followingreconstitution, as shown in Table 1 . Infection of micewith exogenous EBV following engraftment with PBLfrom either EBV-seropositive or seronegative donorsled to the development of SCID/BCL in 100% ofanimals at 5 to 10 weeks following reconstitutionand infection .

82 T J. Boyle et al .

Hu-IgG : Human Immunoglobulin G detected, no murine immunoglobulinsdetected . EBV: Animals infected with exogeneous EBV as described . No . ;Number of animals. BCL% : Percentage of animals developing human B-celllymphoma. Survival : Survival of animals not developing BCL ; necropsyconfirmation of no tumours. Std Dev: Standard deviation . *P<0.001 versusEBV-seropositive, student t-test.

Tumours typically were found in the liver, theregion of the porta hepatis, the thymus, intestinalmesentery, retroperitoneal lymph nodes and spleen(Fig. 1). Immunohistochemical analysis of fresh-frozen tumours confirmed them to be of human B-cellorigin (data not shown) . Histologic analysis of for-malin-fixed specimens revealed features charac-teristic of aggressive B-cell lymphoma (Fig . 2) .

Effects of cyclosporine on the developmentSCID/BCL

Administration of CsA had no effect on PBL engraft-ment. Furthermore, CsA had no demonstrable effecton the development of B-cell lymphomas if micewere actively infected with exogenous EBV . How-

ever, administration of CsA at each of the dosestested prevented the development of human B-celllymphomas in SCID mice latently infected with EBVfollowing engraftment with PBL from EBV-seroposi-tive donors. One of five mice treated with CsA at adose of 50 mg kg - ' day - ' died following a laboratoryaccident during interim retro-orbital bleed at 77 daysfollowing reconstitution . Autopsy revealed nomacroscopic evidence of BCL development. Resultsare shown in Table 2 .

DISCUSSION

The pathogenesis of lymphoproliferative diseaseand B-cell lymphoma occurring in post-transplant

Table 1 . Human B-cell lymphomadevelopment in hu-PBL-SCID mice :influence of serologic status of PBLdonor and exogenous EBV

Figure 1 . Macroscopic appearanceof human B-cell lymphomasdeveloping in hu-PBL-SCID mice .(a) Ventral surface of liverdemonstrating invasion andreplacement of normal parenchymaby BCL . (b) Mesenteric tumour inclose approximation to stomach andduodenum. (c) Intestinal mesentericBCL . (d) Spleen demonstratingcomplete replacement of upper pole(to right of picture) by tumour .

EBV-Serologic status Hu-IgG EBV No . BCL% Time to BCLof PBL donor

Development (days)Survival(No. BCL)

Mean Range Std Dev

EBV-seropositive

+

- 10 70 89 77-98 ±7.16 180 daysEBV-seropositive

+ 5 100 43 39-57 ±8.11* --EBV-seronegative

+

- 5 0 - - - 180 daysEBV-seronegative

+

+ 5 100 37 35-40 ±4.1* -

Figure 2. Histologic appearance ofhigh-grade large B-cellimmunoblastic lymophoma invadingnormal liver parenchyma . Note themonomorphic appearance oflymphocytoid cells, withconspicuous plasmacytoid cytologicatypia (haematoxylin & eosin, x40) .

Table 2. Human B-cell lymphoma development in hu-PBL SCID mice : effects of cyclosporine

patients remains incompletely defined, but a role ofEpstein-Barr virus and immunosuppression in theaetiology of these conditions appears certain . EBV-infected B-lymphocytes enter a virus-transformed,non-virus-producing cycle in which B-cells arestimulated to proliferate and to produce immuno-globulins [28, 29] . In the transplant recipient, B-cellproliferation may continue unchecked because offailure of the suppressed immune system torespond normally to these cells [6] .

In the immunocompetent patient primary EBVinfection is followed by early humoral and non-specific cell-mediated immune responses . Cell-mediated immune responses are most important inthe outcome of EBV infection [7, 30] . Early prolifera-tive T-cell responses appear to be secondary to

CsA effects on human lymphoma development in vivo 83

Hu-IgG : Human immunoglobulin G detected, no murine immunoglobulins detected . EBV: Animals infected withexogeneous EBV as described . No . : Number of animals . BCL%: Percentage of animals developing human B-celllymphoma. Survival : Survival of animals not developing BCL : necropsy confirmation of no tumours . Std Dev: Standarddeviation .

polyclonal T-cell activation . Relatively non-specificlysis of EBV-infected B-cells by these effectorsresemble natural killer (NK) or lymphokine activatedkiller (LAK) cell killing [31] .

Between 2 and 8 weeks following primary infec-tion, a more EBV-specific T-cell response can beidentified by in vitro proliferation to viral antigenand EBV-specific, HLA-restricted lysis of EBV-infected B cells [32] . This EBV-specific cytotoxic T-cellresponse appears crucial in the control of EBV-induced polyclonal B-cell proliferation, whichsuggests that cytoxic memory T-cells are crucial inthe immune surveillance of latently infected B-cells .

There is much evidence implicating post-trans-plantation immunosuppression in the etiology ofLPD and BCL . Immunosuppressive agents in current

EBV-Serologic status ofPBL donor

Hu-IgG EBV No . BCL% Time to BCLDevelopment (days)

Survival(No. BCL)

Mean Range Std Dev

EBV-seropositive+CsA (10 my kg ' day- ')

+ 10 0 - 140 daysEBV-seropositive I CsA (25 mg kg - ' day- ')

+ 5 0 - 140 daysEBV-seropositive+CsA (50 mg kg - ' day- ')

+

- 4 0 - - 140 daysEBV-seropositive+CsA (10 mg kg - ' day- ')

+

+ 5 100 47 34-51 ±8.5EBV-seronegative+CsA (10 mg kg - ' day- ')

+

+ 5 100 38 35-42 ±3.3

84

T J. Boyle et al .

clinical use function by blockade of certain aspectsof T-cell activation [25] . The net effect is inhibition ofthe functions of cells mediating rejection (CD8'cytotoxic T-cells, CD4' helper T-cells, antibody-forming B-cells and macrophages) . A side effect,however, is the inhibition of EBV-specific cytotoxicT-cell activity, and EBV-induced polyclonal B-cellproliferation may be permitted to continueunchecked. Ultimately, transformed B cells undergomalignant change . Reports of regression of LPD andBCL following reduction or cessation of immuno-suppressive therapy provides further evidence forthe role of immunosuppression in the etiology ofthese conditions in organ transplant recipients [6,33] .

It was anticipated initially that treatment of hu-PBL-SCID with CsA would induce more rapid tumourdevelopment. The results reported were surprisingat first, however it is now felt that EBV reactivationin vivo may require T-cell-13-cell interactions . Stimulithat activate latently infected B-cells in vivo are notdefined. EBV replication can be activated in vitro bytreatment of latently infected cells with a number ifreagents, including 12-O-tetradecanoylphorbol-13-acetate (TPA) and antibody to immunoglobulin [18] .These reagents are thought to act by inducingexpression of ZEBRA, the gene product encoded bythe EBV immediate early gene BZLF1 [34] . Morerecently, it has been shown that expression ofZEBRA is sufficient to activate productive EBV repli-cation [34-36] . While expression of even a smallamount of ZEBRA appears to be sufficient to induceEBV reactivation, spontaneous reactivation in vitrooccurs at a rate of only one in 10 3 to one in 106B-cells, suggesting that activation of the BZLF1 geneis a rare event.

T-lymphocyte activation in response to antigenmay lead to activation of EBV by the production ofcytokines . Following binding of T-cells to antigen byspecific cell surface receptors, activation is initiated[23] . Bound antigen induces second messenger for-mation inn the cytoplasm. Second messengersinduce IL-1 production by activating T-cells, which inturn induces IL-6 production by macrophages .T-cells are stimulated by antigen plus IL-6 to secreteIL-2 (T-cell growth factor) and to express de novocell surface IL-2 receptors . In addition to its functionas an autocrine growth factor for T-cells, IL-2 servesa crucial role in eliciting the secretion by T-cells ofbioactive lymphokines that activate T-cells, B-cellsand macrophages .

Antigen-activated, IL-2-stimulated T-cells releaseB-cell activating factors, including IL-4 and IL-5, thatstimulate antigen-activated B-cells to proliferate andelaborate high affinity, high titre anti-antigen anti-bodies. We postulate that in the SCID mouse modelof human B-cell lymphoma these factors are impor-tant in the activation of EBV-latently infected B-cells,possibly via an effect on BZLF1 expression . Wespeculate that following reconstitution of mice withPBL a degree of graft-versus-host disease (GVHD)occurs, with polyclonal T-cell activation . B-cellgrowth factors elaborated by T-cells activate latentlyinfected B-cells, and polyclonal proliferation pro-ceeds to the development of malignant lymphoma .The administration of cyclosporine to mice followingtheir engraftment with EBV-seropositive PBL inhibitsthis process . If this is indeed the case, similar eventsmay underlie, at least in part, the pathogenesis ofLPD and BCL in transplant recipients.

An anticipated extension of the reported experi-ments is the engraftment of defined subpopulationsof T- or B-cells into SCID mice to determine the con-tribution of lymphocyte subtypes to lymphomadevelopment. Engraftment of such defined sub-populations into SCID mice has not been reported todate. A report of PBL engraftment in another murinemodel (beige/nude/xid mice) [37] suggests that co-engraftment of T lineage cells is important forlymphoma development . Beige/nude/xid miceselectively accept human B-cell but not T-cell grafts .These mice do not develop BCL when engraftedwith EBV seropositive donor PBL, but developlymphomas when exogenous EBV is injected .

The SCID mouse provides an ideal model for thestudy of human B-cell lymphoma . Elucidation of themechanisms underlying the aetiology and patho-genesis of this disease process may have importantimplications for the development of strategies forthe prevention, detection and treatment of lympho-proliferative disease and malignant B-cell lymphomaassociated with immunosuppression in the trans-plant recipient .

REFERENCES

1 . Zitelli BJ, Malatack JJ, Urbach AH, et al. Pediatrichepatology : a three-year experience with pediatricliver transplantation . In : Winter PM, Kang YG, eds .Hepatic transplantation : anaesthetic and perioperativemanagement. Westport: Praeger Special Studies,Praeger Scientific 1986 : 61-73 .

2 . Penn I . The occurrence of cancer in immune deficien-

cies. CurrProb Cancer 1982; 6(10) : 1-64 .

3. Penn I, Brunson ME. Cancers after cyclosporine

therapy . Transplant Proc 1988; 20 (suppl . 3( : 885-92 .

4. Penn I . Cancer is a long term hazard of immuno-

suppressive therapy, JAutoimmunity 1988 ; 1 : 545-58 .

5. Penn I. Why do immunosuppressed patients develop

cancer? In : Pimental E, ed . CRC Critical Reviews in

Oncogenesis. Boca Raton : CRC Press Inc . 1989 :

27-52 .

6. Malatack JJ, Gartner Jr JC, Urbach AH, Zitelli BJ .

Orthotropic liver transplantation, Epstein-Barr virus,

cyclosporine, and lymphoproliferative disease: a

growing concern . JPediatr 1991 ; 118 : 667-75 .

7. Okano M, Thiele GM, Davis JR, Grierson HL, Purtilo

DT. Epstein-Barr Virus and Human Diseases : Recent

Advances in Diagnosis . Clin Microbiol Rev 1988; 1 :

300-12 .

8. Kinlen JJ, Incidence of Cancer in Rheumatoid Arthritis

and Other Disorders after Immunosuppressive Treat-

ment . Am J Med 1985 ; 78 (suppl . 1A) : 44-9.

9. Klein G, Purtilo DT, Symposium on Epstein-Barr virus-

induced lymphoproliferative diseases in immuno-

deficient patients . Cancer Res 1981 ; 41 : 4209-302 .

10. Straugh B, Andrews L-L, Seigel N, Miller G . Oro-

pharyngeal excretion of Epstein-Barr virus by renal

transplant recipients and other patients treated withimmunosuppressive drugs . Lancet 1974 ; 1 : 234-7 .

11 . Borisch-Chappius B, Nezelof C, Muller H, Muller-

Hermelink HK . Different Epstein-Barr virus expression

in lymphomas from immunocompromised patients .

Am J Pathol 1990 ; 136: 751-8 .

12, deThe G, Zeng Y. Population screening for EBV

markers : toward improvement of nasopharygeal

carcinoma control . In: Epstein MA, Achong BG, eds .

The Epstein-Barr virus: Recent Advances . New York :

John Wiley & Sons 1986; 237-49 .

13. deThe G, Geser A, Day NE, et al. Epidemiologic

evidence for causal relationship between Epstein-Barr

virus and Burkitt's lymphoma from Ugandan prospec-

tive study . Nature 1978; 274 : 756-61 .14. Henle G, Henle W, Clifford P, et at Antibodies to

Epstein-Barr Virus in Burkitt's lymphoma and controlgroups . J Natl Cancer Institute 1969 ; 43 : 1147-57 .

15. Henle W, Henle G, Ho JH . Epstein-Barr virus-related

serology in nasopharyngeal carcinoma and controls,

1ARC Sci Pub! 1978 ; 20 : 427-37 .16. Rowe M, Rowe DT, Gregory CD, et al. Differences in B

cell growth phenotype reflect novel patterns of

Epstein-Barr virus latent gene expression in Burkitt's

lymphoma cells . EMBO J 1987; 6: 2743-51 .

17. Young LS, Dawson CW, Clark D, et al. Epstein-Barrvirus gene expression in nasopharyngeal carcinoma .

JGen Virol1988 ; 69 : 1051-65 .

CsA effects on human lymphoma development in vivo

85

18. Kieff E, Liebowitz D . In : Fields BN, ed . Virology. New

York : Raven Press 1990 : 1889-920 .

19. Bosma GC, Custer RP, Bosma MJ . A severe combined

immunodeficiency mutation in the mouse . Nature

1983; 301 : 527-30 .

20. McCune JM, Namikawa R, Kaneshima H, Schultz LD,

Lieberman M, Weissman IL. The SCID-hu mouse :

murine model for the analysis of human hemato-

lymphoid differentiation and function . Science 1988 ;

241 : 1632-9 .

21 . Mosier DE, Gulizia RJ, Baird SM, Wilson DB. Transfer

of a functional human immune system to mice with

severe combined immunodeficiency. Nature 1988 ;

335:1333-7.

22. Cannon MJ, Pisa P, Fox RI, Cooper NR . Epstein-Barr

Virus Induces Aggressive Lymphoproliferative Dis-

orders of Human B Cell Origin in SCID/hu Chimeric

mice . J Clin Invest 1990; 85: 1333-7 .

23. Moss DJ, Rickenson AB, Pope JH . Long term T-cell-

mediated immunity to Epstein-Barr virus in man . Ill .

Activation of cytotoxic T-cells in virus-infected leuko-

cyte cultures . IntJCancer 1979; 23 : 618 .

24. Bird AG, McLachlan SM, Britton S . Cyclosporin A pro-

motes spontaneous outgrowth in vitro of Epstein-Barr

virus-induced B-cell lines . Nature 1981 ; 289: 300-1 .

25. Strom TB . Immunosuppression in tissue and organ

transplantation . In : Brent L, Sells RA, eds . OrganTransplantation : Current Clinical and Immunological

Concepts . London : Bailliere Tindall 1989 : 39-56 .

26. Okano M, Taguchi Y, Nakamine H, et al. Charac-terization of Epstein-barr virus-induced lymphoproli-

feration derived from human peripheral blood

mononuclear cells transferred to severe combined

immunodeficiency mice . Am J Pathol 1990; 137 :

517-22 .

27. Nakamine H, Yokote H, Itakura T, et al. Non-Hodgkin's

lymphoma involving the brain . Diagnostic usefulness

of stereotactic needle biopsy in combination withparaffin section immunohistochemistry . Acre Neuro-

pathol 1989; 78: 462-71 .

28 . AlIday M, Crawford DH . Role of epithelium in EBV per-sistance and pathogenesis of B-cell tumors . Lancet

1988 ; 1 : 855-7 .

29. Henle W, Henle G, Lennette E . The Epstein-Barr Virus .

Scientific American 1979 ; 241 : 48-59 .

30. Svedmer E, Jondal M . Cytoxic effector cells specific

for B cell lines transformed by Epstein-Barr virus are

present in patients with infectious mononucleosis .

Proc NatAcad Sci USA 1975 ; 72: 1622-6 .

31 . Masucci MG, Bejarano MT, Masucci G, Klein E. Large

granular lymphocytes inhibit the in vitro growth of

autologous Epstein-Barr virus-infected B cells . Cell

lmmunol1983 ; 76: 311 .

32. Moss DJ, Rickenson AB, Wallace LE, Epstein MA .

86

T. J. Boyle et al .

Sequential appearance of Epstein-Barr virus nuclear

and lymphocyte-detected membrane antigens in B cell

transformation . Nature 1981 ; 291 : 664 .

33. Penn I . Principles of Tumor Immunity; Immuno-

compromized Patients . In : DeVita Jr VT, Hellman S,

Rosenberg SA, eds . AIDS-Etiology, Diagnosis, Treat-ment and Prevention . Philadelphia : WB Saunders

1990: 1-14 .34. Miller G. The switch between latency and replication

of Epstein-Barr virus, J Infec Dis 1990; 161 : 833-44-35. Countryman J, Jewson H, Seibl R, Wolf H, Miller G .

Polymorphic proteins encoded within BZLF1 of defec-

tive and standard Epstein-Barr viruses disrupt latency .

J Virol 1987 ; 61 : 3672-9 .

36. Rooney CM, Rowe DT, Ragot T, Farrell PJ . The spliced

BZLF1 gene of Epstein-Barr virus IEBV) transactivates

an early EBV promoter and induces the virus produc-

tive cycle . J Virol 1989; 63: 3109-16 .

37. Dosch HM, Cochrane DMG, Cook VA, Leader JS,

Cheung RK. Exogenous but not endogenous EBV

induces lymphomas in beige/nude/xid mice carrying

human lymphoid xenografts . Int Immunol 1991 ; 3 :731-35 .