Exp Toxic Pathol 1992; 44: 375-380 Gustav Fischer Verlag Jena Cancer Research Center of the Russian Academy of Medical Science!), Moscow, Russia International Agency for Research on Cancer2), Lyon, France
Metastatic capacity of BURKITT'S lymphoma cell lines in nude mice V. E. GURTSEVITCH 1), V. S. TURUSOVI) and, G. M. LENOIR2) With 3 figures and 2 tables Received: April 24, 1991; Revised: October 28,1991; Accepted: November 21,1991 Address for correspondence: Dr. V. E. GURTSEVITCH, Cancer Research Center, Kasltirskoe shosse 24, 115478, Moscow, Russia. Key words: BURKITT'S lymphoma; Lymphoma, BURKITT'S; Metastases
Summary BURKITT'S lymphoma (BL) cell lines estimated in a previous study as having a high, low and no tumourigenicity (7) were intravenously (i.v.) injeced into preirradiated (480 rad) nude mice. BL cell lines with a high tumourigenic potential produced metastatic tumours in the brain, spinal cord, bone marrow, stomach and kidney, but did not disseminate into the lung, liver, ovary and spleen. The survival time of the tumour bearing animals ranged from 2 to 10 weeks. The majority of mice i.v. injected with highly tumourigenic BL cell lines showed paresis or paralysis of the hind legs. This was associated with the presence of neoplastic nodules either in the brain and/or in the spinal cord. In animals with metastasis to the stomach and kidney progressive cachexy was observed. The described experimental model of metastatic BL tumours in nude mice can effectively be used for the in vivo study of new therapeutic molecules such as monoclonal antibodies coupled or not to substances, toxic to tumour cells. This model can also be useful for the identification and analysis of homing properties of BL cells and their implication in BL pathogenesis.
Introduction BL is a unique human tumour for which a number of stages have been suggested and in which such factors as a virus (Epstein-Barr virus, EBV), specific chromosomal translocations (t(8; 14), t(8; 22) and (2; 8)), and deregulation of the proto-oncogene (c-myc) expression are involved (10). The evaluation of the significance of events associated with a cascade of cellular responses triggered by oncogene activation may lead to development of effective therapeutical approaches based on the intervention at some stages of this cascade to stop or even to prevent the malignant process. An animal model of metastatic BL tumours in nude mice would be very essential ana useful for these purposes. In the present paper, we describe the metastatic model of BL in nude mice. In particular, we demonstrate that pre-
irradiated (480 rad) Swiss nu/nu mice being intravenously (i/v) injected with BL cell lines (prevously estimated (7) as highly tumourigenic for these animals under s.c. inoculation) developed tumours in such organs as the brain, spinal cord, bone marrow, stomach and kidney. Nude mice with metastasis to brain and/or spinal cord revealed the paresis or paralysis of the hind legs while the animals with tumour growth in the kidney or stomach showed progressive cachexy.
Materials and methods Cell lines BL lines with high (BL 2, BL 29, BL 30 and BL 60), low (BL 57 and BL 72) and no (BL 16 and BL 37) tumorigenicity, as reported in a previous paper (7), were selected for the intravenous inoculation. Two lymphoblastoid cell lines (LCLs), i.e. IARC 139 established from peripheral blood lymphocytes of a BL patient, and IARC 728 established from a non BL patient were also i. v. inoculated. All cell lines were cultured in RPMI -1640 medium supplemented with 20 % foetal calf serum and antibiotics. The cells were harvested and inoculated into nude mice only if their viability, checked by Trypan-blue dye, was over 75%.
Animals and experimental conditions Five to six-week old female nu/nu mice outbred from Swiss stock (IFFAICREDO, Lyon, France) were used. One or two days before inoculation, the animals underwent whole-body irradiation at a dose of 4.8 Gy (137 Cs with 0.48 Gy/min). Groups of 3-5 mice were used to evaluate the metastatic capacity of each line; lOx 10 cells Exp Toxic Pathol 44 (1992) 7
375
were injected into the lateral tail vein in a volume of 0.2 rn1 of the complete cultural medium. Each cell line was independently tested 2 or 3 times. The animals were observed weekly during 12 weeks. Those animals which showed progressive cachexy or any neurological symptoms were sacrificed earlier.
Histopathology All animals were necropsied, and the brain, vertebral column with spinal cord, lungs, liver, kidney, spleen, lymph nodes, ovary, heart and other organs were fixed in 10 % buffered formalin and examined histologically. Paraffin sections were stained with haematoxylin and eosin.
Results Three selected groups of BL cell lines, estimated in previous study as having a high, low or no tumorigenicity (7), as well as two LCLs, were repeatedly tested under the described conditions. A total of 99 nude mice were used in this study. Practically the same results were obtained with each cell line in the different experiments, which are summarized in table 1. As seen, the metastasizing capacity of BL cell lines differed significantly. Only BL lines with a
in 11 of 36 (31 %), stomach - in 7 of 36 (19 %) and kidney - in 7 of 41 (17 %). Each BL line, even belonging to the same tumourigenic potential group revealed, however, an individual organ preference for metastases. For instance, BL 02 and BL 60 colonized the brain more efficiently than BL 29 and BL 30: in 8 of 10 (80%) and in 7 of 9 (78 %) against 4 of 10 (40 %) and 4 of 10 (40 %) ilv inoculated animals, respectively. The frequency of metastatic tumours in other organs differed even more significantly. In particular, the rate of tumour metastatic growth in the spinal cord ranged from 0% for BL 30 to 75 % (6/8) for BL 60; in the bone marrow, from 0 % for BL 30 to 44 % (4/9) for BL 29 and BL 60; in the stomach, from 0% for BL 30 and BL 60 to 44 % (4/9) for BL 29; and in the kidney, from 0 % for BL 60 to 42 % (5/12) for BL 29. Affected nude mice showed different clinical manifestations, however, the lungs and the liver were not affected at all. Those with metastases in the brain and spinal cord revealed the spectrum of neurological manifestations, from paresis to paralysis of the hind legs. Hard breathing was also frequenly observed. Nude mice with metastatic tumour growth in the kidney or stomach demonstrated a progressive loss of weight leading to cachexy. Simultaneous tumour development in the brain and in any of the mentioned
Table 1. Metastatic capacity of different BL cell-lines after intravenous injection into nude mice. Cell lines
Tumorigenicity*
No of No of experiments injected animals
Number of mice with tumours/Number of injected mice Brain
Spinal cord
Affected organs Bone Kidney marrow
Stomach
02 29 30 60
High
2 3 2 2
10 14 10 9
8/10 4/10 4/10 7/9
3/6 3/8 0/8 6/8
3/10 4/9 0/8 4/9
1/10 5/12 1/10 0/9
3/10 4/9 0/10 017
BL 57 BL 72
Low
3 2
12 7
0/12 017
0/12 017
0/12 017
0/12 117
0/12 017
BL 16 BL 37
No
3 2
13 8
0/13 0/8
0/13 0/8
0/13 0/8
0/13 0/8
0/13 0/8
IARC 139 IARC 728
No
2 2
8 8
0/8 0/8
0/8 0/8
0/8 0/8
0/8 0/8
0/8 0/8
BL BL BL BL
*
Tumorigenicity was assessed in earlier experiments on the basis of subcutaneous injection (Ref. 7).
high tumourigenic potential (BL 02, BL 25, BL 30 and BL 60) metastasized effectively in various organs, while those with low tumorigenicity (BL 57, and BL 72) being i. v. inoculated into 19 nude mice initiated metastatic tumour only in one (kidney). Non-tumourigenic BL lines (BL 16 and BL 37), as well as LCLs (lARC 139 and IARC 728) were unable to metastasize to any organ. For BL lines with a high tumourigenicity, the ,.overall frequency of metastatic growth in different organs was as follows: brain - in 23 of 39 (59%), spinal cord - in 12 of30 (40%), bone marrow376
Exp Toxic Pathol44 (1992) 7
visceral organs (the kidney, stomach) was accompanied by a combined symptomatology. Metastatic growth in the brain in some cases was confined only to meninges while in other mice it spread to brain tissue. Not infrequently, both meningeal infiltration and numerous perivascular tumour deposits were present (fig. 1). In few cases accumulation of tumour cells replaced an important part of the hemisphere and they were seen with the naked eye. In the stomach, both parts, forestomach and glandular stomach or either of them, could be affected. Initial stages of
Fig. 1. BL Cell infiltration in the brain. I a - meninges are diffusely infiltrated with BL cells; small perivascular infiltrates in the brain tissue. H. & E. x80. 1b - large BL cell infiltrates in the brain tissue. H. & E. x 140. tissue with a subsequent destruction of the epithelium and the tumour cell spread to the muscular layer and serosa . Macroscopically, the diffuse or local thickening of the stomach wall was observed leading in some cases to the transformation of the stomach into huge tumourous mass. The lymphomatous infiltration in the kidneys was identified either microscopically only or had a form of a huge tumour replacing the kidney almost completely. Lymphoma cells had a big nucleus of a round, oval or irregular shape with a fine chromatin and easily discernable nucleolus (fig. 3). In some areas, mitoses were frequent, in others many nuclei underwent pycnotic changes . "starry-sky" pattern was not observed.
Discussion
Fig. 2. Diffuse BL cell infiltration under the squamous epithelium of the forestomach . H. & E. x500. metastasis were manIfested in the focal infiltration by tumorcells of the mucous membrane (fig. 2). This progressed to the total replacement of the mucous membrane by tumour 12
The majority of BL cell lines like those of both primary and metastatic human tumours of other origin grow successfully as a subcutaneous graft in nude mice, but manifest limited if no metastatic protential (3, 6, 19). In our experiments, we demonstrated that i . v. inoculation of BL cells gave rise to formation of metastases in various organs, namely in the brain, spinal cord, bone marrow, stomach, and kidney. The frequency and the spectrum of the affected organs did not depend on the patient's origin (Caucasian, Reunion or North Africa), the origin of biopsied tumour specimen used for establishment of BL cell lines (lymph nodes, abdomen, bone marrow, maxillar and so on), the presence of EBV or type of cytogenetic translocation (8: 14 or 8:22) (table 2). On the other hand it was found that the metastatic capacity of BL cell lines in nude mice strictly correlated with their tumourigenicity upon s.c. inoculation (7) . Only tumourigenic BL cell lines possessed the capacity to form experimental metastases; non tumourigenic BL lines did not colonize any organ (BL 16, BL 37) and behaved like LCLs (IARC 139, Exp Toxic Pathol 44 (1992) 7
377
Fig. 3. BL cell infiltration of the kidney. 3 a - large infiltrate in the kidney cortex. H. & E. x200. 3 b - to demonstrate the fine structure of BL cells. H. & E. X 1200.
Table 2. The main characterics of BL cell lines used for intravenous injection into nude mice. Nb 1 2 3 4 5 6 7 8 9 10
Cell lines
Patient/Origin
Specimen Origin
BL 02 BL 16 BL 29 BL 30 BL 37 BL 57 BL 60 BL 72 IARCIl39 IARCn28
Caucasian ReuniQP Reunion Caucasian Caucasian Caucasian Nth African Nth African LCL (BL 30) LCL (Lymphoma)
Biopsy LN (Dx) Abd. Asc. (Dx) Abd. Biopsy (Dx) Bone marrow (R) Bone marrow (Dx) Periph. Blood (+ B 95) Maxill. Punct. (Dx) Abd. Biopsy (Dx) Blood (+ B 95) Blood (+ B 95)
Presence Cytogenetic EBV translocation
+ + + + + + +
t(8; t(8; t(8; t(8; t(8; t(8; t(8; t(8;
22) 14) 14) 14) 22) 14) 22) 14)
LN - Lymph node Dx - specimen was taken at the moment of diagnosis R - specimen was taken at the state of relapse + B 95 - peripheral blood lymphocytes were infected with B 95 EBV strain
IARC 728). It was recently reported, however, that human EBV-positive LCL, tenned ZS, after irradiation in vitro aquired the capacity to invade the lymph nodes of nude mice (17). Such unusual feature of mentioned LCL can be explained by the effect of irradiation, which was likely the cause of chromosomal rearrangements in ZS karyotype and may have induced mutation of a number of relevant oncogenes. The significance of c-myc deregulation for an unlimited in vivo growth potential of B lymphoblastoid cells is a subject of many investigations (8, 11). In particular, it was recently demonstrated that in vivo down-regulation of the c-myc gene does not contribute to the-BVLCL hybrid graft regression in nude mice. The level of the c-myc transcription pattern detected in the parental BL cell lines was the same in both 378
Exp Toxic Pathol 44 (1992) 7
growing and regressing hybrid grafts whose in vivo growth phenotype was indistinguishable from that of the parental LCL (20). Our data are in line with the findings of GHETIE et a1. (5) who recently demonstrated that following i.v. inoculation the human BL cell line Daudi could be grown in scm mice as dissiminated tumours affecting the lungs, kidneys, ovaries and adipose tissue. Microscopic tumour infiltrates were also found in the spleen, bone marrow, spinal column and femur. In contrast to other findings in the current study tumour cells disseminating into the lung, liver, ovaries and spleen have never been observed. As distinct from human BL no "starry-sky" pattern was observed histologically. Likewise "starry-sky" was only rarely found after s.c. inoculation of the same BL lines (7). Our data are also
,-,
L'-"J"
.................
"" ........... _ .... _ ..... "" ....
_-... ..
r ...
~r-~
... J
-~·r
...
-~
...... .... J
~
...
~-
...
4. GALLATIN M, ST JOHN TP, SIEGELMAN M, et al.: Lymphocyte homing receptors. Cell 1986; 44: 673-680. 5. GHETIE M-A, RICHARDSON J, TUCKER T, et al.: Dissiminated or localized growth of a human B-cell tumour (DAUDI) in scm mice. Int J Cancer 1990; 45: 481-485. 6. GIOVANELLA BC, FOGH 1: The nude mouse in cancer research. Advanc Cancer Res 1985; 44: 69-120. 7. GURTSEVITCH VE, O'CONOR GT, LENOIR GM: BURKITT'S lymphoma cell lines reveal different degrees of tumorigenicity in nude mice. Int J Cancer 1988; 41: 87-95. 8. HOTCHIN NA, ALLDAY MJ, CRAWFORD DH: Deregulated c-myc expression in EpSTEIN-BARR virus immortalized Bcells induces altered growth properties and surface phenotype but not tumorigenecity. Int J Cancer 1990; 46: 566-571. 9. LEMERLE J: Rapporteur's Report. In: BURKITT'S Lymphoma: a Human Model. (eds. GM LENOIR, GT O'CONOR, CLM OLWENY), pp. 149-151. IARC Scientific Publications No. 60, Lyon 1985. 10. LENOIR GM: Role of the virus, chromosomal translocations and cellular oncogenes in the aetiology of BURKITT'S lymphoma. In: The EpSTEIN-BARR Virus: Recent Advances. (eds MA EpSTEIN, BG ACHONG), pp. 183-200. William Heinemann Medical Books 1986. 11. LOMBARDI L, NEWCOMB EW, DALLA-FAVERA R: Pathogenesis of BURKITT lymphoma: expression of an activated c-myc oncogene causes the tumorigenic conversion of EBV-infected human B-Iymphoblasts. Cell 1987; 49: 161-170. 12. MAGRATH IT, SARIBAN E: Clinical feature of BURKITT'S lymphoma in the USA. In: BURKITT'S Lymphoma: a Human Model. (eds. GM LENOIR, GT O'CONOR, CLM OLWENY), pp. 119-127. IARC Scientific Publications No. 60, Lyon 1985. 13. NICOLSON GL, DULSKI K, BASSON C, et al.: Preferential organ attachment and invasion in-vitro by B 16 melanoma cells selected for differing metastatic colonization and invasive properties. Invas Metast 1985; 5: 144-158. 14. NICOLSON GL, DULSKI KM: Organ specificity of metastatic tumor colonization is related to organ-selective groth properties of malignant cells. Int J Cancer 1986; 38: 289-294. 15. NKRUMAH FK, OLWENY CLM: Clinical features of BURKITT'S lymphoma: The African Experience. In BURKITT'S Acknowledgement: The authors are indebted Mrs. M. Lymphoma: a Human Model. (eds. GM LENOIR, GT LAVAL and Mrs. N. LYANDRAT for technical help and Mr. O'CONOR, CLM OLWENY), pp. 87-95. IARC Scientific 1. MaLON for the preparation of photographs. Publications No. 60, Lyon 1985. 16. PHILIP T: BURKITT'S lymphoma in Europe. In: BURKITT'S References lymphoma: a Human Model. (eds. GM LENOIR, GT O'CONOR, CLM OLWENY), pp. 107 -118. IARC Scien1. ANAISSIS E, GEHA S, ALLAM C, et al.: BURKITT'S Lymtific Publications No. 60, Lyon 1985. phoma in the Middle East. In: BURKITT'S Lymphoma: a 17. RONCELLA S, RAMARLI D, COOPER MK, et al.: EstablishHuman model. (eds. GM LENOIR, GT O'CONOR, CLM' ment of an EBV -positive lymphoblastoid cell line that OLWENY) pp. 129-134. IARC Scientific Publications grows as a lymphoma in nude mice and expresses memNo. 60, Lyon 1985. brane CD2 molecules. Int J Cancer 1990; 45: 299-307. 2. BERNSTEIN SC, WEINBERG RA: Expression of the metas18. THORGEIRSSON UP, TURPEENIEMI-HIJANEN T, WILLIAMS tatic phenotype in cells transfected with human DNA. JE, et al.: NIH/3T3 cells transfected with human tumor Proc Natl Acad Sci USA 1985; 82: 1726-1730.
experiments only tumourigenic cell lines possesed the metastatic capacity. The mechanism by which BL and other tumor cells preferentially form metastases in certain organs, is still unknown. From experiments with mouse 816 melanoma cell sublines, one can conclude that organ specificity of metastatic tumor colonization is determined by differences in the ability of the tested cells to adhere to and invade particular target organs (14) and/or by organ-selective growth properties of malignant cells (13). On the other hand, it should not be excluded that the mechanism of migration of circulating cells from the blood stream to a certain organ could be similar to that of specific recognition and adherence of mature lymphocytes, possessing so-called homing receptors, to the specialized high endothelial cells of the postcapillar venules (HEV) located in some lymphoid organs (see for review 4). Recognition of such receptors on the surface of metastazing BL cells would confirm such a suggestion. When comparing the pattern of metastasis in our experiments to that of human BL both the common features and dissimilarities can be noted. The spectrum of organs affected in nude mice reminds, on the whole, of that of human BL with exception of the lungs, liver and ovaries, which are frequently involved in human BL (9) and were never affected in nude mice. The jaw which is a typical site of tumour manifestation in African BL (21, 15) was never a target organ in mice. But one has to remember that in areas other than Africa the jaw is also only rarely, - if ever involved in BL (1, 16, 12). Different BL cell lines showed dissimilar organ distribution of metastasis as well. Thus, the model we have developed might represent an effective and reptoducible metastatic model system to study drug targeting by monoclonal antibodies since testing of monoclonal antibodies coupled to toxins has usually been performed either in in vitro systems, or in vivo on mouse leukemia. Our model can be also useful for the study of Burkitt's and other human lymphoma's pathogenesis as well as for analysis of homing properties of lymphomatous cells.
12*
Exp Toxic Pathol 44 (1992) 7
379
DNA containing activated ras oncogenes express themetastatic phenotype in nude mice . Molec Cell Bioi 1985; 5: 259-262. 19. WANG WR, SORDAT B, PIGUET D, et al.: Human colon tumors in nude mice: implatation site and expression of invasive phenotype. In: Immune-deficient Animals. (ed. B SORDAT), pp. 239-245. Karger, Basel, 1984.
20. WOLF J, KLEVENTZ B, PAVLITA M, et al.: Regressing nude mose grafts of BURKITTS'S lymphoma x Iymphoblastoid cell hybrids show deregulation of the c-myc gene and expression of the EBV latent membrane protein. Int J Cancer 1991; 47: 99-104. 21. ZIEGLER JL, BLiMING AZ, FASS L, et al.: Relapse patterns in Burkitt's lymphoma. Cancer Res 1972; 32: 1267-1272.
Exp Toxic Pathol 1992; 44: 380 Gustav Fischer Verlag lena
Announcement
The Ernest Witebsky Center for Immunology The following will be offered this Spring by The Ernest Witebsky Center for Immunology, University at Buffalo, State University of New York:
Recombinant DNA technology: methods and applications Hands-on laboratory course Monday, April 19, through Sunday, April 25, 1993 This seven-day course, presented in conjunction with ExonIntron, Inc., provides both fundamental and advanced training in recombinant DNA technology . It is designed to provide and expand laboratory skills in molecular cloning techniques for individuals with a background in biological sciences but who have little or no knowledge of recombinant DNA techniques. Course registration fee of $ 1,200.00 covers course materials, beverage breaks, and seven (7) days' luncheons. Registration will be limited to 24 persons.
Methods of immunologic research and diagnosis An at-the-bench laboratory training program
Monday, May 10, through Thursday, May 20, 1993 This ten-day program consists of daily practical laboratory exercises personally carried out by participants, accompanied by lectures, demonstrations and discussions. Exercises range
380
Exp Toxic Pathol 44 (1992) 7
from the basic techniques of antiserum preparation to advanced techniques of immunoassay. Open to individuals at the postdoctoral level, supervisory senior technologists and graduate students. Course registration of $ 700.00 includes course materials and beverage breaks. Registration will be limited to 36 individuals.
Latex particle immunoassays Hands-on laboratory course
Friday and Saturday, May 21-22,1993 This course is designed to provide an overview of preparation, formulation, properties and applications of latex particles in medical laboratory diagnosis . Geared toward laboratory researchers and technicians, and in-vitro diagnostic manufacturers. Course registration fee of $ 295.00 covers course materials, beverage breaks, and lunch on both days. Registration will be limited to 24 persons.
For further information on any of the above courses, contact Roger K. Cunningham, Ph.D., Director, The Ernest Witebsky Center for Immunology, 423 Sherman Hall, University at Buffalo, 3435 Main Street, Buffalo, New York 14214; Phone: 716/ 829-2901, Fax: 716/829-2158. Because class size is limited in all courses, we recommend registrations be submitted at least one month prior to respective starting dates for best possibility of acceptance.
Metastatic capacity of BURKITT'S lymphoma cell lines in nude mice V. E. GURTSEVITCH 1), V. S. TURUSOVI) and, G. M. LENOIR2) With 3 figures and 2 tables Received: April 24, 1991; Revised: October 28,1991; Accepted: November 21,1991 Address for correspondence: Dr. V. E. GURTSEVITCH, Cancer Research Center, Kasltirskoe shosse 24, 115478, Moscow, Russia. Key words: BURKITT'S lymphoma; Lymphoma, BURKITT'S; Metastases
Summary BURKITT'S lymphoma (BL) cell lines estimated in a previous study as having a high, low and no tumourigenicity (7) were intravenously (i.v.) injeced into preirradiated (480 rad) nude mice. BL cell lines with a high tumourigenic potential produced metastatic tumours in the brain, spinal cord, bone marrow, stomach and kidney, but did not disseminate into the lung, liver, ovary and spleen. The survival time of the tumour bearing animals ranged from 2 to 10 weeks. The majority of mice i.v. injected with highly tumourigenic BL cell lines showed paresis or paralysis of the hind legs. This was associated with the presence of neoplastic nodules either in the brain and/or in the spinal cord. In animals with metastasis to the stomach and kidney progressive cachexy was observed. The described experimental model of metastatic BL tumours in nude mice can effectively be used for the in vivo study of new therapeutic molecules such as monoclonal antibodies coupled or not to substances, toxic to tumour cells. This model can also be useful for the identification and analysis of homing properties of BL cells and their implication in BL pathogenesis.
Introduction BL is a unique human tumour for which a number of stages have been suggested and in which such factors as a virus (Epstein-Barr virus, EBV), specific chromosomal translocations (t(8; 14), t(8; 22) and (2; 8)), and deregulation of the proto-oncogene (c-myc) expression are involved (10). The evaluation of the significance of events associated with a cascade of cellular responses triggered by oncogene activation may lead to development of effective therapeutical approaches based on the intervention at some stages of this cascade to stop or even to prevent the malignant process. An animal model of metastatic BL tumours in nude mice would be very essential ana useful for these purposes. In the present paper, we describe the metastatic model of BL in nude mice. In particular, we demonstrate that pre-
irradiated (480 rad) Swiss nu/nu mice being intravenously (i/v) injected with BL cell lines (prevously estimated (7) as highly tumourigenic for these animals under s.c. inoculation) developed tumours in such organs as the brain, spinal cord, bone marrow, stomach and kidney. Nude mice with metastasis to brain and/or spinal cord revealed the paresis or paralysis of the hind legs while the animals with tumour growth in the kidney or stomach showed progressive cachexy.
Materials and methods Cell lines BL lines with high (BL 2, BL 29, BL 30 and BL 60), low (BL 57 and BL 72) and no (BL 16 and BL 37) tumorigenicity, as reported in a previous paper (7), were selected for the intravenous inoculation. Two lymphoblastoid cell lines (LCLs), i.e. IARC 139 established from peripheral blood lymphocytes of a BL patient, and IARC 728 established from a non BL patient were also i. v. inoculated. All cell lines were cultured in RPMI -1640 medium supplemented with 20 % foetal calf serum and antibiotics. The cells were harvested and inoculated into nude mice only if their viability, checked by Trypan-blue dye, was over 75%.
Animals and experimental conditions Five to six-week old female nu/nu mice outbred from Swiss stock (IFFAICREDO, Lyon, France) were used. One or two days before inoculation, the animals underwent whole-body irradiation at a dose of 4.8 Gy (137 Cs with 0.48 Gy/min). Groups of 3-5 mice were used to evaluate the metastatic capacity of each line; lOx 10 cells Exp Toxic Pathol 44 (1992) 7
375
were injected into the lateral tail vein in a volume of 0.2 rn1 of the complete cultural medium. Each cell line was independently tested 2 or 3 times. The animals were observed weekly during 12 weeks. Those animals which showed progressive cachexy or any neurological symptoms were sacrificed earlier.
Histopathology All animals were necropsied, and the brain, vertebral column with spinal cord, lungs, liver, kidney, spleen, lymph nodes, ovary, heart and other organs were fixed in 10 % buffered formalin and examined histologically. Paraffin sections were stained with haematoxylin and eosin.
Results Three selected groups of BL cell lines, estimated in previous study as having a high, low or no tumorigenicity (7), as well as two LCLs, were repeatedly tested under the described conditions. A total of 99 nude mice were used in this study. Practically the same results were obtained with each cell line in the different experiments, which are summarized in table 1. As seen, the metastasizing capacity of BL cell lines differed significantly. Only BL lines with a
in 11 of 36 (31 %), stomach - in 7 of 36 (19 %) and kidney - in 7 of 41 (17 %). Each BL line, even belonging to the same tumourigenic potential group revealed, however, an individual organ preference for metastases. For instance, BL 02 and BL 60 colonized the brain more efficiently than BL 29 and BL 30: in 8 of 10 (80%) and in 7 of 9 (78 %) against 4 of 10 (40 %) and 4 of 10 (40 %) ilv inoculated animals, respectively. The frequency of metastatic tumours in other organs differed even more significantly. In particular, the rate of tumour metastatic growth in the spinal cord ranged from 0% for BL 30 to 75 % (6/8) for BL 60; in the bone marrow, from 0 % for BL 30 to 44 % (4/9) for BL 29 and BL 60; in the stomach, from 0% for BL 30 and BL 60 to 44 % (4/9) for BL 29; and in the kidney, from 0 % for BL 60 to 42 % (5/12) for BL 29. Affected nude mice showed different clinical manifestations, however, the lungs and the liver were not affected at all. Those with metastases in the brain and spinal cord revealed the spectrum of neurological manifestations, from paresis to paralysis of the hind legs. Hard breathing was also frequenly observed. Nude mice with metastatic tumour growth in the kidney or stomach demonstrated a progressive loss of weight leading to cachexy. Simultaneous tumour development in the brain and in any of the mentioned
Table 1. Metastatic capacity of different BL cell-lines after intravenous injection into nude mice. Cell lines
Tumorigenicity*
No of No of experiments injected animals
Number of mice with tumours/Number of injected mice Brain
Spinal cord
Affected organs Bone Kidney marrow
Stomach
02 29 30 60
High
2 3 2 2
10 14 10 9
8/10 4/10 4/10 7/9
3/6 3/8 0/8 6/8
3/10 4/9 0/8 4/9
1/10 5/12 1/10 0/9
3/10 4/9 0/10 017
BL 57 BL 72
Low
3 2
12 7
0/12 017
0/12 017
0/12 017
0/12 117
0/12 017
BL 16 BL 37
No
3 2
13 8
0/13 0/8
0/13 0/8
0/13 0/8
0/13 0/8
0/13 0/8
IARC 139 IARC 728
No
2 2
8 8
0/8 0/8
0/8 0/8
0/8 0/8
0/8 0/8
0/8 0/8
BL BL BL BL
*
Tumorigenicity was assessed in earlier experiments on the basis of subcutaneous injection (Ref. 7).
high tumourigenic potential (BL 02, BL 25, BL 30 and BL 60) metastasized effectively in various organs, while those with low tumorigenicity (BL 57, and BL 72) being i. v. inoculated into 19 nude mice initiated metastatic tumour only in one (kidney). Non-tumourigenic BL lines (BL 16 and BL 37), as well as LCLs (lARC 139 and IARC 728) were unable to metastasize to any organ. For BL lines with a high tumourigenicity, the ,.overall frequency of metastatic growth in different organs was as follows: brain - in 23 of 39 (59%), spinal cord - in 12 of30 (40%), bone marrow376
Exp Toxic Pathol44 (1992) 7
visceral organs (the kidney, stomach) was accompanied by a combined symptomatology. Metastatic growth in the brain in some cases was confined only to meninges while in other mice it spread to brain tissue. Not infrequently, both meningeal infiltration and numerous perivascular tumour deposits were present (fig. 1). In few cases accumulation of tumour cells replaced an important part of the hemisphere and they were seen with the naked eye. In the stomach, both parts, forestomach and glandular stomach or either of them, could be affected. Initial stages of
Fig. 1. BL Cell infiltration in the brain. I a - meninges are diffusely infiltrated with BL cells; small perivascular infiltrates in the brain tissue. H. & E. x80. 1b - large BL cell infiltrates in the brain tissue. H. & E. x 140. tissue with a subsequent destruction of the epithelium and the tumour cell spread to the muscular layer and serosa . Macroscopically, the diffuse or local thickening of the stomach wall was observed leading in some cases to the transformation of the stomach into huge tumourous mass. The lymphomatous infiltration in the kidneys was identified either microscopically only or had a form of a huge tumour replacing the kidney almost completely. Lymphoma cells had a big nucleus of a round, oval or irregular shape with a fine chromatin and easily discernable nucleolus (fig. 3). In some areas, mitoses were frequent, in others many nuclei underwent pycnotic changes . "starry-sky" pattern was not observed.
Discussion
Fig. 2. Diffuse BL cell infiltration under the squamous epithelium of the forestomach . H. & E. x500. metastasis were manIfested in the focal infiltration by tumorcells of the mucous membrane (fig. 2). This progressed to the total replacement of the mucous membrane by tumour 12
The majority of BL cell lines like those of both primary and metastatic human tumours of other origin grow successfully as a subcutaneous graft in nude mice, but manifest limited if no metastatic protential (3, 6, 19). In our experiments, we demonstrated that i . v. inoculation of BL cells gave rise to formation of metastases in various organs, namely in the brain, spinal cord, bone marrow, stomach, and kidney. The frequency and the spectrum of the affected organs did not depend on the patient's origin (Caucasian, Reunion or North Africa), the origin of biopsied tumour specimen used for establishment of BL cell lines (lymph nodes, abdomen, bone marrow, maxillar and so on), the presence of EBV or type of cytogenetic translocation (8: 14 or 8:22) (table 2). On the other hand it was found that the metastatic capacity of BL cell lines in nude mice strictly correlated with their tumourigenicity upon s.c. inoculation (7) . Only tumourigenic BL cell lines possessed the capacity to form experimental metastases; non tumourigenic BL lines did not colonize any organ (BL 16, BL 37) and behaved like LCLs (IARC 139, Exp Toxic Pathol 44 (1992) 7
377
Fig. 3. BL cell infiltration of the kidney. 3 a - large infiltrate in the kidney cortex. H. & E. x200. 3 b - to demonstrate the fine structure of BL cells. H. & E. X 1200.
Table 2. The main characterics of BL cell lines used for intravenous injection into nude mice. Nb 1 2 3 4 5 6 7 8 9 10
Cell lines
Patient/Origin
Specimen Origin
BL 02 BL 16 BL 29 BL 30 BL 37 BL 57 BL 60 BL 72 IARCIl39 IARCn28
Caucasian ReuniQP Reunion Caucasian Caucasian Caucasian Nth African Nth African LCL (BL 30) LCL (Lymphoma)
Biopsy LN (Dx) Abd. Asc. (Dx) Abd. Biopsy (Dx) Bone marrow (R) Bone marrow (Dx) Periph. Blood (+ B 95) Maxill. Punct. (Dx) Abd. Biopsy (Dx) Blood (+ B 95) Blood (+ B 95)
Presence Cytogenetic EBV translocation
+ + + + + + +
t(8; t(8; t(8; t(8; t(8; t(8; t(8; t(8;
22) 14) 14) 14) 22) 14) 22) 14)
LN - Lymph node Dx - specimen was taken at the moment of diagnosis R - specimen was taken at the state of relapse + B 95 - peripheral blood lymphocytes were infected with B 95 EBV strain
IARC 728). It was recently reported, however, that human EBV-positive LCL, tenned ZS, after irradiation in vitro aquired the capacity to invade the lymph nodes of nude mice (17). Such unusual feature of mentioned LCL can be explained by the effect of irradiation, which was likely the cause of chromosomal rearrangements in ZS karyotype and may have induced mutation of a number of relevant oncogenes. The significance of c-myc deregulation for an unlimited in vivo growth potential of B lymphoblastoid cells is a subject of many investigations (8, 11). In particular, it was recently demonstrated that in vivo down-regulation of the c-myc gene does not contribute to the-BVLCL hybrid graft regression in nude mice. The level of the c-myc transcription pattern detected in the parental BL cell lines was the same in both 378
Exp Toxic Pathol 44 (1992) 7
growing and regressing hybrid grafts whose in vivo growth phenotype was indistinguishable from that of the parental LCL (20). Our data are in line with the findings of GHETIE et a1. (5) who recently demonstrated that following i.v. inoculation the human BL cell line Daudi could be grown in scm mice as dissiminated tumours affecting the lungs, kidneys, ovaries and adipose tissue. Microscopic tumour infiltrates were also found in the spleen, bone marrow, spinal column and femur. In contrast to other findings in the current study tumour cells disseminating into the lung, liver, ovaries and spleen have never been observed. As distinct from human BL no "starry-sky" pattern was observed histologically. Likewise "starry-sky" was only rarely found after s.c. inoculation of the same BL lines (7). Our data are also
,-,
L'-"J"
.................
"" ........... _ .... _ ..... "" ....
_-... ..
r ...
~r-~
... J
-~·r
...
-~
...... .... J
~
...
~-
...
4. GALLATIN M, ST JOHN TP, SIEGELMAN M, et al.: Lymphocyte homing receptors. Cell 1986; 44: 673-680. 5. GHETIE M-A, RICHARDSON J, TUCKER T, et al.: Dissiminated or localized growth of a human B-cell tumour (DAUDI) in scm mice. Int J Cancer 1990; 45: 481-485. 6. GIOVANELLA BC, FOGH 1: The nude mouse in cancer research. Advanc Cancer Res 1985; 44: 69-120. 7. GURTSEVITCH VE, O'CONOR GT, LENOIR GM: BURKITT'S lymphoma cell lines reveal different degrees of tumorigenicity in nude mice. Int J Cancer 1988; 41: 87-95. 8. HOTCHIN NA, ALLDAY MJ, CRAWFORD DH: Deregulated c-myc expression in EpSTEIN-BARR virus immortalized Bcells induces altered growth properties and surface phenotype but not tumorigenecity. Int J Cancer 1990; 46: 566-571. 9. LEMERLE J: Rapporteur's Report. In: BURKITT'S Lymphoma: a Human Model. (eds. GM LENOIR, GT O'CONOR, CLM OLWENY), pp. 149-151. IARC Scientific Publications No. 60, Lyon 1985. 10. LENOIR GM: Role of the virus, chromosomal translocations and cellular oncogenes in the aetiology of BURKITT'S lymphoma. In: The EpSTEIN-BARR Virus: Recent Advances. (eds MA EpSTEIN, BG ACHONG), pp. 183-200. William Heinemann Medical Books 1986. 11. LOMBARDI L, NEWCOMB EW, DALLA-FAVERA R: Pathogenesis of BURKITT lymphoma: expression of an activated c-myc oncogene causes the tumorigenic conversion of EBV-infected human B-Iymphoblasts. Cell 1987; 49: 161-170. 12. MAGRATH IT, SARIBAN E: Clinical feature of BURKITT'S lymphoma in the USA. In: BURKITT'S Lymphoma: a Human Model. (eds. GM LENOIR, GT O'CONOR, CLM OLWENY), pp. 119-127. IARC Scientific Publications No. 60, Lyon 1985. 13. NICOLSON GL, DULSKI K, BASSON C, et al.: Preferential organ attachment and invasion in-vitro by B 16 melanoma cells selected for differing metastatic colonization and invasive properties. Invas Metast 1985; 5: 144-158. 14. NICOLSON GL, DULSKI KM: Organ specificity of metastatic tumor colonization is related to organ-selective groth properties of malignant cells. Int J Cancer 1986; 38: 289-294. 15. NKRUMAH FK, OLWENY CLM: Clinical features of BURKITT'S lymphoma: The African Experience. In BURKITT'S Acknowledgement: The authors are indebted Mrs. M. Lymphoma: a Human Model. (eds. GM LENOIR, GT LAVAL and Mrs. N. LYANDRAT for technical help and Mr. O'CONOR, CLM OLWENY), pp. 87-95. IARC Scientific 1. MaLON for the preparation of photographs. Publications No. 60, Lyon 1985. 16. PHILIP T: BURKITT'S lymphoma in Europe. In: BURKITT'S References lymphoma: a Human Model. (eds. GM LENOIR, GT O'CONOR, CLM OLWENY), pp. 107 -118. IARC Scien1. ANAISSIS E, GEHA S, ALLAM C, et al.: BURKITT'S Lymtific Publications No. 60, Lyon 1985. phoma in the Middle East. In: BURKITT'S Lymphoma: a 17. RONCELLA S, RAMARLI D, COOPER MK, et al.: EstablishHuman model. (eds. GM LENOIR, GT O'CONOR, CLM' ment of an EBV -positive lymphoblastoid cell line that OLWENY) pp. 129-134. IARC Scientific Publications grows as a lymphoma in nude mice and expresses memNo. 60, Lyon 1985. brane CD2 molecules. Int J Cancer 1990; 45: 299-307. 2. BERNSTEIN SC, WEINBERG RA: Expression of the metas18. THORGEIRSSON UP, TURPEENIEMI-HIJANEN T, WILLIAMS tatic phenotype in cells transfected with human DNA. JE, et al.: NIH/3T3 cells transfected with human tumor Proc Natl Acad Sci USA 1985; 82: 1726-1730.
experiments only tumourigenic cell lines possesed the metastatic capacity. The mechanism by which BL and other tumor cells preferentially form metastases in certain organs, is still unknown. From experiments with mouse 816 melanoma cell sublines, one can conclude that organ specificity of metastatic tumor colonization is determined by differences in the ability of the tested cells to adhere to and invade particular target organs (14) and/or by organ-selective growth properties of malignant cells (13). On the other hand, it should not be excluded that the mechanism of migration of circulating cells from the blood stream to a certain organ could be similar to that of specific recognition and adherence of mature lymphocytes, possessing so-called homing receptors, to the specialized high endothelial cells of the postcapillar venules (HEV) located in some lymphoid organs (see for review 4). Recognition of such receptors on the surface of metastazing BL cells would confirm such a suggestion. When comparing the pattern of metastasis in our experiments to that of human BL both the common features and dissimilarities can be noted. The spectrum of organs affected in nude mice reminds, on the whole, of that of human BL with exception of the lungs, liver and ovaries, which are frequently involved in human BL (9) and were never affected in nude mice. The jaw which is a typical site of tumour manifestation in African BL (21, 15) was never a target organ in mice. But one has to remember that in areas other than Africa the jaw is also only rarely, - if ever involved in BL (1, 16, 12). Different BL cell lines showed dissimilar organ distribution of metastasis as well. Thus, the model we have developed might represent an effective and reptoducible metastatic model system to study drug targeting by monoclonal antibodies since testing of monoclonal antibodies coupled to toxins has usually been performed either in in vitro systems, or in vivo on mouse leukemia. Our model can be also useful for the study of Burkitt's and other human lymphoma's pathogenesis as well as for analysis of homing properties of lymphomatous cells.
12*
Exp Toxic Pathol 44 (1992) 7
379
DNA containing activated ras oncogenes express themetastatic phenotype in nude mice . Molec Cell Bioi 1985; 5: 259-262. 19. WANG WR, SORDAT B, PIGUET D, et al.: Human colon tumors in nude mice: implatation site and expression of invasive phenotype. In: Immune-deficient Animals. (ed. B SORDAT), pp. 239-245. Karger, Basel, 1984.
20. WOLF J, KLEVENTZ B, PAVLITA M, et al.: Regressing nude mose grafts of BURKITTS'S lymphoma x Iymphoblastoid cell hybrids show deregulation of the c-myc gene and expression of the EBV latent membrane protein. Int J Cancer 1991; 47: 99-104. 21. ZIEGLER JL, BLiMING AZ, FASS L, et al.: Relapse patterns in Burkitt's lymphoma. Cancer Res 1972; 32: 1267-1272.
Exp Toxic Pathol 1992; 44: 380 Gustav Fischer Verlag lena
Announcement
The Ernest Witebsky Center for Immunology The following will be offered this Spring by The Ernest Witebsky Center for Immunology, University at Buffalo, State University of New York:
Recombinant DNA technology: methods and applications Hands-on laboratory course Monday, April 19, through Sunday, April 25, 1993 This seven-day course, presented in conjunction with ExonIntron, Inc., provides both fundamental and advanced training in recombinant DNA technology . It is designed to provide and expand laboratory skills in molecular cloning techniques for individuals with a background in biological sciences but who have little or no knowledge of recombinant DNA techniques. Course registration fee of $ 1,200.00 covers course materials, beverage breaks, and seven (7) days' luncheons. Registration will be limited to 24 persons.
Methods of immunologic research and diagnosis An at-the-bench laboratory training program
Monday, May 10, through Thursday, May 20, 1993 This ten-day program consists of daily practical laboratory exercises personally carried out by participants, accompanied by lectures, demonstrations and discussions. Exercises range
380
Exp Toxic Pathol 44 (1992) 7
from the basic techniques of antiserum preparation to advanced techniques of immunoassay. Open to individuals at the postdoctoral level, supervisory senior technologists and graduate students. Course registration of $ 700.00 includes course materials and beverage breaks. Registration will be limited to 36 individuals.
Latex particle immunoassays Hands-on laboratory course
Friday and Saturday, May 21-22,1993 This course is designed to provide an overview of preparation, formulation, properties and applications of latex particles in medical laboratory diagnosis . Geared toward laboratory researchers and technicians, and in-vitro diagnostic manufacturers. Course registration fee of $ 295.00 covers course materials, beverage breaks, and lunch on both days. Registration will be limited to 24 persons.
For further information on any of the above courses, contact Roger K. Cunningham, Ph.D., Director, The Ernest Witebsky Center for Immunology, 423 Sherman Hall, University at Buffalo, 3435 Main Street, Buffalo, New York 14214; Phone: 716/ 829-2901, Fax: 716/829-2158. Because class size is limited in all courses, we recommend registrations be submitted at least one month prior to respective starting dates for best possibility of acceptance.
