from a T-ALL sample engrafted in SCID mice Establishment of a ...

1993 81: 2714-2722

A Cesano, R O'Connor, PC Nowell, B Lange, SC Clark and D Santoli from a T-ALL sample engrafted in SCID miceEstablishment of a karyotypically normal cytotoxic leukemic T-cell line

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

reserved.Copyright 2011 by The American Society of Hematology; all rights900, Washington DC 20036.weekly by the American Society of Hematology, 2021 L St, NW, Suite Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published

use only.For personal at PENN STATE UNIVERSITY on February 23, 2013. bloodjournal.hematologylibrary.orgFrom

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprints

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtml

http://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtl

http://bloodjournal.hematologylibrary.org/


Establishment of a Karyotypically Normal Cytotoxic Leukemic T-ce l l Line From a T-ALL Sample Engrafted in SCID Mice

By Alessandra Cesano, Rosemary O’Connor, Peter C. Nowell, Beverly Lange, Steven C. Clark, and Daniela Santoli

Bone marrow (BM) cells from a child with an immature (CD3-) acute T lymphoblastic leukemia (T-ALL) bearing no chromosomal abnormalities failed to grow in long-term culture in the presence or absence of recombinant human (rh) growth factors but could be engrafted in severe combined immunodeficient (SCID) mice and induced leukemia. The leukemic cells recovered from the animal tissues could be adapted to grow in vitro in the presence of rh interleukin- 2 (IL-2) and give rise to a growth factor-dependent cell line designated TALL-1 07. This cell line expresses T-cell-specific mature markers (CD2, CDB/T-cell receptor [TCR] d?, CD8, CD56), and its growth can be inhibited by IL-4 of all the cytokines tested. Similar to the original leukemic blasts, TALL-1 07 cells are clonal, have rearranged TCR-8, y, and 6 loci, and a normal 46 XY karyotype. However, unlike the

EVERAL LABORATORIES have reported the estab- S lishment of cell lines from patients with acute T lymphoblastic leukemia (T-ALL) either directly or on mitogen and interleukin-2 (IL-2) In a recent report: we have indicated that the ability of T-ALL cells to become established and grow permanently in culture is positively correlated with the presence of chromosomal abnormalities. Specifically, we generated four T-ALL cell lines carrying the t(8;14) (q24;qll) chromosomal translocation and one with the t( 11;14) (p13;qll) translocation.6 These translocations involving chromosome 14 result in the association of a pu- tative proto-oncogene (c-myc in the case of chromosome 8 and TCL-2 in the case ofchromosome 1 I ) with T-cell receptor (TCR) coding regions. Oncogene involvement is thought to alter drastically either the pattern of growth or the differen- tiation pathway of leukemic cells leading to uncontrolled growth.’,’ All attempts by us and other investigators to establish in culture T-ALL cells with normal karyotypic features have so far failed, indicating that in vitro culture conditions are not adequate for the expansion of leukemic cells that bear no obvious translocations and oncogene involvement.

We have recently observed that the severe combined immunodeficient (SCID) mouse strain is able to act as an in vivo tissue culture vessel allowing the propagation of lym-

From The Wistar Institute ofAnntomy and Biology, the Children’s Hospital of Philadelphia, the University of Pennsylvania School of Medicine, Philadelphia, PA; and the Genetics Institute, Cambridge, MA.

Submitted October 22, 1992; accepted December 30, 1992. Supported by Research Grants No. CA-47589, CA-10815, and CA-

42232from the National Institutes of Health and Grant No. CH-527 from the American Cancer Society.

Address reprint requests to Daniela Santoli, Ph.D, The Wistar In- stitute of Anatomy and Biology, 36th and Spruce Sts. Philadelphia, PA 19104.

The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section I734 solely to indicate this fact.

0 1993 by The American Society of Hematology. 0006-4971/93/811 0-00l4$3.00/0

patient‘s BM cells, the TALL-1 07 cell line displays potent tumoricidal activity that is not major histocompatibility complex restricted. The magnitude of mRNA expression of perforin and serine esterases and of lytic activity depends on the doses of IL-2 added. TALL-107 cells can also be triggered by CD3- and CD2-specific monoclonal antibodies (MoAbs) to mediate reverse tumor cell lysis. In addition, this cell line produces high levels of interferon gamma and tumor necrosis factor alpha on stimulation with anti-CD3 and/or anti-CD2 MoAb both in the presence or absence of IL-2. The overall data indicate that the SCID mouse is able to support the functional maturation and expansion of a cytotoxic T-cell subset from some types of T-ALL. 0 1993 by The American Society of Hematology.

phoid leukemic cells (both of pre-B-cell and T-cell lineage) from samples that fail to adapt to in vitro culture conditions even in the presence of recombinant hematopoietic growth factor^.^^'^ In general, however, on recovery from SCID mouse tissues, the leukemic cells are unable to grow in vitro in the presence and absence of cytokine~.~.’~ These findings indicate that the SCID mouse provides a more suitable environment for the expansion of human ALL cells than in vitro culture conditions, which is likely through the release of unknown growth factors by stromal cells.

In this study, the bone marrow (BM) cells from an immature (CD3- CD2’ CDY) pediatric T-ALL sample that could not be maintained in long-term culture in vitro were able to infiltrate SCID mouse tissues inducing leukemia reminiscent of the patient clinical picture. We have examined the growth factor requirements of the leukemic cells recovered from the mouse pathologic tissues and were able to establish an IL-2-dependent cell line (TALL- 107) that displays mature T-cell characteristics. Overall, the results indicate that engraftment of T-ALL cells in SCID mice allows the selective expansion and maturation of cytotoxic clones within the leukemic T-cell population and suggest that there is a particular T-cell subset in pediatric leukemia that can acquire killer function on appropriate activation. The TALL-I07 cell line established and characterized herein adds to the list of permanent cytotoxic cell lines that we have established from T- ALL cases.6,11,‘2

MATERIALS AND METHODS

The patient’s BM sample was obtained from the Chil- dren’s Hospital of Philadelphia, PA. Use of excess diagnostic material for research was approved by the Institutional Review Board and by The Wistar Institute in accord with an assurance filed with and approved by the US Department of Health and Human Services. The patient was an 8-year-old boy who presented with high-risk T-cell ALL (LI morphology). Clinically, he had marked anterior cervical lymphoadenopathy, an anterior mediastinal mass, and hepatosplenomegaly. The white blood cells at diagnosis were 149,OOO/pL with 77% blasts, hematocrit (Ht) was 29, and platelets were 93,000/ pL. The spinal fluid was negative for leukemic cells. Karyotypic analysis showed no abnormalities in chromosome number or morphology. After treatment with a protocol for high-risk ALL and prophylactic

Patient.

2714 Blood, Vol 81, No 10(May 15). 1993: pp 2714-2722




A KARYOTYPICALLY NORMAL KILLER T-ALL CLONE 2715

central nervous system irradiation, the patient went into complete remission that still continues 2 years after diagnosis.

Pure recombinant human (rh) IL-2 (specific activity IO’ U/mg protein) was a gift from Dr Maurice Gately (Hoff- mann-LaRoche, Nutley, NJ). Stem cell factor (SCF) half maximally active at 25 ng/mL in a proliferation assay using the TF-1 cell line was purchased from Amgen (Thousand Oaks, CA). The following hematopoietic growth factors were supplied by the Genetics Institute Pilot Development Laboratory (Cambridge, MA): homogeneous rhIL- 3 (CHO cell-derived, specific activity 4.3 X lo6 U/mg protein), IL- la (COS cell-conditioned medium [CM], half maximally active at 9 X IO-’ in a murine thymocyte proliferation assay), IL-4 (COS cell CM, half maximally active at 9 X lo-’ in a proliferation assay using M07e cells), IL-5 (COS cell CM, maximally active at 9 X lo-’ in a CML proliferation assay), IL-6 (Escherichia coli-derived, specific activity 2 X IO6 U/mg), and IL-7 (COS cell CM, half maximally active at 3 X IO-’ in a thymocyte assay).

The patient’s BM sample was subjected to Ficoll-Hypaque gradient centrifugation. The leukemic cells were partly frozen in liquid nitrogen and partly seeded in a 24-well Linbro plate (Flow Laboratory, McLean, VA) at the concentration of 1 X 106/ml in Iscove’s modified Dulbecco’s medium (GIBCO, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS; Hy- clone, Logan, UT) and antibiotics (complete medium). IL-2 (100 U/ mL), IL-7 (5 X IO-’), or IL-3 (10 UjmL) were added to distinct wells. Control wells contained no growth factors. Medium and growth factors were replaced biweekly.

CB 17-scid/scid, originally obtained from Dr Me1 Bosma (Fox Chase Cancer Center, Philadelphia, PA) were bred and maintained in a pathogen-free environment in The Wistar Institute Animal Facility. Cryopreserved leukemic cells were thawed, suspended in phosphate-buffered saline, and inoculated intraperitoneally (IP) at 2 X IO’ (in 0.5 mL volume) in two 10-week-old IgM nonproducer mice, which had not been pretreated with immunosuppressants nor irradiated.

Five months after leukemic cell transfer, the mice were killed despite the absence of clinical symptoms. Peripheral blood (PB) and BM were collected. The enlarged spleen was minced through metal grids followed by a Ficoll-Hypaque gradient to remove dead cells, erythrocytes, and de- bris. Cells recovered from the spleen were cultured in complete medium with or without IL-2 (100 U/mL) and expanded for phenotypic, genotypic, karyotypic, and functional analyses. Part of the cells was transferred in a second group of SCID mice, which had been injected IP with cyclophosphamide (CTX 100 mg/kg) 4 days earlier.

Mouse PB smears and BM samples concentrated on slides by cytocentrifugation were stained with May-Grunwald- Giemsa. Histopathologic analysis was performed on spleen, liver, kidneys, lung-heart-thymus complex, and brain, which had been fixed in 10% formalin and paraffin embedded; 4-pm sections were cut and stained with hematoxylin-eosin.

Short-term proliferation of the TALL- 107 cell line in response to rh growth factors was measured as described.6 Briefly, cells were seeded in complete medium at 5 X 104/well in 96- well microtiter plates (Falcon, Becton Dickinson, Oxnard, CA) in the presence and absence of the indicated concentrations of growth factors. After 3 days, the cells were pulsed with I rCi [’Hlthymidine (2 Ci/mmol) for 16 hours and harvested on fiber glass filters by an automated cell harvester. Isotope incorporation was measured with a Packard Matrix 96 direct beta counter (Packard Instrument B.V., Vigersmaneg, The Netherlands).

Karyotypic analysis was performed by standard techniques” on the diagnostic sample and on leukemic cells recovered from the SCID mouse tissues. A minimum of 20 meta- phases was examined for each sample.

Growth factors.

Leukemic cell culture.

Mice.

Recovery of the patient cells from mouse tissues.

Histopathology.

Proliferation assay.

Cytogenetic analysis.

Phenotypic characterization. Expression of cell surface antigens was measured by indirect immunofluorescence assay using an Ortho cytofluorograph cell sorter, as described! Fluorescein isothiocyanate- conjugated F(ab’)* fragments of goat antibodies to mouse Ig (Cappel Laboratories, Cochranville, PA) were used to detect binding. A fluorescence-intensity threshold was established where 99% of the control cells were negative, and the mean fluorescence intensity was determined in a range with an upper limit of 200. The following monoclonal antibodies (MoAbs) were used OKT3 (CD3), OKT4 (CD4), OKT8 (CD8) (Ortho Pharmaceuticals, Raritan, NJ), TCR- I (TCR- cup), anti-Leu- I9 (CD56) (Becton Dickinson, Mountain View, CA), 6TCS-1 (TCR-76) (T-cell Science, Inc, Cambridge, MA), B67.1 (CD2), B36.1 (CD5) (gifts from Dr Bice Perussia, Thomas Jefferson Uni- versity, Philadelphia, PA), and 3A1 (CD7) (American Type Culture Collection [ATCC], Rockville, MD).

Genomic DNA was isolated from the cells using a “DNA single-tube’’ method (Biotecx Labora- tories Inc, Houston, TX). After restriction enzyme digestion and electrophoresis in 1% agarose gels, the DNA was blotted to nylon membranes (Zetabind, AMF, Cuno, Meridien, CT) and covalently bound by UV irradiati011.l~ Total RNA was extracted from cells using the RNAzol method (Cinna/Biotex Laboratories Inc, Friendswood, TX). RNA samples were fractionated in a 1% agarose-formaldehyde gel, transferred to nylon membrane by capillary action, and covalently bound by UV irradiati~n.’~ Prehybridization, hybridization, and washing procedures were done as previously described.15 Probes were labeled by the random priming method (Amersham, Arlington, IL), and the blots were exposed to Kodak XAR film (Eastman Kodak, Rochester, NY) at -70°C for I to 6 days.

The HP-10 (pore-forming protein [PFP], perforin)16 and HSE26.1 (serine esterase [SE]2, granzyme B)” probes were obtained from Dr John Ding-E Young (Laboratory ofCellular Physiology and Immunology, New York, NY). The HuHF probe for SEI (granzyme A)’* was a gift from Dr Irvin Weissman (Stanford University School of Medicine, Stanford, CA). The TCR-0 chain constant region probe was the HincII CY fragment of the human cDNA clone YTJ-2I9 provided by Dr T. W. Mak (University of Toronto, Ontario, Canada). The phTa-92 genomic probe mapping to the 62 region of the TCR- 6 chainz0 was a gift of Dr Louise Showe (The Wistar Institute, Phil- adelphia, PA). The pH60 probe” a 700-bp EcoRI Hind111 fragment

Southern and Northern blot analysis.

Probes.

Table 1. lmmunophenotype of Primary (Uncultured) BM Cells, of Cells Recovered From SClD Mouse Tissues, and

of the Established TALL-107 Cell Line

MoAb

OKT3 OKT4 OKT8 867.1 TCR-1 BTCS-1 3A 1 Anti-Leu- 19 836.1

CD Group or Specificity

CD3 CD4 CD8 CD2 TCR-mP TCR-76 CD7 CD56 CD5

Cells From

Diagnostic EM

~ ~

Established SClD Mouse TALL-107

Tissues’ Cell Line

30 6 1

2 98 30 < I 39

2 97

44 92 7 3

61 90 100 99 30 81 < I <1 98 96 54 93

NT NT

Values are the percent reactive cells as detected by immunofluores-

Abbreviation: NT, not tested. * Similar results were obtained with leukemic cells recovered from the

cence analysis.

spleen or marrow.




CESANO ET AL 2716

Patient H L60 PBL BM TALL-1 07

Fig 1 . Southern blot analysis of the TCR-B locus in BM patient cells (diagnostic sample) and in the established TALL-1 07 cell line. HL60 cells and normal PB lymphocytes were used as a control germline and as a control for clonality, respectively. Genomic DNA was digested with EcoRl (A), BamHl (B), and Hindlll (C) and hybridized with a CB DNA probe. The pattern of abnormal bands indicates the presence

A B C A B C A B C' A B C ofrearrangementstothe~B~~ocus.

that contains the Jy1.3 gene segment and cohybridizes to the J72.3 region was purchased from ATCC.

The cytotoxic activity of primary patient cells both uncultured and stimulated with IL-2 (100 U/mL for 3 days), and of IL-2-maintained TALL-IO7 cells was measured in a 4- hour "Cr release assay, as previously described!."." The natural killer (NK)-susceptible K562 (erythroleukemia) and U937 (monocytic leukemia) cell lines were used to measure NK activity. whereas the NK-resistant cell lines HL60 (promyelocytic leukemia), Raji (Burkitt's lymphoma), ALL-I, ALL-2. ALL-3 (pre-B ALLs)? TALL-106 (T- ALL)." and fresh leukemic samples (case no. 125. a monocytic leukemia, and case no. 22. a T-ALL) were used lo measure lymphokine- activated killing. In experiments of reverse antibody-dependent cell- mediated cytotoxicity (ADCC), MoAb specific for CD2 (B67.1). CD3 (OKT3). CD8 (OKT8). and CD56 (anti-Leu-19) were added to the effector cells immediately before the addition of the IgG Fc receptor- positive IgG-FcR+ murine mastocytoma P815 (H-2 d DBA/2) target.

A fixed number of "Cr-labeled target cells ( 104/well) was tested against four effector cell concentrations. Specific "Cr release was cal- culated from the mean of three replicates and the results were expressed as 40010 lytic units (LU)/IO* effectors as de~cribed!.~." N- henz!huj,carhon!.l- L-/wine /li,w/~enz,rl esrer (BLT) iwi~rase re-

lease assq: The ability of MoAb specific for CD2, CD3. CD8. and CD56 antigens to trigger the release ofcytotoxic granules from TALL- 107 cells was tested as described." Briefly, I pg/mL MoAb in NaCO,/ NaHC02 buffer was immobilized on 96-well microtiter plates by overnight incubation at 0°C. After washing with RPMl 1640 medium supplemented with 5010 FBS and I O pmol/L HEPES buffer. TALL- 107 cells (5 X I05/well) were added and incubated for 4 hours at

C~~/o/o.~icI/,17 arsayr.

37OC cell-free supematants were then collected. BLT esterase activity was measured using a modification of the method described by Tak- ayama et al." Total BLT esterase activity contained in the cells was measured in cell lysates obtained by repeated freezing and thawing.

OKT3 and B67. I MoAb ( I pg/

mL) were immobilized on 96-well plates by dilution in carbonate buffer. After overnight incubation at 0°C. the plates were washed with medium supplemented with 5 8 FBS. TALL-I07 cells were then plated (2 X I05/wcll) and incubated at 37°C in the presence and absence of IL-2 ( 100 U/mL). Cell-free supernatants were harvested 18 hours later and tested for the presence of interferon-y (IFN-y) and tumor necrosis factor-n (TNF-n) by radioimmunoassay. as described previou~ly.".'~

Inrlrrc/ion c?/'lwipliokinc relcusc

RESULTS

Cliaructeristics qf thc dicignmtic T-ALL sumpkc. The diagnostic sample contained immature T cells as judged by the high expression of CD2 and CD5 antigens and the small number of cells expressing the CD3/TCR-a@ complex (Table 1). Neither B-cell (CDIO, CD19. CD20) nor myeloid (CD13. CD33, CD34) markers could be detected (data not shown). The leukemic blasts displayed a normal 46 XY karyotype. Rearrangement of the TCR genes were determined by Southern blotting with probes for the p, 6. and y loci. The leukemic cells showed band rearrangements to the Cp2 locus as determined by the BurnHl and ilindlll restriction patterns (Fig 1). Using the pH60 probe, which hybridizes to Jy1.3 and Jy2.3, rearranged bands also were detected on BumHI

Fig 2. Histopathologic analysis of tissues from SClD mice injected with the T-ALL patient cells. (A) P8 smear of a mouse showing a discrete number of circulating leukemic cells. (B) Highly infiltrated BM from the same mouse; murine hemopoietic elements are indicated by arrows. (C) Liver section of another mouse injected with T-ALL cells showing infiltration by lymphoblasts (indicated by arrows). (D) Lung section from the same mouse. N, peribronchial nodular tumor. (E) Brain (Br) section from the same mouse (arrows indicate the presence of leukemic cells in the area beneath the meninges). (F) Kidney section from the same mouse showing a marked leukemic infiltration of the parenchyma (IC, leukemic cells; t, tubule). Original magnification X 200 (A, C through F) and X 1,000 (B).




2718 CESANO ET AL

and HindIII restriction but not with EcoRI (not shown). Using the pHTa92 probe for the 562 region, rearranged bands were detected on HindIII restriction but not with EcoRI and BamHI (not shown). The pattern of rearrangements obtained with BamHI and HindIII, coupled with the lack of germline bands, suggested that the leukemic cells in examination rep- resented a clonal population. The cells lacked the ability to lyse NK-sensitive and NK-resistant targets both sponta- neously and on activation with rhIL-2 (100 U/mL) for 3 days (not shown). When cultured in the presence of rhIL-3 (10 U/mL), IL-7 (5 X or in the absence of growth factors, the leukemic blasts failed to proliferate and died within 1 week. In the presence of rhIL-2 (100 U/mL), the cells demonstrated a limited proliferative capacity and ceased dividing within 1 month.

Engraftment of cells of patient with T-ALL in SCID mice. Cells from the diagnostic BM sample were engrafted IP in two SCID mice, neither ofwhich showed clinical symptoms during a 5-month observation period. However, at autopsy, both mice presented a severe anatomic-pathologic picture characterized by marked hepatosplenomegaly (spleen 20 times heavier than control) and mediastinal lymphadenopathy. Neither ascites nor tumor masses were detected in the peritoneal cavity. Microscopic analysis (Fig 2) showed L1 blast infiltration in the PB (A), BM (B), liver (C), lungs (D), and kidneys (F). Despite the absence of neurologic symptoms, microscopic analysis of the brain showed the presence of leukemic cells underlying the meninges (Fig 2E). Cells recovered from the spleen of one mouse were transferred IP to a second group of mice pretreated with CTX to accel- erate the engraftment. These mice were selectively killed after 2.5 months and were found to display the same anatomic- pathologic findings as the first group of mice.

Cytogenetic and phenotypic analysis of T-ALL cells recov- eredfrom mouse tissues. Cytogenetic analysis on leukemic cells recovered from the spleens of the mice confirmed the presence of a normal 46 XY karyotype, as observed in the diagnostic sample. Phenotypically, the leukemic cells recovered from mouse tissues displayed a more mature phenotype as compared with the original cells, with a slight increase in the percent of CD3 positivity and the appearance of CD8 and CD56 antigens (Table 1).

Leukemic cells recovered from the BM and spleen of the SCID mice were cultured in the presence and absence of rhIL-2 (100 U/mL). Whereas the cultures without growth factor died within a few days, those maintained in IL-2 (both BM and spleen) were able to proliferate and become established as growth factor- dependent cell lines with a mature T-cell phenotype (close to 100% CD2+, CD3+, CD7+, CDV, CD56+; Table 1). The BM and spleen subcultures were combined together and the cell line, designated TALL- 107, has now been in continuous culture for 8 months.

Genotypic analysis. TCR-P-, y-, and 6-chain gene rearrangements were studied in the newly established cell line and compared whh the ones shown by the parental leukemic cells. The rearrangement patterns to the CP, (Fig I ) and Jy, (not shown) loci (as determined by the BamHI and HindIII restriction enzymes) and to the J6* locus (as determined by

Establishment of the TALL-I07 cell line.

rhlL-4: 0 0 0 1110,Ooo

112,500 A 111,250

115,oQo

0 I 5 10 50 100

rhlL-2 (U/ml)

Fig 3. Proliferative responses of the established TALL-107 cell line to rhlL-2 and IL-4. Cells were starved from IL-2 overnight and plated at 5 X 1 O"/well in complete medium in the presence and absence of IL-2, and/or IL-4. r3H]Thymidine was added 4 days later and isotope incorporation measured after 16 hours. Results shown are from one representative experiment of three performed with the same results at different times between 1 and 4 months after establishment of the cell line.

the HindIII restriction enzyme; not shown) were exactly the same in TALL-107 cells as in the parental leukemic cells confirming the derivation of the TALL-107 cell line from the original leukemic population.

Short-term prolijerative responses of TALL-I 07 cells to hematopoietic growth factors. The proliferative response of the established TALL-107 cell line to rhIL-la ( 1 X to 2.5 X lo-'), IL-2 (I to 100 U/mL), IL-3, IL-6 (both 1 to 20 U/mL), IL-4, IL-5 (both 5 X to 2.5 X IO-*), IL-7 ( I X

and SCF (10 to 500 ng/mL) were analyzed using short-term [3H]thymidine incorporation assays. TALL- 107 cells showed a dose-dependent response to rhIL-2 (Fig 3) and failed to respond to all the other growth factors (not shown). IL-4 inhibited the IL-2-dependent growth of TALL- 107 cells in a dose-dependent manner (Fig 3). Such an in- hibition by IL-4 cannot be attributed to possible toxic prod- ucts present in the COS cell CM because IL-la, IL5, and IL-7 (also derived from COS cells) were not effective. More- over, a rabbit antihuman IL-4 antibody (Genzyme, Boston, MA) at 10 pg/mL completely blocked the inhibitory effects of IL-4 (not shown).

The morphology of May-Griin- wald-Giemsa-stained TALL- 107 cells was examined under a light microscope. Figure 4 shows the typical appearance of TALL-107 cells with a nucleus to cytoplasm ratio of about 1 : 1 and a uniform basophilic cytoplasm containing few azurophilic granules typical of mature cytotoxic lymphocytes.

The ability of the established cell line to lyse NK-susceptible and NK-resistant targets was measured

to I X

Morphologic analysis.

Cytotoxic activity.





efficiency. The extent of killing capability of these cells cor- relates positively with IL-2 stimulation and PFP and SE expression: Cells maintained in IL-2 ( 100 U/mL) display basal levels of these messages (Fig 6). whereas deprivation of IL-2 results in a gradual loss of cytotoxicity, intracytoplasmic granules, and PFP and SE transcripts (not shown). After IL- 2 deprivation and restimulation with 5 to 100 U/mL IL-2 for 16 hours. the cells express again all three transcripts at levels corresponding to the doses of IL-2 added (Fig 6).

BLT esterase secretion by TALL- 107 cells in response to specific interaction with MoAbs was investigated. Table 2 summarizes the results of one representative experiment of three performed with the same results. Secretion of BLT esterase by TALL-IO7 cells was efficiently

BLT cslcrasc rclcmc.

lo00

In Fig 4. Morphologic features of TALL-1 07 cells in IL-2. Note the

high nucleus/cytoplasm ratio and the basophilic cytoplasm containing few azurophilic granules (original magnification X 1,000).

in a 4-hour “Cr release assay (Fig 5). TALL-IO7 cells maintained in IL-2 ( 100 U/mL) can lyse very efficiently both NK- sensitive (K562 and U937) and NK-resistant (Raji, HL60, ALL-I, ALL-2, ALL-3, T-ALL-106) tumor cells as well as fresh leukemic samples (case no. 123 [M5] and case no. 22 [T-ALL]) (Fig 5A). The possibility of triggering the lytic ma- chinery of these cells using MoAbs that recognize specific T- cell surface molecules was investigated in reverse ADCC assays against IgG FcR’ P8 15 cells by incubating the effectors in the presence of 100 U/mL IL-2 and MoAbs specific to CD2 (B67.1). CD3 (OKT3). CD8 (OKT8), or CD56 (anti- Leu-19). Although background lysis of PSI5 cells was observed also in the absence of MoAb stimulation, the highest levels of cytotoxicity were always induced on triggering with

producible levels of cytotoxicity. By contrast, stimulation of OKT3 (Fig 5B). Stimulation via CD2 induced low but re- lo00

no mAb OKr3 867.1 OKr3ll367.1

the effector cells with-MoAbs specific to CD8 and CD56 did not increase the basal level of lysis (not shown). In addition, B67. I (Fig 5B), OKT8, and anti-Leu-19 (not shown) MoAbs did not alter the cytotoxicity triggered by 0 ~ ~ 3 antibody.

Northern blot hybridization was performed to analyze mRNA expression Of PFP, SEI, and SE2 by TALL- 107 cells and to correlate with killing

Fig 5. cytotoxic of TALL-1o7 cells. (A) Direct lysis of NK-sensitive (K562 and U937). NK-resistant (HL60. Raji, ALL-1, ALL-2, ALL-3, TALL-1 06). and fresh leukemic (cases nos. 123 and 22) targets. (B) Reverse ADCC assay; soluble MoAbs were added to TALL-1 0 7 cells just before addition ofthe 6’Cr-labeled FcR+ P815 target. Both panels show the of one representative expe+ merit of three performed between 1 and 6 months after the establishment of the cell line with similar results.

EWrc,”,”ion (?f c?’~o!ll.~in mRNA.




2720

kb 2.9 -

1.1 -

rhlL-2( U/ml)

0 5 10 20 50100

0.9 - ;

28s - I

HP10 ( perfor in)

HSE26.1

Fig 6. Expression of mRNA for PFP, SE1, and SE2 by TALL-1 0 7 cells. Cells were starved overnight and then restimulated with 5 to 100 U/mL of IL-2 for 1 6 hours.

and reproducibly observed on triggering (both in the presence and absence of IL-2) with immobilized anti-CD3 MoAbs but not with antLCD2 MoAbs. Simultaneous triggering of CD3 and CD2 antigens did not increase the secretion of BLT esterase induced by OKT3.

As measured by radioimmunoassay, unstimulated TALL-IO7 cells produce little, if any. IFN-y and TNF-n (Fig 7). Immobilized OKT3 triggers the production of high levels of these cytokines (up to I20 U/mL) both in the presence and absence of IL-2. The anti- CD2 MoAb B67.1 is unable to induce IFN-y release but induces a slight (though reproducible) increase in the secretion of TNF-n in the presence of IL-2 (Fig 7). The two antibodies display additive effects on TNF-n release, but not IFN-y release both in the presence and absence of IL-2. Thus, the magnitude of TNF-a production by TALL-IO7 cells is re- markably higher on simultaneous triggering via the CD2 and CD3 signaling pathways. whereas the production of IFN-.I is not.

In several experiments. immobilized MoAbs specific for either CD8 or CD56 did not stimulate IFN-y nor TNF-a production in TALL-IO7 cells (not shown).

IFN-y and TNF-a prodiction.

DISCUSSION

The present study demonstrates that ( I ) the SClD mouse could support the growth of a karyotypically normal T-ALL sample that had failed to grow in culture conditions, and (2) it was possible to establish a permanent cell line after recovery of the leukemic cells from the mouse tissues. Failure of this T-ALL sample to adapt to long-term growth in culture is consistent with previous data indicating that only T-ALLs featuring chromosomal abnormalities can be established as permanent cell lines in vitro." Another T-ALL sample with the t( I 1;14) translocation could be propagated in vitro for only 3 weeks but after transfer in SCID mice and recovery

CESANO ET AL

Table 2. BLT Esterase Release in MoAb-Stimulated TALL-1 0 7 Cells

% BLT Release

MoAb No IL-2 IL-2 (1 00 U/mL)

None 24.0 26.0 OKT3 52.7 53.0 867.1 43.7 38.7 OKT3/867.1 49.2 48.8

from their pathologic tissues became established in vitro as a permanent growth factor-independent cell line (TALL-108) (unpublished data, January 1992).

At difference with all the T-ALL samples investigated so far that were found to induce fatal leukemia in SClD mice

140

120

100

h - E 80 2 ?

60

40

20

0

180

160

140

120 E 3 100 23

L i 80 z t-

60

40

20

0

-

h - Y

a nomAb O R 3 867.1 OKT3667.1

Fig 7. Production of IFN-y and TNF-u by TALL-107 cells on stimulation with MoAbs. TALL-1 0 7 cells were stimulated in the presence (0) or absence (m) of IL-2 with immobilized OKT3 and/or 867.1 MoAb. The release of IFN-y and TNF-u in the supematants was tested after 1 8 hours by radioimmunoassay.





within 7 to I O weeks: the BM cells of the patient with T- ALL in the present study displayed a slower growth in the animal tissues and the animals remained clinically asymp- tomatic for the whole 5-month observation period. However, the anatomic-pathologic findings at autopsy revealed a heavy leukemic picture reminiscent of the patient disease characterized by infiltration of the lymphoid organs (especially BM and PB), mediastinal lymphadenopathy, and hepatosplenomegaly. The slow growth of this sample in the SCID mouse tissues reflects the indolent behavior of the leukemic cells in the patient of origin; this patient, in fact, is still in remission 2 years after the onset of the illness.

As noted in two previous studies on the engraftment of hematopoietic malignancies in SCID m i ~ e , ~ , ’ ~ the possibility of expanding in vivo primary leukemic cell that cannot be supported in long-term in vitro culture represents a very useful approach for investigating the homing, progression, and ag- gressiveness of each leukemia (thus providing invaluable diagnostic and prognostic information) and allowing the development of appropriate treatment strategies. The observation that primary T-ALL cells can adapt to permanent in vitro growth after being passed in SCID mice represents a new finding that has important implications. In fact, because the recovered cells are genotypically and karyotypically iden- tical to the original malignant cell^,^*'^ the cell lines derived from the mouse tissues could be used for a number of different studies. The ability of the SCID mouse to support the growth of T-ALL cells that are either unresponsive to growth factors or lose responsiveness after a limited period of time in culture indicates that the SCID mouse is able to provide the nutrients or conditions necessary to stimulate and maintain the leukemic cells in a state of continuous activation. Consequently, after recovery from the mouse tissues, these cells either lose growth factor dependency (and no longer require the mouse microenvironment for continuous propagation in vitro) or need an exogenous cytokine that is able to replace the stimulus previously offered by the murine tissues. In the case of the T-ALL cells in this study, it appears that the SCID mouse allowed the phenotypical and functional maturation and the selective expansion of a cytotoxic T-cell clone originally present, though undetectable, in the leukemic population. This is supported by the finding that cells recovered from the mouse spleen or BM had acquired CD8 and CD56 antigens (which represent cytotoxicity-associated markers) and grad- ually acquired killer ability on incubation in IL-2 analogous to the situation with conventional lymphokine-activated killer cells. Although the IL-2 provided by the mouse environment could not have been responsible for the in vivo maturation and expansion of this cytotoxic subset (murine IL-2 does not stimulate human cells), other stimuli, possibly derived from stromal cells, might have selectively triggered the growth of cells expressing CD3, CD7, CD8, and CD56 antigens. With time in culture, these markers became expressed on almost 100% of the cells, and the TALL107-derived cell line became totally dependent on IL-2 for continuous growth and cytotoxic function. A recent report by Maziarz et alZ5 has shown that fresh leukemic blasts from a patient with a CD3/TCRy@ T-ALL were endowed with major histocompatibility complex (MHC)-nonrestricted NK-like cytotoxic activity against a

limited target cell panel. These authors suggested that this T- ALL potentially had in vivo functional cytolytic activity; but they were not able to determine whether this activity con- tributed to the patient’s disease. In our study, both the original patient blasts and the cells freshly recovered from SCID mouse tissues were agranular and totally devoid of cytotoxic activity; in contrast, after culture in IL-2, the cells acquired azurophilic intracytoplasmic granules, typical of cytotoxic cells, along with MHC-nonrestricted killer activity against a broad spectrum of NK-sensitive and NK-resistant tumor targets. In this respect, the TALL-107 cell line is comparable to T lymphokine-activated killer (T-LAK) cells (CD3’ CD8+ CD56’) that are generated by incubation of normal PB lymphocytes in IL-2 and to a leukemic T-cell line, TALL-104, which was derived from another pediatric case of T-ALL.6%12 These two cell lines (CD3/TCRaP+) share a number of characteristics, being very similar phenotypically, morphologi- cally, and functionally. As shown in the case of TALL-104 cells,’2 the extent of killing capability of TALL-107 cells cor- relates with cytolysin (PFP, SE) expression, and with the IL- 2 doses added. Deprivation of IL-2 results in a gradual loss of intracytoplasmic granules, cytolysins, and cytotoxic function. However, at difference with TALL- 104 cells, the karyotype of TALL-107 cells is normal, thus resembling more closely LAK cells from healthy donors. In addition, different from TALL-104 cells, but similar to the CD3/TCR-y6+ TALL- 103/2 cell line also established in this laboratory,6.’ I,’’

the IL-2-dependent growth of TALL-107 cells can be inhibited dose dependently by rhIL-4.

The demonstrated abilities of the TALL-1049224 and TALL- 107 cell lines to (1) produce high levels of cytokines in response to specific stimuli; (2) efficiently lyse a broad panel of tumor target cells; (3) grow briskly in vitro in response to IL-2; (4) infiltrate SCID mouse tissues; and (5) display tumoricidal activity in this in vivo model (A. Cesano et al, manuscript in preparation, 1992) offer a unique opportunity to develop improved protocols for adoptive transfer therapy to cancer in a preclinical setting.

ACKNOWLEDGMENT

We thank Maurice Gately (Hoffmann-LaRoche, Nutley, NJ) for the generous gift of recombinant human IL-2, Janet Finan for karyotypic analysis, Jeffrey S . Faust for assistance at the cell sorter, Mike Sidelski for care of the SCID mouse colony, and the Wistar Editorial Office for preparing the manuscript.

REFERENCES 1. Kess UR, Lukeis R, Ford J, Garson OM: Establishment and

characterization of a childhood T-cell acute lymphoblastic leukemia cell line, PER-255, with chromosome abnormalities involving 7q32- 34 in association with T-cell receptor+ gene rearrangement. Blood 74:369, 1989

2. Huang CC, Hou Y, Woods LK, Moore GE, Minowada J: Cy- togenetic study of human lymphoid T-cell lines derived from lym- phocytic leukemia. J Natl Cancer Inst 53655, 1974

3. Gjerset R, Yu A, Haas M: Establishment of continuous cultures of T-cell acute lymphoblastic leukemia cells at diagnosis. Cancer Res 50:10, 1990

4. Minowada J, Sagawa K, Trowbridge IA, Kung PD, Goldstein G Marker profiles of 55 human leukemia lymphoma cell lines, in




2722 CESANO ET AL

Rosenberg SA, Kaplan HS (eds): Malignant Lymphomas: Etiology, Immunology, Pathology, Treatment. San bego, CA, Academic Press, 1982, p 53

5. Kees UR, Ford J, Ranford PR, Dawson VM: Characterization of an interleukin-2 dependent human leukemic cell line, PER 3 15, with an immature T-cell phenotype which does not express the Tac antigen. Leukemia 4:292, 1990

6. OConnor R, Cesano A, Lange B, Finan J, Nowell PC, Clark SC, Raimondi S, Rovera G, Santoli D: Growth factor requirements of childhood acute-T-lymphoblastic leukemia: Correlation between presence of chromosomal abnormalities and ability to grow permanently in vitro, Blood 77:1534, 1991

7. Erikson J, Finger L, Sun L, Ar-Rushdi A, Nishikura K, Mi- nowada J, Finan J, Emanuel BS, Nowell PC, Croce CM: Deregulation of c-myc by translocation of the a locus of the T-cell receptor in leukemias. Science 232:884, 1986

8. Champagne E, Takihara U, Sagman U, de Sousa J, Burrow S, Lewis WH, Mak TW, Minden MD: The T-cell receptor delta chain locus is disrupted in the T-ALL associated t( I1;14) (p13-;qlI) translocation. Blood 73:1672, 1989

9. Cesano A, OConnor R, Lange B, Finan J, Rovera G, Santoli D: Homing and progression patterns of childhood acute lymphoblastic leukemias in severe combined immunodeficiency mice. Blood 77: 2463, 1991

IO. Cesano A, Hoxie JA, Lange B, Nowell P, Bishop J, Santoli D The severe combined immunodeficient (SCID) mouse as a model for human myeloid leukemias. Oncogene 7927, 1992

1 1. Santoli D, OConnor R, Cesano A, Phillips P, Colt TL, Lange B, Clark SC, Rovera G: Synergistic and antagonistic effects of IL-la and IL-4 respectively on the IL-2 dependent growth of a T cell receptor-y6+ human T cell line. J Immunol 144:4703, 1990

12. Cesano A, Santoli D: Two unique human leukemic T-cell lines endowed with a stable cytotoxic function and a different spectrum of target reactivity. Analysis and modulation of their lytic mechanisms. In Vitro Cell Dev Biol 28A:648, 1992

13. Seabright M: A rapid banding technique for human chro- mosomes. Lancet 2:97 I , 197 1

14. Church GM, Gilbert W: Genomic sequencing. Roc Natl Acad Sci USA 81:1991, 1984

15. Feinberg AP, Vouelstein B: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 137:266, 1984

16. Young JD, Liu CC, Persechini PM, Cohn AZ: Perforin-dependent and independent pathways of cytotoxicity mediated by lymphocytes. Immunol Rev, 103: I6 I , 1988

17. Trapani JA, Klein JL, White PC, DuPont B: Molecular cloning of an inducible serine esterase gene from human cytotoxic lymphocytes. Proc Natl Acad Sci USA 896926, 1988

18. Gershenfeld HK, Hershberger RJ, Shows TB, Weissman I L Cloning and chromosomal alignment of a human cDNA encoding a T-cell and natural killer cell-specific trypsin-like serine protease. Proc Natl Acad Sci USA 85:1184, 1988

19. Yanagi M, Yoshikai Y, Leggett K, Clark SP, Aleksander I, Mak TW: A human T-cell-specific cDNA encodes A protein having extensive homology to immunoglobulin chains. Nature 308: 145, 1984

20. Harvey RC, Marteneire C, Sun LHK, Williams D, Showe L C Translocations and rearrangements in T-cell acute leukemias with the t( 1 1 ; 14) (p 1 3:q 1 I ) chromosomal translocations. Oncogene 6:34 I , 1989

2 1. Le Franc MP, Forster A, Baer R, Stinson MA, Rabbitts TH: Diversity and rearrangement of the human T-cell rearranging y genes: Nine germ line variable genes belonging to two subgroups. Cell 45: 237, 1986

22. Takayama H, Trenn G, Sitkovsky MV: A novel cytotoxic T lymphocyte activation assay. Optimized conditions for antigen receptor-triggered granule enzymatic secretion. J Immunol Methods 104:183, 1987

23. Murphy M, Loudon R, Kobayashi M, Trinchieri G: Gamma interferon and lymphotoxin released by activated T cells, synergize to inhibit granulocyte-monocyte colony formation. J Exp Med 164: 263, 1986

24. Cesano A, Santoli D: Inducible expression of granulocyte- macrophage colony-stimulating factor, tumor necrosis factor a, and interferon-? in two human cytotoxic leukemic T-cell lines. In Vitro Cell Dev Biol 28A:657, 1992

25. Maziarz RT, Arceci RJ, Bernstein SC, Frazier L, Smith BR, Kasai M, Tautravahi R, Strominger J L A y6+ T-cell leukemia bearing a novel t(8; 14) (q24;q 1 1) translocation demonstrates spontaneous in vitro natural killer-like activity. Blood 79: 1523, 1992




from a T-ALL sample engrafted in SCID mice Establishment of a ...

Documents

Transcript of from a T-ALL sample engrafted in SCID mice Establishment of a ...