Determinants of Deoxyadenosine Toxicity in Hybrids Between Human T

[CANCER RESEARCH 45. 1579-1586, April 1985]

Determinants of Deoxyadenosine Toxicity in Hybrids between Human T- and B- Lymphoblasts as a Model for the Development of Drug Resistance in T-Cell Acute Lymphoblastic Leukemia

Joanne Kurtzberg1 and Michael S. Hershfield2

Department of Pediatrics, Division of Hematology-Oncology [J. K.Â¡,and Departments of Medicine and Biochemistry, Division of Rheumatology and Genetics [M. S. H.], Duke University Medical Center, Durham, North Carolina 27710

ABSTRACT lymphoid cancers with a potent ADA inhibitor, dCF (42, 49-51). Several specific effects of the ADA substrates Ado and dAdo Cultured human T-lymphoblastoid cell lines are more sensitive are known, including altered metabolism of cAMP (60), inhibition than B-cell lines to 2'-deoxyadenosine in the presence of 2'- of pyrimidine nucleotide synthesis (25), S-adenosylmethionine- deoxycoformycin, a potent inhibitor of adenosine deaminase. dependent transmethylation (28, 30, 32, 43, 45), and ribonucle- This difference is related to the greater efficiency with which T- otide reducÃase(40,41,47,58), and activation of ATP catabolism lymphoblasts accumulate cytotoxic levels of dATP derived from the adenosine deaminase substrate 2'-deoxyadenosine (dAdo). (5,54,62). Together or individually, these effects could contribute to the immune defect in genetic ADA deficiency in various ways, Previous work has shown that differences in dATP accumulation by interfering with lymphocyte differentiation, function, or viabil by cultured T- and B-lymphoblastoid cell lines cannot be ex ity. However, in determining the effectiveness of ADA inhibition plained by large differences in the levels of dAdo-phosphorylating as a treatment for lymphocytic leukemia, the most important or dAdo nucleotide (dAXP)-degrading activities in cytoplasmic factor is selective cytotoxicity, which is due primarily to the extracts of these cells, although it has been proposed that intact expansion of the intracellular pool of dATP derived from dAdo B-cell lines may catabolize intracellular dAXP more rapidly than (8,12,13,17). do T-cell lines. To further examine the determinants of dAdo Of the various types of ALL, T-ALL has been most responsive sensitivity in T- and B-lymphoblasts, we have studied dAdo and dAXP metabolism in the human T- and B-cell lines CEM and Wl- to treatment with dCF (26, 42, 51, 53, 56). Both in vivo during dCF treatment and in vitro when incubated with dCF and dAdo, L2 and in hybrids generated by fusion of these cell lines. The T-lymphoblasts accumulate large amounts of dATP and are killed hybrid nature of the fusion products was established by nutri (26, 56). In contrast, cultured B-lymphoblastoid cell lines accu tional studies and by analyses of cellular surface antigens, DNA mulate much less dATP from dAdo than do T-cell lines, and they content, and enzymatic activities. We found that WI-L2 x CEM are more resistant to dAdo toxicity (7,9,48, 52). Mutations that hybrids and another T x B hybrid derived from fusion of the SB abolish the ability of T-cell lines to trap dAdo as intracellular human B-cell line with CEM were 30- to 40-fold less sensitive to dAdo and about 10-fold less sensitive to the dAdo analogue 9- dATP, such as loss of ability to transport or phosphorylate dAdo, markedly diminish the sensitivity of T-cell lines to dAdo in vitro /Ã®-D-arabinofuranosyladeninethan was CEM, or about as resist ant as were their B-cell parental lines. Our studies confirm that (29, 59). Likewise, we have observed the emergence of relative resistance to dCF therapy in a patient with T-ALL, which was CEM avidly accumulates dAXP from dAdo but does not ca associated with loss of the ability of the patient's leukemia cells tabolize intracellular dAXP. In contrast, WI-L2, SB, and WI-L2 x CEM and SB x CEM hybrids rapidly degraded intracellular dAXP, to accumulate dATP in vivo (31). which limited their ability to undergo dAXP pool expansion. Attempts have been made to determine the basis for the 1Ci to 100-fold greater rate of dATP accumulation that occurs in T- Expression of dAXP catabolic activity in T x B hybrids behaved as a dominant mechanism, conferring resistance to dAdo- and lymphoblasts compared with B-lymphoblasts when they are dAdo-related nucleosides to T x B hybrids. It has been postu incubated with dAdo in the presence of ADA inhibitors. Different lated that cell fusion may play a role in the progression of tumors rates of dATP accumulation could result from different levels of and contribute to diversity among the cells that compose clonal enzymes that phosphorylate dAdo or degrade dAdo nucleotides tumors. We have speculated that fusion of a malignant T-lym- (dAXP). The levels of total dAdo-phosphorylating activity and of phoblast with an activated B-cell might be a mechanism for the the two specific enzymes capable of catalyzing this reaction, evolution of drug resistance in acute T-cell leukemia. dCyd kinase and Ado kinase, differ by no more than 2- to 4-fold in cytoplasmic extracts of a number of T- and B-cell lines (9, 29). More rapid dAXP catabolism by B-cell lines than by T-cell lines INTRODUCTION has been proposed (7,10, 61). However, the enzyme(s) specifi Genetic deficiency of ADA3 (EC 3.5.4.4) results in profound cally involved in dAXP catabolism has not been identified and, as with dAdo phosphorylation, only 2- to 4-fold differences in lymphopenia, with relative sparing of other organ systems (23, 44). This selectivity has provided a rationale for the treatment of total cytoplasmic nucleotidase activity have been found in ex tracts of T- and B-cell lines (10). Received 8/13/84; revised 12/26/84; accepted 12/31/84. 1Supported by NIH Training Grant 5T32 AM 07015. Fellow of the Leukemia mycin; Ado, adenosine; dAdo, 2'-deoxyadenosine; ALL, acute lymphoblastic leu Society of America and a recipient of a Junior Faculty Clinical Fellowship of the kemia; PEG, polyethylene glycol; HGPRT, hypoxanthine-guanine phosphoribosyl- American Cancer Society. To whom requests for reprints should be addressed, at transferase; QUA, ouabain; OHA, ouabain + hypoxanthine + azaserine; PNP, Box 2916, Duke University Medical Center, Durham, NC 27710. purine nucleoside phosphorylase; dCyd, deoxycytidine; AdoHcyase, S-adenosyl- 2 Supported by NIH Grant AM 20902 and Research Career Development Award homocysteine hydrolase; dAXP, total deoxyadenosyl nucleotides; ECÂ»,50% effec AM 00434. tive concentration (concentration of drug required to inhibit growth by 50%); ara- 3The abbreviations used are: ADA, adenosine deaminase; dCF, 2'-deoxycofor- A, 9-/3-c-arabinofuranosyladenine; ara-C, 1-0-D-arabinofuranosylcytosine.

CANCER RESEARCH VOL. 45 APRIL 1985

The finding of similar levels of nucleoside kinase and nucleo- line had been grown to cause the conversion of dAdo to adenine (in the tidase activities in extracts of T- and B-lymphoblastoid cells, presence of 5 ^M dCF to inhibit ADA) or thymidine to thymine or by a despite large differences in their rates of dATP accumulation, cytochemical stain procedure (11). suggests that there are significant differences in the regulation OHA selective medium consisted of RPM11640 + 10% horse serum of dAdo phosphorylation or dAXP catabolism in intact T- and B- + 1 mw pyruvate + nonessential amino acids + 50 nw OUA + 100 ^M hypoxanthine + 5 Â»Mazaserine. These drug concentrations were se lymphoblasts. In earlier studies, we examined the effects of mutational loss of dCyd or Ado kinases on dAdo-mediated dATP lected to completely inhibit growth of the appropriate parent but slow growth of the other parent by no more than 25%. Growth in OHA medium pool expansion in the human T- and B-cell lines CEM and WI-L2 selects for cells that are OUA-resistant and possess HGPRT activity, (29, 30, 59). We found that dAXP accumulation was primarily properties expected of the desired hybrids. dependent on dCyd kinase in the T- (29) and Ado kinase in the Cell Fusion. Fusions were performed by a method originally developed B- (30,59) cell line. Moreover, both kinases appeared to operate for murine myeloma-splenocyte fusions (22) with minor modifications. far more actively in dAdo phosphorylation in the intact T-cell line Briefly, 2 x 106 CEM and 2 x 106 WI-L2 were mixed in a 15-ml conical than in the B-cell line. We postulated that rapid dAXP degradation centrifuge tube and pelleted; 0.5 ml of a 50% solution of PEG in RPMI might account for the apparent inactivity of dAdo-phosphorylat- 1640 medium without serum was added dropwise over 1 min, followed ing enzymes in WI-L2 (29). by dropwise addition of 11 ml of RPMI 1640 without serum over the next 6 min. The cells were pelleted, resuspended in 20 ml of OHA In order to obtain more information about the nature of the selective medium, and distributed in 150-^1 aliquots in 96-well tissue mechanisms that regulate dAdo nucleotide accumulation in intact culture plates (Costar) at a final density of 2 x 10* cells/well. In every T- and B-cell lines, we have used the somatic cell genetic fusion, controls were performed by omitting PEG and by carrying out the technique of generating cell hybrids by fusing WI-L2 and CEM. procedure with each parent alone. We have compared the hybrid and parental cell lines with respect Cell Surface Antigenic Phenotype. HLA-A.B.C. and DR typing were to their sensitivity to dAdo and other nucleosides and their performed by the Duke Medical Center Serologie Laboratories by com abilities to accumulate and degrade intracellular dAXP. In addition plement mediated microcytotoxicity (2). to shedding light on the regulation of dAdo metabolism, our Chromosome Analysis. Logarithmically growing cells were exposed studies suggest a possible mechanism by which resistance to to Colcemid (0.1 n9/ml) for 2 h. Cells were then harvested, and slides chemotherapeutic nucleoside analogues may develop in lymph- were prepared for analysis by Giemsa staining according to established protocols (63). At least 50 mitoses were examined by photomicroscopy oid cancers. for determination of chromosome number. DNA content per cell was determined by flow microcytofluorimetry after staining with acridine orange (3, 14). Ten ml of logarithmically MATERIALS AND METHODS growing cells in growth medium were pelleted and resuspended in phosphate-buffered saline containing 0.5% Nonidet P-40 + 20% fetal Materials. Hypoxanthine, aminopterin, QUA, 6-thioguanine, Nonidet calf serum. Ten min later, the cells were stained by addition of 1 ml of a P-40, azaserine, ara-C, ara-A, 8-azaguanine, and dAdo were purchased solution containing 0.002% acridine orange, 0.1 M NaCI, and 0.01 M from Sigma Chemical Co. (St. Louis, MO). [8-14C]Adenosine and citrate-phosphate buffer at pH 3.8. After 15 min at room temperature, [2-3H]dAdo were obtained from Moravek Biochemicals (Brea, CA), [5- the fluorescence intensities of the individual cells were measured in a 3H]Cyd and [8-14C]AMP were from Amersham/Searle Corp. (Arlington flow cytofluorometer (Cytofluorograf 50H; Ortho Diagnostic Instruments, Heights, IL), and [8-14C]dAdo, [2-14C]dCyd, and [8-14C]hypoxanthine Bio/Physics Systems, Westwood, MA). Detailed description of the in were from New England Nuclear (Boston, MA). Polyethyleneimine cellu strument is presented elsewhere (6,18). The results were analyzed using lose thin-layer plates were from Merck. PEG (Carbowax PEG 1000) was a multichannel analyzer and plotted as frequency distribution histograms purchased from Fisher. Acridine orange was purchased from Polysci- of DNA content. ences, Inc. (Warrington, PA). dCF (Pentostatin) was a gift from Warner Enzyme Assays. Extracts of 2 to 5 x 107 cells were prepared for Lambert-Parke Davis (Detroit, Ml). Tiazofurin was provided by the Na assay by methods that involved disrupting cells by freezing and thawing, tional Cancer Institute. Other chemicals and materials used were of the followed by centrifugation and chromatography of supernatants over highest available quality and were purchased as described in the refer 0.7- x 20-cm Sephadex G-25 columns (30). The assay conditions have ences cited. been described previously for measurement of AdoHcyase (28), ADA Cell Lines and Culture Methods. The HGPRT-contaming (HGPRT -). (57), HGPRT (33), total dAdo-phosphorylating activity (30), dCyd kinase OUA-sensitive (OUAS) human splenic B-lymphoblastoid cell line WI-L2 (29), ecto-5'-nucleotidase (61), PNP (21), Ado kinase (29), and dCyd (46) has been used in this laboratory for several years. Cell line CEM.AG1.OU1.5 is an HGPRT-deficient (HGPRT~), OUA-resistant deaminase(16). Growth Inhibition by Nucleosides. The effects of the ADA substrates (OUAR) clone of the human malignant T-lymphoblastoid cell line CCRF- dAdo and ara-A (in the presence of 5 UMdCF) and of ara-C and tiazofurin CEM (19). In the text and charts, CEM.AG1.OU1.5 is referred to simply on growth of the parent and hybrid cell lines were examined after 72 to as CEM or "the T-cell parent." The T x B hybrid cell line SB.CEM.1 (34), 96 h of exposure. Results were compared to controls exposed to dCF derived by fusing the human B-cell line SB (1) with a CEM.AG1.OU1.5 alone or to no drug using the formula clone, was obtained from D. Howell and P. Cresswell, Department of [Final cell count - initial cell count] in presence of drug Immunology, Duke University Medical Center. Cells were routinely cul tured in RPMI 1640 (Grand Island Biological Co., Grand Island, NY) [Final cell count - initial cell count] in absence of drug supplemented with 10% horse serum, nonessential amino acids, and 1 mM pyruvate under an atmosphere of 5% CO2 in air at 37 Â°C.Cloning Results were expressed as ECÂ»(concentration of drug required to inhibit was performed by limiting dilution either in growth medium or in semisolid growth by 50%). medium containing 0.22% agarose (Sea-Kern; Marine Colloids, Inc., Measurement of dAXP Accumulation. Cultures at a cell density of 5 x 105 cells/ml in growth medium were incubated at 37Â°Cwith 5 UM dCF Rockville, ME), using as feeder cells (in a separate, lower 0.22% agarose layer) mouse macrophages obtained by sterile 0.9% NaCI solution irri for 1 h, followed by addition of various concentrations of dAdo (range, 5 gation of the peritoneal cavity or human fetal foreskin fibroblasts, which to 500 pM). After 3 or 4 more h of incubation, duplicate aliquots were were kindly provided by Dr. Sheldon Pinnell. All experiments and cell removed for harvesting of cells and extraction with HCIO4 as described fusions were performed with cultures in the logarithmic phase of growth (30). Total adenine deoxyribonucleotides (dAXP = dATP + dADP + (4 to 8 x 105 cells/ml). Cultures were deemed free of Mycoplasma dAMP) in neutralized HCIO4 extracts of cells were measured by high- contamination by the inability of conditioned medium in which the cell pressure liquid chromatography after dephosphorylation of these nucleo-

CANCER RESEARCH VOL. 45 APRIL 1985 1580 Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1985 American Association for Cancer Research. dAdO TOXICITY IN LYMPHOBLAST HYBRIDS tides to dAdo by a modification (29) of a procedure described previously Table 1 (30). dAdo was quantitated by high-pressure liquid chromatography on Growth ol parental cell lines and hybrid offspring in selective medium a Partisil PXS 10/25 SCX column eluted with 50 mw NaH2PO4, pH 2.6, Growth at 72 or 96 h" with at a flow rate of 1.5 ml/min. the following cell line6 Measurement of dAXP Degradation. Duplicate cultures of CEM, Wl- T-cell B-cell L2, SB, the WI-L2 x CEM hybrid (TBJ.1E6), and the SB x CEM hybrid, Selective condition parent parent Hybrid each at a cell density of 5 x 105 cells/ml in growth medium, were incubated with 5 //M dCF for 1 h at 37 Â°C.Then, WI-L2, SB, and the OUA (50nM)Azaserine *IM)Azaserine(5 hybrids were incubated for an additional 3 h with 500 Ã•/MdAdo and CEM >iM)OHA6-Thioguanine(5 ^M) + hypoxanthine (100 for 2 h with 20 /IM dAdo, conditions chosen to cause approximately equal dATP accumulation in the B-cell lines and CEM (see "Results"). ^M)Aminopterin (10 The cultures were chilled on ice for 5 to 10 min and centrifuged at 4 Â°C, pu)HATHAT (1 and the medium was decanted. The cell pellets were washed once with + OUA (50 nM)+<=000+00000+000+0+0++00++ 10 ml of ice-cold medium containing 5 UM dCF without dAdo. Centrifuga- "Controls grown in parallel in growth medium containing no additions underwent tion was repeated, the wash medium was aspirated, and the cells were 3 to 4 cell doublings. resuspended at the original cell density in 37 Â°Cgrowth medium that ^â€¢-cellparent, CEM.AG1.OU1.5; B-cell parent, WI-L2; hybrid, TBJ.1E6; OHA, contained 5 MM dCF without dAdo. Duplicate aliquots of cultures were 50 nw QUA + 100 UM hypoxanthine + 5 Â»Mazaserine; HAT, 100 Â¿JMhypoxanthine + 1 >iMaminopterin + 20 UMthymidine. then harvested for measurement of intracellular dAXP as described (29, C0, <10% of control growth; +, >75% of control growth. 30), immediately (0 time) and after further incubation at 37Â°for 30, 60, 90,180, and 300 min.

RESULTS

General Characterization of Hybrids

A fusion of CEM.AG1.OU1.5 with WI-L2 yielded growth in 3 wells from two 96-well tissue culture plates. Cells from these wells were cloned in OHA medium in 0.22% agarose layers. Six hybrid clones were recovered and placed in long-term culture in OHA for 2 months and then in regular growth medium. No growth was observed in control experiments in which either parental line was fused with itself or when PEG was omitted from the fusion protocol. As an additional control, we also developed a T- x T- cell hybrid by crossing a CEM.AG1 .OU1.5 clone with the HGPRT+, OUAS CEM line from which it was derived. The behav ior of this T- x T-cell hybrid cell line is discussed in various FLUORESCENCE INTENSITY Â» sections below. We also conducted some studies with one other Chart 1. DNA histograms of T-cell [CEM.AG1 .OU1.5 ( )] and B-cell [WI-L2 T- x B-cell hybrid clone, SB.CEM.1, which was derived from the ( )] parents and representative CEM x WI-L2 cell hybrid clones [TBJ.1E6.C2 ( ) and TBJ.1 E9.C1 (â€”)]. DNA content was measured via flow microcytofluor- fusion of a clone of CEM.AG1 .OU1.5 with the B-cell line SB. imetry of acridine stained cultured cells as outlined in "Materials and Methods." The hybrid nature of the cloned CEM x WI-L2 fusion products Similar studies of the SB x CEM hybrid showed that it contained higher DNA was established on the basis of their growth in various selective content than did either SB or CEM (data not shown). conditions, their karyotype and DMA content, the expression of surface antigens, and by measurement of cell-associated en number for the CEM x WI-L2 hybrids was 122 (range, 117 to zyme activities. The 6 cloned products of CEM x WI-L2 fusions 130), compared with 92 (range, 86 to 96) for CEM and 89 (range, were found to have very similar characteristics, which have 70 to 93) for WI-L2. remained stable after almost 2 years in culture. Representative Surface Antigens. Expression of human class I and class II data from one isolate (TBJ.1E6) are presented in more detail; major histocompatibility antigens was examined in the hybrid data from the other hybrid clones are summarized. and parental cell lines (Table 2). With the exception of CW2 (a Growth in Selective Media. The hybrids all grew in selective class I antigen expressed only on the B-cell parent), the HLA A, media that failed to support the growth of one or both parents. B, and C antigens and DR antigens of both parents were Sensitivity of the hybrids to thioguanine and to azaserine in the expressed by the hybrids. In addition, the hybrids expressed one absence of hypoxanthine established that they contained new antigen, expressed on neither parent, at each HLA locus. HGPRT activity and were not simply azaserine-resistant mutants The phenomenon of novel class II (DR) antigen expression in T- of the OUA-resistant T-cell parent. The T- x T-cell hybrid control cell-B-cell hybrid lymphoblastoid cell lines has been observed exhibited identical growth properties in the media shown in Table previously in studies of the regulation of HLA phenotype in T- 1 as the T- x B-cell hybrids, but it could be easily distinguished and B-cells (34, 35). It was concluded that the new DR marker from them on the basis of HLA markers and deoxynucleoside was encoded by the genome of the T-cell parent but was not metabolism, as discussed below. expressed in the undifferentiated or unactivated state. Our find DNA Content. Based on cytofluorometric analysis of nuclear ings suggest that hybrid formation may also result in activation DMA content, WI-L2 and CEM cells contained similar amounts of class I as well as class II loci, but it is unclear whether these of DNA, while the DNA content of the hybrids was clearly greater loci were derived from the T- or the B-cell parent. than that of either parent (Chart 1). These relationships were Enzyme Activities. Cytoplasmic enzyme activities were mea confirmed by chromosome counts. The modal chromosome sured in Sephadex G-25-treated cell extracts for CEM, WI-L2,

CANCER RESEARCH VOL. 45 APRIL 1985 1581

Table 2 systems were >10-fold more resistant to ara-A-mediated growth HLA phenotypes of parental and hybrid cell lines inhibition than was CEM and were slightly more resistant than Expression by the following cell line8 was either B-cell parent. HLA antigen T-cell parent B-cell parent Hybrid Hypothetically, resistance to nucleosides might be a general property of T- x B-lymphoblast hybrids. To assess this possibil A1 A2 ity, we compared the sensitivities of the hybrid and parental lines Aw 19 to the nucleoside analogues tiazofurin and ara-C. Tiazofurin is B8 converted to an analog of NAD, which inhibits inosinic acid B17 Bw4 dehydrogenase (36). Our unpublished studies indicate that tia Bw6 zofurin, unlike dAdo and ara-A, is activated by neither dCyd nor Cw2 Ado kinase. In contrast to dAdo and ara-A, tiazofurin is about Cw3 DR3 10-fold more toxic to WI-L2 than to CEM, with EC.Â»values for DR7 the B- and T-cell lines of 6 and 60 nu, respectively. The hybrid DRw8 was as sensitive to tiazofurin as was WI-L2 (Chart 2C). Like "As in Tabte 1. dAdo and ara-A, ara-C is phosphorylated by dCyd kinase, but ara-C is only about 2-fold more toxic to CEM than WI-L2 (Chart Table3 2D). The WI-L2 x CEM hybrid TBJ.1E6 was about as sensitive Enzymespecific activities in parental and hybrid cell lines to ara-C as was WI-L2. These studies show no evidence of a Activities were measuredin Sephadex G-25-treated cell lysates as described in "Materials and Methods" using the indicated concentrations of radioactive nucleo- general resistance of the hybrids to nucleosides, so that it is side substrate. valid to address the resistance of the hybrids to dAdo specifically with the following cell in terms of the regulation of dAdo and dAdo nucleotide metab line*T-cell8175 tration of olism by the parental T- and B-cell lines. labeled There is evidence that dCyd kinase in WI-L2 might be less substrate efficient than the enzyme in CEM in activating dAdo. Thus, dAdo- M175 B-cellhybrid2560 induced dAXP pool expansion by WI-L2 was unaffected by ADAAdo 3170 kinase 100 20.0 11.0 15.0 mutational loss of dCyd kinase, but it was eliminated by loss of dCyd kinase 50 9.0 2.0 2.0 Ado kinase (30,59). Similar kinds of studies with kinase-deficient dAdo kinase" 100 19.0 7.0 7.0 AdoHcyase 150 320 320 320 mutants indicated that, in CEM, dCyd kinase is primarily respon PNP 200 2976 2997 2467 sible for phosphorylation of dAdo at concentrations <20 /IM, dCyd deaminaseConcen 200nmol/h/mg0.13TxB0.68nmol/h/10.97 while Ado kinase is active at higher concentrations (29). Con 0scells sistent with these observations, it has been found that exoge Ecto-5'-nucleotidase 1000 9915 2310 nous dCyd was much less effective in blocking dAdo-mediated "As in legend to Table 1. dATP pool expansion or toxicity in WI-L2 (30,59) compared with ^"otal phosphorylating activity for dAdo, since this nucleosidecan be phospho- CEM (9). In light of these observations, we examined the effect rylated by both dAdo kinase and dCyd kinase. of dCyd on dAdo toxicity to WI-L2 x CEM hybrids. dCyd (20 JIM) increased the ECÂ«Â»valuesfor dAdo of CEM by 4- to 6-fold but and several of the WI-L2 x GEM hybrids. The most complete had almost no effect on dAdo toxicity to WI-L2 or the CEM x data were obtained with the TBJ.1E6 clone (Table 3). Several WI-L2 hybrid. It is possible that the ineffectiveness of dCyd in activities that were 2- to 4-fold higher in GEM than WI-L2 were diminishing dAdo toxicity to WI-L2 and the WI-L2 x CEM hybrid expressed at the WI-L2 level in the hybrid [ADA, total dAdo- is related to the 5- to 7-fold higher dCyd deaminase activity in phosphorylating activity (dAdo kinase), and dCyd kinase]. PNP, these lines as compared to CEM (Table 3). AdoHcyase, and Ado kinase activities were similar in the T- and B-cell lines and the hybrid. Ecto-5'-nucleotidase, a cell surface- The greater dependence of dAXP accumulation on dCyd ki nase in CEM compared with WI-L2 does not appear to be the associated activity that is expressed at much higher levels on B- lymphoblasts than on T-lymphoblasts (7, 61), was undetectable result of different properties of the dCyd kinases present in these on OEM and was expressed on the hybrid at about one-fourth cell lines. The apparent Kms for dAdo of dCyd kinase partially purified from WI-L2 and CEM were about 120 ^M (29, 30). The the level of WÃ•-L2.dCyd deaminase activity, 7-fold higher in WÃ•- L2 than in CEM, was 5-fold higher in the hybrid than in GEM. value for the similarly isolated enzyme from the hybrid TBJ.1E6 was somewhat lower, 39 UM, but there was no evidence of the Nucleoside Sensitivity and Metabolism presence of 2 kinetically distinct forms of the enzyme (Chart 3). Moreover, our earlier studies had indicated that Ado kinase, as Growth Inhibition by Nucleosides. Growth of the parental well as dCyd kinase, operated very inefficiently in WÃ•-L2com and TBJ.1E6 hybrid cell lines in the presence of varying concen pared with CEM (29). It seems unlikely that T- and B-lympho- trations of dAdo or ara-A plus 5 MMdCF is illustrated in Chart 2, blasts possess different forms of both Ado kinase and dCyd A and B, and data from all 6 of the WI-L2 x CEM, the SB x kinase. A more likely explanation for the apparent inefficiency of CEM, and the CEM x CEM hybrid clones are summarized in dAdo-phosphorylating enzymes in WI-L2 is that the cell line Table 4. The B-cell parents, SB and WI-L2, were 65- and 19-fold possesses a dominantly acting inhibitor of either the kinases more resistant to dAdo-mediated growth inhibition than was the themselves or an enhanced dAXP degrading activity. The latter T-cell parent, CEM. Both the WI-L2 x CEM and the SB x CEM mechanism is suggested by the following studies of dAXP ac hybrids displayed "B-like" resistance to dAdo, with EC60values cumulation and degradation in the intact hybrid and parental cell 30- to 40-fold higher than the value for CEM. In contrast, the lines. CEM x CEM hybrid was as sensitive as was CEM to dAdo Accumulation of Adenine Deoxyribonucleotides. dAXP ac (Table 4). The hybrids in both the WI-L2/CEM and SB/CEM cumulation was measured in ADA-inhibited cells after 4 h of

CANCER RESEARCH VOL. 45 APRIL 1985 1582

100

; 80 O 60

025 05 1 25 IO 50 200 [2'deoxyodenosine] M M

100 J 100 I tâ€” Â§80 Â§80 oOC 260 2 60 o Iu 40 u U. o O 20 ** 20

-I 1 001 01 1 5 10 50 100 0.01 005 0.1 025 [Tiazofunn] >u.M [Ara-Cl/uM

Chart 2. Effects of dAdo (A),ara-A (B),tiazofurin (C),and ara-C (D)on the growth rates of T-cell [CEM.AG1.OU1.5(â€¢)],B-cell[WI-L2 (â€¢)],andT- x B-hybrid [TBJ.1E6 (A)] cell lines. The indicated concentrations of dAdo or ara-A were added to exponentially growing cells in the presence of 5 IÃ•MdCF.dCF was not used in studies with tiazofurin and ara-C. Cell growth is plotted as a percentageof cell growth after 72 to 96 h in the absenceof nucleoside

Table 4 Comparisonof sensitivity to dAdo and ara-A in T-cell, B-cell, and hybrid cell lines All experiments contained 5 *IMdCF. Growth inhibition[ECÂ»OÂ»M)]: CelllineCEM(T) dAdo0.74 ara-A0.48 Â±0.66s(7)" Â±0.24(3) WI-L2 (B) 48.5 Â±9.47 (4) 3.9 Â±1.86(3) SB(B) 14.0 Â±4.36 (3) 1.4C CEM x WI-L2 (T x B) 27.7 Â±5.35 (7) 5.3 Â±2.4 (3) CEM x SB (T x B) 21.8 Â±5.89 (5) 3.5C CEM x CEM (T x T)By 0.52 Â±0.39 (5)By Not done 'Mean Â±SD. "Numbers in parentheses, number of experiments. Â°Experimentperformedonce. incubation in growth medium containing 5, 25, or 125 /Â¿MdAdo 025 (Table 5). At lower concentrations of dAdo, dAXP accumulation 005 Q01 0 15 by the B-cell lines was barely detectable and, at 125 /JM, they 1/dAdo, /uM accumulated roughly 10% of that by CEM. The WI-L2 x CEM Charts. Effect of substrate concentration on the rate of phosphorylation of deoxyadenosine by dCyd kinase from a CEM/WI-L2 hybrid. Extracts of 5 x 107 hybrid accumulated only slightly more dAXP than did WI-L2. The cells of TBJ.1E9 were prepared, centnfuged. and chrornatographed on Utrogel SB.CEM.1 hybrid accumulated about 20% as much dAXP as AcA 34 as described previously (29). Abscissa, reciprocal of the concentration of CEM at 25 tiu dAdo and nearly as much as CEM at 125 MM-At ['HJdAdo concentration,y-ax/s, reciprocalof the reaction velocity. Inset, ptot of the concentration of [3H]dAdo(x-axis)versus velocity (y-axis, arbitrary units). very high concentrations of dAdo (500 MMor greater), the accu mulation of dAXP by the WI-L2 x CEM hybrid also approached that by CEM (not shown). transferred to dAdo-free medium and the rate of fall in dAXP Catabolism of Intracellular dAXP. In order to study the rate content is followed. It is desirable but difficult to achieve equal of catabolism of intracellular dAXP, it is necessary to first expand levels of intracellular dAXP in the cells under study, given the this pool by incubating cells with dAdo, after which the cells are marked difference in rates of dAXP accumulation between T-

CANCER RESEARCH VOL. 45 APRIL 1985 1583

Tables DISCUSSION dAdo nucleotide accumulation in parental and hybrid cell lines Cultures were incubated with 5 JIMdCF plus the indicated concentration of dAdo for 4 h at 37Â°C,after which cells were harvested and the intracellular content of The objective of this study was to determine whether hybrids dAXP was determined by high-pressure liquid chromatography, as described in â€¢Materialsand Methods." The dAXP content in cells not exposed to dAdo has formed by fusing a dAdo-sensitive T-cell line with a dAdo- resistant B-cell line would retain the sensitivity of one or the been subtracted from the values shown in the table. lines8dAdo nmol dAXP/10* cells with cell other parent or simply express intermediate sensitivity. We hoped to evaluate our results in terms of the mechanisms that B may underlie the different sensitivities of the parental cell lines UM)A. hybrid2 to dAdo and related nucleosides. Since T-cell lines are also more WI-L2/CEM 5 sensitive than are B-cell lines to the cytotoxic effects of thymi- 25 145 6 1055 125 524 20 dine, we thought that use of the standard hypoxanthine-aminop- B. SB/CEM 5 72 <5 32 terin-thymidine medium to select for hybrids [which had been 348 <5 25 68 used in isolating the SB x CEM hybrid (34)] might result in 125T-cell42 1096B-cell<1 52Tx 960 preferential isolation of thymidine-resistant clones that might be aT-cell line, CEM.AG1 .OU1.5; B-cell line, WI-L2 (Experiment A), SB (Experiment cross-resistant to dAdo. We therefore selected for WI-L2 x CEM B); T x B, TBJ.1E6 (Experiment A), SB.CEM.1 (Experiment B). hybrids in medium containing azaserine, which inhibits purine synthesis but not thymidylate synthesis, and hypoxanthine, which is a nontoxic source of purines to cells that possess HGPRT activity. A similar type of selection has been developed independently for the isolation of hybrids involving peripheral blood T-lymphocytes (20). == 2 The 6 WI-L2 x CEM hybrid clones we isolated were all much more resistant to dAdo toxicity than was their T-cell parent and were nearly as dAdo-resistant as was their B-cell parent. Like wise, the SB.CEM.1 hybrid was as resistant to dAdo as was its B-cell parent, SB. "B-ness" was also dominant in both the Wl- 06 L2 x CEM and SB x CEM systems with respect to the toxicity of ara-A, which is phosphorylated by the same kinases that act on dAdo and which may achieve toxicity in part by mechanisms 02 .WI-L2Â«CEM similar to those of dAdo. These results contrast with the finding that tiazofurin, a nucleoside that differs from dAdo and ara-A in O I 2345 its mechanism of activation and toxicity, was about 10-fold more HOURS toxic to WI-L2 and a WI-L2 x CEM hybrid than to CEM. Thus, Chart 4. Catabolism of intracellular dAXP by T-cells [CEM (O)], B-cells [WI-L2 resistance to nucleoside toxicity is not an inevitable attribute of IÂ«)and SB (D)J, and T- x B-hybnds [WI-L2 X CEM (A) and SB x CEM (A)]. T- x B-hybrids. Duplicate cultures (3 to 5 x 105 cells/ml in mid-kxj phase) were preincubated in The limited ability of dAdo to cause dATP pool expansion has growth medium with dAdo plus 5 M" dCF to expand intracellular dAXP pools. Then cells were chilled, washed, and resuspended in medium at 37 Â°Cthat contained been proposed as the sole explanation for the relative resistance dCF without dAdo, and the intracellular dAXP concentration was measured imme of WI-L2 and other B-cell lines to dAdo (30). However, there may diately (0 h) and at several time points over a 5-h period. The preincubation also be differences between B- and T-cell lines in the intracellular conditions were chosen so as to expand the dAXP pools in the T-cells and B-cells to the same degree. To achieve this, the B-cells were incubated for 3 h with 500 target(s) affected by dATP, as suggested by the finding that JIMdAdo, and the T-cells were incubated for 2 h with 20 Â«MdAdo. The T- x B-cell dAdo arrested CEM cells in the d phase of the cell cycle but hybrids were incubated under the same conditions as those of the B-cells. blocked WI-L2 cells in S phase (37). Thus, our present results, which show that B-like resistance is dominant in T- x B-hybrids, and B-cell lines. For the experiment shown in Chart 4, we could indicate either that the dATP target in B-cells is less attempted to achieve similar levels of dAXP in CEM and the B- sensitive to dATP than that of T-cells or that a dominant-acting cell lines (WI-L2 and SB) by incubating the T-cell line with 20 UM mechanism in the B-cell lines is responsible for limiting dAdo- dAdo for 2 h and the B-cell lines with 500 ?M dAdo for 3 h (both mediated dATP pool expansion. Our studies shed no direct light in the presence of dCF); the hybrids (TBJ.1E6 and SB.CEM.1) on the identity of the dATP target, but the fact that the SB.CEM.1 were incubated under the same conditions as those of the B-cell hybrid accumulated substantially more dATP than did its B-cell lines. After transfer to dAdo-free medium, the dAXP content of parent, SB, but was as resistant to dAdo as was SB, is consistent WI-L2 and SB returned to base-line levels within 1 h. In contrast, with the possibility that SB contains a dATP-insensitive target. there was virtually no decrease in dAXP concentration in CEM Our studies with the WI-L2/CEM and SB/CEM systems pro in 5 h. dAXP content fell about as rapidly in the hybrids as in the vide direct evidence in favor of the hypothesis (7, 61) that more B-cell parental lines. The initial rate of dAXP catabolism in the B- rapid catabolism of dAXP is an important factor in explaining the cell lines and T- x B-cell hybrids was at least 20-fold greater difference in ability of dAdo to cause dATP pool expansion in T- than that of the T-cell. The catabolism of intracellular dATP has and B-cell lines. It is clear from our results that a dAXP-degrading also been examined in CEM and the B-cell line 894 (7). In that activity, present in WI-L2 and SB but not CEM cells, is expressed more limited experiment, the disappearance of dATP was fol in WI-L2 x CEM and SB x CEM hybrids. At the low concentra lowed for less than 30 min after cells were placed in dAdo-free tions of dAdo that are cytotoxic to CEM cells, the rate of medium at unphysiologically high cell density. No decline in dATP operation of this nucleotidase activity appears sufficient to pre was observed in CEM, while the level in the B-cell line, which vent dAXP accumulation and toxicity in WI-L2 and SB and in Wl- was about 10% that of CEM, fell by about one-half. L2 x CEM and SB x CEM hybrids. The identity of this nucleoti-

CANCER RESEARCH VOL. 45 APRIL 1985 1584

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1985 American Association for Cancer Research. dAdO TOXICITY IN LYMPHOBLAST HYBRIDS dase has not been definitively established, although it seems as a means of evaluating the occurrence and consequences of likely that it is a cytoplasmic rather than a cell surface-associated hybrid formation in vivo. activity. If expression of this activity in T- x B-lymphoblastoid cell hybrids is a general phenomenon and, if this expression ACKNOWLEDGMENTS confers a dAdo-resistant phenotype to hybrids, then it may be possible to map the chromosomal location of the gene for the The authors appreciatethe expert technical assistance of Janet E. Misenheimer nucleotidase by constructing hybrids between a dAdo-sensitive and the assistance of Laurita G. Melton in the preparation of the manuscript. We murine T-cell line and a dAdo-resistant human B-cell line such as also gratefully acknowledge the expert help of Dorothy Reedand Dr. Jaciyn Siegel in studies determiningchromosome number and helpfuldiscussions with Dr. David WI-L2. Moreover, expression of this activity should be domi- Howell concerning expression of HLA antigens by the hybrids. nantly selectable in T-cell lines, which could provide a strategy for isolation of its gene using DMA transfer methodology. REFERENCES It has been proposed that cell fusion involving malignant and nonmalignant cells or between phenotypically variant tumor cells 1. Adams, R. A. Formaldiscussion:the role of transplantation in the experimental investigationof humanleukemiaand lymphoma.Cancer Res., 27:2479-2482, may contribute to the progression of tumors or to the evolution 1967. of phenotypic heterogeneity common in tumors (4, 15, 27, 38, 2. Amos, D. B., and Pool, P. HLA typing. In: N. R. Rose and H. Friedman(eds.), 39). Evidence of such fusion has been obtained experimentally Manual of Clinical Immunology, pp. 797-804. Washington, DC: American Society for Microbiology, 1976. in studies of transplantable lymphoid tumors (24), and Sinkovics 3. Ashihara,T., TrÃ¡ganos,F.,Baserga, R., and Darzynkiewicz, Z. A comparison (55) has purposed that hybrid formation involving antibody- of cell cycle-related changes in postmitotic and quiescent AF8 cells as mea producing plasma cells may be a natural process in lymphopro- sured by cytofluorometry after acridine orange staining. Cancer Res., 38: 2514-2518,1978. liferative diseases. In general, discussions of the role of hybrid- 4. Atkin, N. B. Prematurechromosomecondensation in carcinomaof the bladder: presumptiveevidencefor fusion of normal and malignantcells. Cytogenet. Cell generated diversity in tumors have focused on the development Genet., 23:217-219,1979. of more invasive or metastatic characteristics, properties that 5. Bagnara, A. S., and Hershfield,M. S. Mechanismof deoxyadenosine-induced influence the progression of the tumor prior to therapeutic inter catabolismof adenineribonudeotides in adenosinedeaminase-inhibitedhuman T-)ymphoblastoidcells. Proc. Nati. Acad. Sci. USA, 79: 2673-2677,1982. vention. Our studies suggest that fusion between a malignant, 6. Gambier,J. C., Havran, W. L., Fernandezde Albornoz, T., and Cortey, R. B. dAdo-sensitive T-cell with an activated B-cell might result in the Identification of a brain 0-positive, secretory cell from hematopoietic tissues. acquisition of characteristics that would confer a selective ad J. Immunol., 727:1685-1691, 1981. 7. Carson, D. A., Kaye, J., Matsumoto, S., Seegmiller,J. E., and Thompson, L. vantage during treatment with dCF or ara-A. In this regard, it is Biochemicalbasis for enhancedtoxicity of deoxyribonucleosidestoward malig interesting to consider clinical and biochemical observations that nant human T-cell lines. Proc. Nati. Acad. Sci. USA, 76: 2430-2433, 1979. we reported recently of a patient with T-cell acute lymphoblastic 8. Carson, D. A., Kaye, J., and Seegmiller, J. E. Lymphospecific toxicity in adenosine deaminase deficiency and purine nucleoside phosphorylase defi leukemia who underwent a series of treatments with dCF (31). ciency: possible role of nucleoside kmase(s) Proc. Nati. Acad. Sci. USA, 74: After initially responding to dCF infusion, this patient's leuke 5677-5681,1977. 9. Carson, D. A., Kaye, J., and Seegmiller,J. E. Differential sensitivity of human mia became relatively resistant to dCF treatment, which still leukemic T-cell lines and B-cell lines to growth inhibition by deoxyadenosine. caused marked elevation in plasma dAdo. Resistance was as J. Immunol., 121:1726-1731,1978. 10. Carson, D. A., Kaye,J., and Wasson, D. B. The potentialimportanceof soluble sociated with a striking decline in accumulation of dAXP in deoxynucleotidase activity in mediating deoxyadenosine toxicity in human circulating leukemia cells in vivo which could not be accounted lymphoblasts.J. Immunol., 126: 348-352,1981. 11. Chen, T. R. In situ detection of mycoplasmacontamination in cell cultures by for by any alteration in ability of dCF or dAdo to enter the fluorescent HOECHST33258 stain. Exp. Cell Res., 104: 255-262,1977. malignant cells or in the activities in cell extracts of dCyd or Ado 12. Cohen, A., Hirschhorn, R., Horowitz, S. D., Rubinstein, A., Polmar, S. H., kinases. Thus, the malignant cells, which still bore T-cell surface Hong, R., and Martin, D. W., Jr. Deoxyadenosinetriphosphate as a potentially markers, had acquired the "B-like" characteristic of inability to toxic metabolite in adenosine deaminase deficiency. Proc. Nati. Acad. Sci. USA, 75:472-476,1978. respond to increased levels of exogenous dAdo by dAXP accu 13. Coleman, M. S.., Donofrio,J., Hutton, J. J., Hahn, L., Daoud, A., Lampkin, B., mulation, despite adequate levels of dAdo-phosphorylating activ and Dyminski,J. Identificationand quantification of adeninedeoxynucleotides in erythrocytes of a patient with adenosine deaminase deficiency and severe ity. combined immunodeficiency.J. Bid. Chem., 253:1619-1626,1978. We were not able to conduct studies of dAXP degradation 14. Darzynkiewicz, Z., TragÃ³nos,F.,Sharpless,T., and Melamed,M. R. Lympho cyte stimulation: a rapid multiparameteranalysis. Proc. Nati. Acad. Sci. USA, with leukemic cells from this patient, but it is not unreasonable 73:2881-2884,1976. to consider the possibility that diminished dAXP accumulation 15. De Baetselier,P. E., Gorelik, E., Eshhar,Z., Ron, Y., Katzav, S., Feldman,M., and Segal, S. Metastatic properties conferred on nonmetastatic tumor hybrid may have resulted from an enhanced rate of dAXP catabolism, ization of spleen B-lymphocyteswith plasmacytomacells. J. Nati. Cancer Inst.. given the present evidence that degradation is a dominant factor 67: 1079-1087,1981. that determines the ability of dAdo to cause dAXP pool expan 16. de Saint Vincent, B. R., Deschamps,M., and Buttin, G. The modulation of the thymidine triphosphate pool of Chinese hamster cells by dCMP deaminase sion. Acquisition of increased capacity to degrade dAXP could and UDPreducÃase.J.Biol. Chem., 255: 162-167, 1980. have arisen by several mechanisms, such as gene amplification 17. Donofrio, J., Coleman, M. S., Hutton, J. J., Daoud, A., Lampkin, B., and Dyminski, J. Overproduction of adenine deoxynucleosides and deoxynucleo or a change in regulation of gene expression. However, in light tides in adenosine deaminase deficiency with severe combined immunodefi of our present studies and proposals regarding the possible role ciency disease.J. Clin. Invest., 62: 884-887,1978. of cell fusion in tumor progression, we think it intriguing to 18. Dorval, G.. Welsh, K. I., and Wigzell. H. Labeled staphylococcal protein A as an immunological probe in the analysis of cell surface markers. Scand J. speculate that resistance to dCF treatment in this patient might Immunol.,3:405-411,1974. have been the result of the fusion of a leukemia T-lymphoblast 19. Foley, G. E., Lazarus, H., Sidney, F., Uzman, B. G., Boone, B. A., and McCarthy, R. E. Continuous culture of human lymphoblasts from peripheral with an activated B-cell. The postulated type of fusion might also blood of a child with acute leukemia.Cancer (Phila.).78: 522-529, 1965. play a role in the evolution of resistance to other chemothera- 20. Foung, S. K. H., Sasaki, D. T., Grumet, F. C., and Engleman,E. G. Production peutic nucleoside analogues such as ara-A and ara-C, which are of functional human T-T-hybridomas in selection medium lacking aminopterin and thymidine. Proc. Nati. Acad. Sci. USA, 79: 7484-7488,1982. converted to cytotoxic nucleotides. It may be fruitful to monitor 21. Fox, I. H., Andres, C. M., Gelfand, E. W., and Biggar, D. Purine nucleoside for changes in karyotype and cell surface antigens, as well as phosphorylase deficiency: altered kinetic properties of a mutant enzyme. Science(Wash. DC), 797:1084-1086,1977. rates of analogue nucleotide accumulation and degradation, in 22. Gerhard, W. Fusion of cells in suspension and outgrowth of hybrids in condi leukemia cells during the course of treatment with these agents tioned medium. In: R. H. Kennett, T. J. McKearn, and K. B. Bechtol (eds.),

CANCER RESEARCH VOL. 45 APRIL 1985 1585

Monoclonal Antibodies, pp. 370-371. New York: Plenum Publishing Corp., 42. Koller,C. A., Mitchell, B. S., Grever, M. R., Mejias, E., Malspeis,L., and Metz, E. N. Treatment of acute lymphoblastic leukemia with 2'-deoxycoformycin: 1980. 23. Giblett, E. R., Anderson, J. E., Cohen, F., Pollara, B., and Meuwissen, H. J. clinical and biochemical consequences of adenosine deaminase inhibition. Adenosmedeammasedeficiency in two patients with severelyimpairedcellular Cancer Treat. Rep., 63:1949-1952,1979. immunity. Lancet, 2: 1067-1069,1972. 43. Kredich, N. M., and Hershfield, M. S. S-Adenosylhomocysteine toxicity in 24. Goldenberg,D. M., PavÃa,R.A., and Tsao, M. C. In vivo hybridisationof human normal and adenosine kinase-deficient lymphoblasts of human origin. Proc. tumour and normal hamster cells. Nature (Lond.), 250: 649-651,1974. Nati. Acad. Sci. USA, 76: 2450-2454,1979. 25. Green, H., and Chan, T. S. Pyrimidine starvation induced by adenosine in 44. Kredich, N. M., and Hershfield, M. S. Immunodeficiencydiseases associated dbroblasts and lymphoid cells: role of adenosine deaminase. Science (Wash. with adenosine deaminase and purine nucleoside phosphorylase deficiency. DC), 182: 836-837,1973. In: J. B. Stanbury, J. B. Wyngaarden, D. S. Fredrickson, J. Goldstein,and M. 26. Grever, M. R., Siaw, M. F. E., Jacob, W. F., Neidhart, J. A., Miser, J. S., Brown (eds.),The Metabolic Basis of Inherited Disease,pp. 1157-1183. New Ciernan, M. S., Hutton, J. J., and Balcerzak,S. P. The biochemicaland clinical York: McGraw-Hill, 1983. consequences of 2'-deoxycoformycin in refractory lymphoprolifarative malig 45. Kredich, N. M., and Martin, D. W., Jr. Role of S-adenosylhomocysteinein nancy. Blood, 57: 406-417,1981. adenosine-mediatedtoxicity in cultured mouse T-lymphoma cells. Cell, 72: 27. Hart, l. R. Tumor cell hybridization and neoplastic progression. In: G. L. 931-938,1977. Nicoison and L. Milas (eds.), Cancer Invasion and Metastasis: Biologic and 46. Levy, J. A., Virolainen, M., and Defendi, V. Human lyrnphobiastoidlines from Therapeutic Aspects, pp. 133-143. New York: Raven Press, 1984. lymph node and spleen. Cancer (Phila.),22: 517-524,1968. 28. Hershfield, M. S. Apparent suicide inactivatkm of human lymphoblast S- 47. Meuth, M., and Green, H. Alterations leading to increased ribonudeotide adenosylhomocysteine hydrolase by 2'-deoxyadenosine and adenine arabi- reducÃaseincells selected for resistance to deoxynucleosides. Cell, 3: 367- noside. J. Bid. Chem., 254. 22-25,1979. 374,1974. 29. Hershfield, M. S., Fetter, J. E., Small, W. C., Bagnara, A. S., Williams, S. R., 48. Mitchell, B. S., Mejias, E., Daddona, P. E., and Kelley, W. N. Purinogenic miman, B., Martin, D. W., Jr., Wasson, 0. B., and Carson, D. A. Effects of immunodeficiencydiseases: selective toxicity of deoxyribonudeosides for T- mutational loss of adenosine kinase and deoxycytidine kinase on deoxyATP cells. Proc. Nat).Acad. Sci. USA, 75: 5011-5014,1978. accumulation and deoxyadenosinetoxicity in cultured CEM human T-lympho- 49. Poplack. D. G., Sallan, S. E., Rivera, G., Holcenberg.J., Murphy, S. B., Blatt, blastoid cells. J. Bioi. Chem., 257: 6380-6386,1982. J., Upton, J. M., Venner, P., Glaubiger, D. L., Ungerleider,R., and Johns, D. 30. Hershfield, M. S., and Kredich. N. M. Resistance of an adenosine kinase- PhaseI study of 2'-deoxycoformycm in acute lymphoblasticleukemia. Cancer deficient human lyrnphobiastoid cell line to effects of deoxyadenosine on Res., 41: 3343-3346,1981. growth, S-adenosylhomocysteine hydrolase mactivation, and dATP accumu 50. Prentice, H. G., Ganeshaguru,R., Bradstock, K. F., Goldstone, A. H., Smyth, lation. Proc. NaÂ«.Acad.Sci. USA, 77: 4292-4296,1980. J. F., Wonke, B., Janossy, G., and Hoffbrand, A. V. Remissioninduction with 31. Hershfield, M. S., Kredich, N. M., Koller, C. A., Mitchell, B. S., Kurtzberg, J., adenosine-deaminaseinhibitor 2'-deoxycoformycin in Thy lymphoblastic leu Kinney, T. R., and Palletta, J. M. S-Adenosylhomocysteine catabolism and kemia. Lancet, 2:170-172,1980. basis for acquired resistance during treatment of T-cell acute lymphoblastic 51. Prentice, H. G., Russell, N. H., Lee, N., Ganeshaguru,K., Blacklock. H., Riga, leukemia with 2'-deoxycoformycin alone and in combination with 9-0-o-Arabi- A.. Smyth, J. F., and Hoffbrand, A. V. Therapeutic selectivityof and prediction nofuranosyladenine.Cancer Res., 43: 3451-3458,1983. of response to 2'-deoxycoformycin in acute leukemia. Lancet, 2: 1250-1253, 32. Hershfield, M. S., Kredich, N. M., Ownby, D. R., Ownby, H., and Buckley, R. 1981. In vivo inactivation of erythrocyte S-adenosylhomocysteinehydrolase by 2'- 52. Reynolds, E. C., Harris, A. W., and Finch, L. R. Deoxyribonucleosidetriphos deoxyadenosine in adenosine deaminase-deficient patients. J. Clin. Invest., phate poolsand differentialthymidmesensitivitiesof cultured mouselymphoma 63:807-811,1979. and myelomacells. Biochim. Biophys. Acta, 567:110-123,1979. 33. Hershfield, M. S., Snyder, F. F., and Seegmiller,J. E. Adenine and adenosine 53. Russell, N. H., Prentice, H. G., Lee, N., Riga,A., Ganeshaguru,K., Smyth, J. are toxic to human lymphoblast mutants defective in purine salvageenzymes. F., and Hoffbrand, A. V. Studies on the biochemical sequelae of therapy in Science (Wash. DC), 797: 1284-1287,1977. Thy-acute lymphoblastic leukemia with the adenosinedeaminase inhibitor 2'- 34. Howell, D. N., Berger, A. E., and Cresswell. P. Human T-B-lymphoblasthybrids deoxycoformycin. Br. J. Hematol., 49:1-9,1981. express HLA-DRspecificitiesnot expressed by either parent. Immunogenetics. 54. Siaw, M. F. E., Mitchell, B. S., Koller, C. A., Coteman,M. S., and Hutton, J. J. 15: 199-206,1982. ATP depletion as a consequence of adenosine deaminase inhibition in man. 35. Howell, D. N., and Cresswell, P. Expression of T-lymphoblast-encodedHLA- Proc. Nati. Acad. Sci. USA, 77:6157-6161,1980. DR antigens on human T-B-lymphoblast hybrids. Immunogenetics, 77: 411- 55. Sinkovics,J. G. Early history of specific antibody-producinglymphocyte hybridomas. Cancer Res., 47:1246-1247,1981. 425,1983. 36. Jayaram. H. N., Ahluwalia. G. S., Dion, R. L., Gebeyehu. G . Marquez, V. A., 56. Smyth, J. F., Paine, R. M., Jackman, A. L., Harrap, K. R., Chassin, M. M., Adamson, R. H., and Johns, D. G. The clinical pharmacologyof the adenosine Kelley.J. A., Robins, R. K., Cooney, D. A., and Johns, D. G. Conversionof 2- deaminase inhibitor 2'-deoxycoformycin. Cancer Chemother. Pharmacol., 5: .f-D-nbofuranosylselenazole-4-carboxamidetoan analogueof NAD with potent IMP dehydrogenase-inhibitory properties. Biochem. Pharmacol., 32: 2633- 93-101,1980. 57. Snyder, F. F., Mendelsohn,J., and Scsgmiller, J. E. Adenosine metabolism in 2636, 1983. phytohemagglutinin-stimulatedhuman lymphocytes. J. Clin. Invest., 58: 654- 37. Kefford, R. F., Helmer, M. A., and Fox, R. M. S-Adenosylhomocysteine hydrolase inhibition in deoxyadenosine-treated human T-lymphoblasts and 666,1976. 58. miman, B., Gudas, L. J., Cohen, A., and Martin, D. W., Jr. Deoxyadenosine resting peripheral blood lymphocytes. Cancer Res., 42:3822-3827,1982. metabolismand cytotoxicity in cultured mouse T-lymphomacells: a model for 38. Kerbel, R. S., Lagarde,A. E., Dennis,J. W., and Donaghue,T. P. Spontaneous immunodeficiencydisease. Cell, 74: 365-375, 1978. fusion in vivo between normal host cells and tumor cells: possible contribution 59. Ulman, B., Levinson, B. B., Hershfield, M. S., and Martin, D. W., Jr. A to tumor progression and metastasis studied with a lectin resistant mutant biochemicalgenetic study of the role of specific nucleoside kinases in deoxy tumor. Mol. Cell. Biol., 3: 523-538,1983. adenosinephosphorylationby cultured human cells. J. Biol. Chem., 256: 848- 39. Kerbel, R. S., Lagarde, A. E., Dennis, J. W., Nestel, F. P., Donaghue, T. P., 852,1981. Siminovitch, L., and Fulchignom-Lataud.M. C. Relevance of spontaneous in 60. Wolberg, G., Zimmerman, T. P., Heimstra, K., Winston, M., and Chu, L. C. vivo tumor-host cell fusion to tumor progression and metastasis evaluated Adenosine inhibition of lymphocyte-mediatedcytolysis: possible role of cyclic using a series of lectin-resistant mutant tumor suWines.In: G. L. Nicoison and adenosinemonophosphate.Science(Wash. DC), 787: 957-959,1975. L. Milas (eds.), Cancer Invasion and Metastasis: Biologic and Therapeutic 61. Wortmann, R. L., Mitchell, B. S., Edwards, N. L., and Fox, I. H. Biochemical Aspects, pp. 47-79. New York: Raven Press, 1984. basis for differential deoxyadenosine toxicity to T- and B-lymphoblasts: rote 40. Klenow, H. On the effect of some adenine derivatives on the incorporation in for 5'nucteotidase. Proc. Nati. Acad. Sci. USA, 76: 2434-2437,1979. vitro of isotopically labelledcompounds into the nucleicacids of ehrlichascites 62. Yu, A. L., Bakay, B., Kung, F. H., and Nyhan, W. L. Effects of 2'-deoxycofor- tumor cells. Biochim. Biophys Acta, 35: 412-421,1959. mycin on the metabolism of purines and the survival of malignant cells in a 41. Klenow, H. Further studies on the effects of deoxyadenosineon the accumu patient with T-cell leukemia. Cancer Res., 47: 2677-2682,1981. lation of deoxyadenosine triphosphate and inhibition of deoxyribonucleic acid 63. Yunis, J. J., Oken, M. M., Kaplan, M. E., Ensrud, K. M., Howe, R. R., and synthesis in Ehrlich ascites tumor cells in vitro. Biochim. Biophys. Acta, 67: Theologides,A. Distinctive chromosomal abnormalities in histologie subtypes 885-896, 1962. of non-Hodgkin's lymphoma. N. Engl. J. Med., 307:1231-1236,1982.

CANCER RESEARCH VOL. 45 APRIL 1985 1586

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1985 American Association for Cancer Research. Determinants of Deoxyadenosine Toxicity in Hybrids between Human T- and B-Lymphoblasts as a Model for the Development of Drug Resistance in T-Cell Acute Lymphoblastic Leukemia

Joanne Kurtzberg and Michael S. Hershfield

Cancer Res 1985;45:1579-1586.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/45/4/1579

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/45/4/1579. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.