BCR-ABL-positive lymphoblastoid cells display limitedproliferative capacity under in vitro culture conditions

ANDREW SPENCER, XIU-HUA YAN, ANDREW CHASE, JOHN M. GOLDMAN AND JUNI A V. MELO

Leukaemia Research Fund Centre for Adult Leukaemia, Royal Postgraduate Medical School, London

Received 18 April 1996; accepted for publication 10 June 1996

Summary. We studied the kinetics of EBV-transformed B-celllines from patients with chronic myeloid leukaemia (CML)using RT-PCR for BCR-ABL transcripts and immunoglobulinheavy chain (IgH) gene rearrangements. Peripheral bloodmononuclear cells, obtained from four patients with CML inchronic phase and from one in accelerated phase, wereincubated with supernatant from the B95-8 EBV producingcell line. In 11/25 (44%) B-cell cultures established wedemonstrated the presence of BCR-ABL transcripts atintervals ranging from 32 to 125 d post EBV transformation.In all but two cases, evidence of BCR-ABL transcriptsdisappeared with time. Cultures were initially polyclonal

with respect to IgH rearrangements but became progressivelyoligoclonal, suggesting the longer-term survival of fewerclones, all of which were BCR-ABL negative. We concludethat BCR-ABL-positive lymphoid cultures can be establishedin the short term from the majority of patients with CML butthey have limited capacity to survive in the longer term.Therefore, in lymphoid cells the presence of the BCR-ABLchimaeric gene appears to confer no survival advantage.

Keywords: BCR-ABL, CML, EBV transformation.

Philadelphia chromosome (Ph) positive chronic myeloidleukaemia (CML) is considered to be a disorder of pluripotenthaemopoietic stem cells with involvement of the myeloid,erythroid and monocytic/megakaryocytic lineages (Fialkowet al, 1977). Several lines of work support the notion that theB-lymphoid lineage is also part of the leukaemic clone, at leastin some patients. Studies in glucose-6-phosphate dehydro-genase (G-6PD) heterozygotes have demonstrated that B-lymphoid lines exclusively expressing the same G-6PD type asthe leukaemic cell population can occur (Fialkow et al, 1978)and that in some patients the apparent excess of such lines isstatistically significant (Raskind et al, 1993). These findingssuggest the possibility of B-lymphoid involvement in theleukaemic clone at a pre-Philadelphiachromosome stage, butremain unsubstantiated. Ph-positive Epstein-Barr virus (EBV)transformed lymphoblastoid cell lines from a small number ofCML patients have been produced, but appear to have onlylimited viability when compared to their Ph-negative counter-parts (Martin et al, 1980; Nitta et al, 1985; Raskind et al,1993). We have examined the immortalization and survivalof lymphoblastoid cells from a further series of CML patientsutilizing the reverse transcriptase-polymerase chain reaction(RT-PCR) for the detection of both the BCR-ABL gene

transcript and rearrangements of the immunoglobulinheavy chain (IgH) gene locus.

MATERIALS AND METHODS

Peripheral blood was obtained from five CML patientsattending the Hammersmith Hospital on an outpatient basisbetween August 1994 and January 1995. One patient wasconsidered to be in accelerated phase on the basis ofthrombocytosis (UPN 387) and two were Ph chromosomenegative but had a BCR-ABL gene rearrangement detected byRT-PCR (UPN 971 and UPN 105). Appropriate consent wasobtained in each case. Following centrifugation on Lympho-prep (Nycomed, Oslo, Norway), 1� 106 peripheral bloodmononuclear cells (PBMC) were resuspended in 1 ml of RPMI1640 (Gibco, Paisley) supplemented with 20% fetal calfserum (Gibco, Paisley). An equal volume of supernatent fromthe B95-8 EBV producing cell line was added and the cellswere then incubated at 378C for 1 h with occasional gentleagitation. Following incubation, the cell suspensions werediluted with medium to a final concentration of 1� 106/mland 2 ml was added to each of five 25 cm3 tissue culture flasksand maintained at 378C with 5% CO2. Two aliquots of1� 107 PBMC from one patient (UPN 105) were establishedat the same time giving a total of 10 cultures. Fresh mediumwas added twice weekly to each flask until a total volume of

British Journal of Haematology, 1996, 94, 654–658

654 # 1996 Blackwell Science Ltd

Correspondence: Dr Junia V. Melo, Haematology Department,Royal PostgraduateMedical School, DuCane Road, London W12 0NN.

655Proliferative Capacity of BCR-ABL�ve Lymphoblastoid Cells

# 1996 Blackwell Science Ltd, British Journal of Haematology 94: 654–658

10 ml was obtained. Therefore each flask was demi-depopu-lated once or twice per week and fresh medium added.

Successful transformation was present by day 30 followingEBV incubation as evidenced by foci of proliferating mono-morphic mononuclear cells, clumping of which was macro-scopically visible in most instances. At various intervals posttransformation 0:5� 105 to 2� 107 cells were harvested(Fig 1) following gentle agitation. The timing of the harvest-ing and pellet size obtained were dependent on theproliferative rate of the different cultures. The pellets weredisrupted by vortexing and the cells were then lysed in 500�lof guanidium-thiocyanate (GTC) solution. If the pelletscontained less than 106 cells, 5�g E. coli tRNA was addedas a carrier to the GTC lysate. Total RNA was prepared fromthe cell lysates by the method of Chomczynski & Sacchi

(1987) and cDNA synthesized as previously described (Crosset al, 1994). A negative control in the form of 500�l of GTCsolution was included in each batch of RNA extractions.

PCR amplifications for BCR-ABL and ABL (to confirmadequate cDNA synthesis) were performed as describedelsewhere (Diamond et al, 1995). A nested PCR using 1�lof first-step product as template was performed if the first-stepamplification for BCR-ABL was negative. RT-PCR for thedetection of IgH rearrangements was performed using theconsensus primers NV2 and NJ2 complementary to the FR3and JH regions of the IgH gene respectively. Amplification wasperformed in 20�l volumes containing 10 mM Tris pH 8.3,50 mM KCl, 1.5 mM MgCl2, 250�M each dATP, dCTP, dGTP anddTTP, 0.5�M of each primer, 0.2 U Taq polymerase and 0.5–1.0�g of cDNA template and carried out on a programmable

Fig 1. Detection of BCR-ABL positivity within B-lymphoblastoid cell populations. At least one positive culture was obtained from each patient. BCR-ABL positivity was detected with either single-step RT-PCR (UPN 661 and UPN 387) or on nested RT-PCR (UPN 971, UPN 105 and UPN 295). Allviable cultures were cryopreserved on the day indicated. An aliquot of cryopreserved cells from UPN 971/1 and UPN 105/1 were subsequently re-established in culture but were BCR-ABL negative on repeat testing. UPN � unique patient number. Open circles�BCR-ABL-negative population;closed circles�BCR-ABL-positive population.

heating block (MJ Research Inc., Watertown, Mass., U.S.A.) for36 cycles of 968C for 30 s, 608C for 50 s and 728C for 1 min,followed by a 10 min extension at 728C. In each case a 5�laliquot of reaction product was electrophoresed on a 2%agarose gel to confirm adequate amplification. A 1�l aliquot ofeach was then reamplified with NV2/NJ2 under the sameconditions incorporating 32P dCTP. The products were thenelectrophoresed on 8% polyacrylamide gels and visualized afterovernight exposure to radiographic film at ÿ708C. Thesequences of the oligonucleotide primers used are shown inTable I.

Cytogenetic analysis using standard methods withoutaddition of B-cell mitogens was performed at the time ofinitial harvesting (UPN 295) or subsequently (UPN 661, day76, and UPN 387, day 160) on the material from the threePh-positive patients. 30 metaphases were analysed in each

instance except in four cases where 8, 10, 12 or 19metaphases were examined respectively.

All surviving cultures were cryopreserved in 10%DMSO from 85 to 266 d post immortalization. Aliquots fromflasks UPN 971/2 and UPN 105/1 were subsequentlysuccessfully re-established in culture and re-analysed for thepresence of BCR-ABL with repeat RT-PCR.

RESULTS

Lymphoblastoid cell lines were established from all fivepatients; in two cases (UPN 387, day 32, and UPN 105,day 36) early manipulation and cell harvesting wasassociated with the loss of viability in some of the cultures.In all other cases the established lines appeared capable ofproliferating indefinitely but were eventually cryopreserved(see Fig 1). Cytogenetic studies were performed at intervals(range 73–107 d post EBV transformation) on cells from thethree Ph-positive patients. Whenever possible, 30 metaphaseswere examined. No Ph-positive metaphases were identified.

RT-PCR for BCR-ABL transcripts was performed at variousintervals (range 32–125 d post EBV transformation) on theproliferating cells (Fig 1). All five cultures established frompatient UPN 387, the only patient in accelerated phase, werepositive when first tested on day 32, the earliest time at whichanalysis was undertaken in any of the patients studied. Thedetection of BCR-ABL positivity was greatest early in theculture period: 5/7 viable cultures which were initiallypositive were found to be negative on repeat testing at alater date. The BCR-ABL positivity seen on day 32 with UPN387 and on day 46 with UPN 661 was demonstrable with asingle-step PCR reaction. The remaining positive results wereseen only after nested PCR.

# 1996 Blackwell Science Ltd, British Journal of Haematology 94: 654–658

656 Andrew Spencer et alTable I. Oligonucleotide primers.

GenePrimer Sequence location

B2A� TTCAGAAGCTTCTCCCTGACAT BCR (b2)B2B� ACAGCATTCCGCTGACCATCAATAA BCR (b2)NIb� CACGAATTCTGGAAAGGGGTACCTATTA ABL (Ib)Jcÿ GGAGTGTTTCTCCAGACTGTTG ABL (III)CA3ÿ TGTTGACTGGCGTGATGTAGTTGCTTGG ABL (III)NV2 GGACAC(GC)GC(CT)G(CT)GTATTACTG FR3NJ2 CT(CT)ACCTGCAGAGAC(AG)GTGACC JH

FR3� framework 3 region of the immunoglobulin heavy chainvariable gene; JH� immunoglobulin heavy chain joining gene.

Fig 2. Sequential IgH gene fingerprints on threecultures which initially were BCR-ABL positive.All display spontaneous progression towardsoligoclonality with coincident loss of BCR-ABLpositivity.

657Proliferative Capacity of BCR-ABL�ve Lymphoblastoid Cells

# 1996 Blackwell Science Ltd, British Journal of Haematology 94: 654–658

RT-PCR for IgH rearrangements was performed on serialspecimens from each culture. The pattern was initiallypolyclonal but later became oligoclonal, an observationconsistent with progressive spontaneous loss of some cloneswith increasing dominance of others. The rate of thisprogression varied between the different cultures and wasassociated in most instances with the loss of initial BCR-ABLpositivity (Fig 2). Two cultures, UPN 971/2 and UPN 105/1,displayed prolonged positivity with BCR-ABL transcriptsdetected on day 125 and day 87 post immortalization,respectively. Neither culture was harvested early aftertransformation and remained oligoclonal at the time ofcryopreservation (data not shown). Aliquots from bothcultures were subsequently thawed and re-establishedunder the same initial culture conditions and maintainedan equivalent pre-cryopreservation proliferative rate. Repeatanalysis for BCR-ABL in both instances proved to be negative,suggesting that BCR-ABL-positive cells could not be immor-talized while their normal counterparts continued to growvigorously.

DISCUSSION

We have demonstrated BCR-ABL positivity in a proportion ofthe lymphoblastoid cell cultures established from all fivepatients studied. This suggests that at least some proportion ofthe circulating B lymphocytes in each of these patients werepart of the leukaemic clone. The highest frequency of BCR-ABL culture positivity was seen on day 32 post transforma-tion in lymphocytes from the patient with accelerated-phasedisease (UPN 387). Whether this reflected a higher leukaemicB-lymphocyte burden than in the chronic phase patients ispurely speculative, because it would also be consistentwith the presence of greater numbers of viable BCR-ABL-positive lymphocytes earlier, rather than later, in the cultureperiod.

We cannot formally exclude the possibility that the RT-PCRpositivity for BCR-ABL observed early after EBV transformationwas due to residual myeloid cells in the culture, but two lines ofevidence argue against this interpretation. First, a comparisonof cord blood mononuclear cells infected with EBV withuninfected control cells showed that only dying cells and celldebris were present in the control cultures at 21 d (Walls &Crawford, 1987). Second, in long-term cultures of PBMC fromCML patients there was a marked reduction in leukaemic cellnumbers within 10 d of initiation (Udomsakdi et al, 1992).Therefore it is highly unlikely that viable myeloid cells stillcapable of generating BCR-ABL transcripts were present in ourcultures at 32 d and even more improbable that such cells couldhave survived beyond 70 d.

In all but two cases, cultures which were initially positivefor BCR-ABL and which remained viable, were negative onsubsequent testing. This is consistent with the progressive lossof viability of the BCR-ABL-positive lymphoblastoid popula-tions over time associated with the gradual emergence ofdominant BCR-ABL-negative clones. Previous attempts toestablish lymphoblastoid cell lines of leukaemic origin fromCML patients have met with limited success. Nitta et al (1985)

grew Ph-positive lines from 2/16 chronic phase patients; themajority of lines from these two cases were Ph negative, 14/17 and 19/20, respectively. This, combined with the casereport of Martin et al (1980) which suggested that Ph-positivelines had limited potential for survival, led to the notion thatleukaemic B lymphocytes may be disadvantaged at thetransformation stage in such experiments and possibly havea limited ability to proliferate (Raskind et al, 1993).

Our findings show that transformation of leukaemic Blymphocytes is possible in a significant number of cases, butafter initial successful establishment these clones are super-seded by normal counterparts as the cultures spontaneouslybecome more oligoclonal. Thus, it would appear that theleukaemic lymphoblastoid cells, unlike the Ph-negativelymphoblastoid cells, show a diminished capacity for trueimmortalization and indefinite proliferation under theseconditions. This is analogous to the situation observed afterthe establishment of primitive marrow progenitors fromCML patients in long-term cultures. A 30-fold decreasein leukaemic long-term culture initiating cells (LTC-ICs)was evident by 10 d associated with a relative maintenancein the number of normal counterparts (Udomsakdi et al,1992) and provided a rationale for autografting CMLpatients with marrow after short-term culture (Barnett et al,1994).

Previous work has shown the relative inefficiency of P210BCR-ABL expression to transform murine lymphoid progenitorsin a mouse model (McLaughlin et al, 1987). The level of BCR-ABL expression in EBV transformed chronic phase B lympho-cytes is significantly lower than the levels in spontaneouslygrowing BCR-ABL-positive myeloid and erythroid cell lines(Konopka et al, 1986). The failure to see a proliferativeadvantage with the BCR-ABL-positive lymphoblastoid cellsmay be due to an inherent lymphoid resistance to BCR-ABLtransformation. Alternatively, an interaction incompatible withlong-term survival may arise due to the introduction of EBV intoa BCR-ABL-positive cell. The ability of Nitta et al (1985) to growpredominantly Ph-positive lines in two patients with advanceddisease, in contrast to the chronic-phase patients, suggests thatadditional oncogenic abnormalities may be necessary toprovide the leukaemic B-lymphoid cell population with anysurvival advantage under in-vitro conditions and is consistentwith our findings.

In conclusion, we have shown that BCR-ABL-positivelymphoblastoid cell lines can be established from CMLpatients, which suggests that leukaemic B lymphocytes arepresent in the circulation in the majority of patients.Furthermore, we have shown that these cells exhibit onlylimited proliferative capacity in comparison to their non-leukaemic counterparts, suggesting that the presence of theBCR-ABL gene rearrangement confers no survival advantageunder in-vitro conditions.

ACKNOWLEDGMENTS

We thank Mr Peter Pollard for providing us with thesequences for the primers NV2 and NJ2.

Dr Andrew Spencer is supported by the LeukaemiaResearch Fund.

REFERENCES

Barnet, M.J., Eaves, C.J., Phillips, G.L., Gascoyne, R.D., Hogge, D.E.,Horsman, D.E., Humphries, R.K., Klingemann, H.G., Lansdorp,P.M., Nantel, S.H., Reece, D.E., Shepherd, J.D., Spinelli, J.J.,Sutherland, H.J. & Eaves, A.C. (1994) Autografting with culturedmarrow in chronic myeloid leukemia: results of a pilot study. Blood,84, 724–732.

Chomczynski, P. & Sacchi, N. (1987) Single-step method of RNAisolation by acid guanidium thiocyanate–phenol–chloroformextraction. Analytical Biochemistry, 162, 156–1659.

Cross, N.C.P., Melo, J.V., Feng, L. & Goldman, J.M. (1994) An optimizedmultiplex polymerase chain reaction (PCR) for detection of BCR-ABLfusion mRNAs in hematological disorders. Leukemia, 8, 186–189.

Diamond, J., Goldman, J.M. & Melo, J.V. (1995) BCR-ABL, ABL-BCR,BCR and ABL genes are all expressed in individual granulocyte-macrophage colony-forming unit colonies derived from bloodof patients with chronic myeloid leukemia. Blood, 85, 2171–2175.

Fialkow, P.J., Denman, A.M., Jacobson, R.J. & Lowenthal, M.N. (1978)Chronic myelocytic leukemia: origin of some lymphocytes fromleukemic stem cells. Journal of Clinical Investigation, 62, 815–823.

Fialkow, P.J., Jacobson, R.J. & Papayannopoulou, T. (1977) Chronicmyelocytic leukemia: clonal origin in a stem cell common to thegranulocyte, erythrocyte, platelet and monocyte/macrophage.American Journal of Medicine, 63, 125–132.

Konopka, J.B., Clark, S., McLaughlin, J., Nitta, M., Kato, Y., Strike, A.,Clarkson, B. & Witte, O.N. (1986) Variable expression of thetranslocated c-abl oncogene in Philadelphia-chromosome-positive

B-lymphoid cell lines from chronic myelogenous leukemia patients.Proceedings of the National Academy of Sciences of the United States ofAmerica, 83, 4049–4052.

Martin, P.J., Najfeld, V., Hansen, J.A., Penfold, G.K., Jacobson, R.J. &Fialkow, P.J. (1980) Involvement of the B-lymphoid system inchronic myelogenous leukaemia. Nature, 287, 49–50.

McLaughlin, J., Chianese, E. & White, P.N. (1987) In vitrotransformation of immature hematopoietic cells by the P210BCR/ABL oncogene product of the Philadelphia chromosome.Proceedings of the National Academy of Sciences of the United States ofAmerica, 84, 6558–6562.

Nitta, M., Kato, Y., Strike, A., Wachter, M., Fried, J., Perez, A.,Jhanwar, S., Duigou-Osterndorf, R., Chaganti, R.S.K. & Clarkson, B.(1985) Incidence of involvement of the B and T lineages in chronicmyelogenous leukemia. Blood, 66, 1053–1061.

Raskind, W.H., Ferraris, A.M., Najfeld, V., Jacobson, R.J., Moohr, J.W. &Fialkow, P.J. (1993) Further evidence for the existence of a clonalPh-negative stage in some cases of Ph-positive chronic myelocyticleukemia. Leukemia, 7, 1163–1167.

Udomsakdi, C., Eaves, C.J., Swolin, B., Reid, D.S., Barnett, M.J. & Eaves,A.C. (1992) Rapid decline of chronic myeloid leukemic cells inlong-term culture due to a defect at the leukemic stem cell level.Proceedings of the National Academy of Sciences of the United States ofAmerica, 89, 6192–6196.

Walls, E.V. & Crawford, D.H. (1987) Generation of human Blymphoblastoid cell lines using Epstein-Barr virus. Lymphocytes:a Practical Approach (ed. by G. G. B. Klaus), pp. 149–162. IRL Press,Washington.

# 1996 Blackwell Science Ltd, British Journal of Haematology 94: 654–658

658 Andrew Spencer et al