British journal oj Haematology. 199 1, 79, 5 1 6-5 19

METHODS PAPER

Non-radioactive detection of immunoglobulin and T cell receptor gene rearrangement in acute lymphoblastic leukaemia

G E R A L D MARTIN A N D EMER LAWLOR Haematology Department, T.C.D. Medical School, Trinity College Dublin, Dublin 2, Ireland

Received 18 April 1991; acceptedfor publication 29 July 1991

Summary. Rearrangements of the heavy chain immuno- globulin gene and T cell receptor 1 gene were investigated in 2 5 patients suffering from precursor I3 cell acute leukaemia and six patients suffering from T cell acute leukaemia using biotinylated DNA probes. All precursor B acute leukaemia patients had IgH gene rearrangements and 63% of those

studied also had TCKP gene rearrangements. All T cell acute leukaemia patients had TCRB gene rearrangements and germline IgH configuration. Dilution experiments indicated that DNA from leukaemic cells representing 1-2% of a tested sample could be detected using this technique which com- pares favourably to radioactive DNA probes.

The ability to detect immunoglobulin (Ig) gene and T cell receptor (TCR) gene rearrangement is becoming an integral part of the evaluation of patients with acute lymphoblastic leukaemia (ALL) (British Society of Haernatology. 199 1). Standard Southern blotting techniques for the detection of rearrangements of the heavy chain lg and TCR /j chain genes have used radiolabelled DNA probes corresponding to the joining regions of these genes (Arnold et al. 1983: Wright et a/. 1987: Felix et al. 1987, 1990). However, due to the potential hazards and other disadvantages associated with the routine handling of radioactive probes, efforts have been made to use alternative, non-isotopic detection systems for Southern blot DNA analysis (Brigati et a/ , 1983; Leary et a!. 1983). The utilization of the highly specific and tenacious interaction between biotin and avidin or streptavidin has become the cornerstone of one such system (Gebeyehu et al. 1989). and advances in hybridization and detection methods have improved the sensitivity of this system to make it comparable to radiolabelled techniques (Chan et a/, 1985: Dykes, 1988). Some reservations have, however, been expressed as to the general success of these techniques (Worwood & Wagstaff. 1990). The aim of this study was to investigate the suitability of the biotin-streptavidin reaction for detecting gene rearrangements in ALL patients, and to compare the sensitivity of this system to radiolabelled probes.

Correspondence: Dr Emer Lawlor. Haematology Department, C.P.L.. St James' Hospital, Dublin 8. Ireland.

MATERIALS A N D METHODS

Patients were diagnosed according to standard French- American-British (FAB) morphological and cytochemical criteria (Bennett t r a/. 1983) and also according to immuno- logical phenotyping criteria defined by a panel of monoclonal antibodies (Janossy & Campana, 1989). Blood and bone marrow mononuclear cells were isolated by standard Ficoll- hypaque (Lyrnphoprep) separation. High molecular weight DNA was extracted from these cells following proteinase K digestion and 6 M NaCl precipitation of proteins (Miller ef a/ , 1988). DNA (10 p g ) was digested to completion with Eco RI. HindIII or HindIII and BamHl (Sigma or BRL), electrophor- esed through 0.6% agarose gels, and transferred to Sureblot nylon membranes (Oncor) by 2 x SSC (saline sodium citrate) Southern transfer (Reed & Mann. 1985).

Biotinylafed probes. J t 1 probes consisted of the nick trans- lation products of the 5.6 kb BamH 1/Hind Ill J,, region fragment (Oncor). TCRP J, and J r probes consisted of the nick translation products of a cloned 2.4 kb JP1 and a cloned 4 .2 kb 182 fragment (Oncor). Hybridization and detection steps were performed with solutions provided in a commercially available kit (Oncor). but with the following alterations to the manufacturer's protocol. Biotinylated DNA probes (1 2 ng/ ml) were hybridized over night at 45OC to the membrane- bound DNA in hybrisol Ill (Oncor). For optimal results, prehybridization and post stringency wash blocking reac- tions were performed at 45°C for 1 h, and the final stringency wash was at 55°C for 35-40 min. 1 0 min streptavidin and alkaline phosphatase detection steps were carried out in 6 . 5 ml buffer plus 6.5 pi detection chemical per 100 cmL membrane. Overnight staining was performed in the dark at

516

Nun-isotopic Detection of Gene Rearrangements 5 1 7 37°C with a solution that had been passed through a 0.2 pm filter immediately prior to use, containing 3.5 ml staining buffer plus 70 pl nitroblue tetrazolium (NBT) and 70 pI bromochloroindolyl phosphate (BCIP) per 100 cm2. All washes and incubations were performed in sealed plastic bags. Stock staining reagents were stored in the dark at 0- 4°C. If required, stained blots were destained using dimethyl- formamide and proteinase K digestion and cleaned for reprobing with formamide (Gebeyehu e t al, 1989).

Radioac-five probes. Kadiolabelled Jll probe (Oncor) was used for DNA dilution experiments to compare the sensitivity of the two systems. Prehybridization was for 2 h at 65°C in 5 x SSC. 1 "/o SDS. 5 x Denhardt's solution ( 5 0 x Denhardt's = 1% ficoll, 1%) PVP. l(x, BSA). 500 pg/rnl heparin and 100 pg/ml denatured salmon sperm DNA. Hybridization was performed overnight at h5'C in 5 x SSC. 1%) SDS. 5 x Denhardt's solution. 100/, dextran sulphate. 1 0 0 jcg/ml denatured salmon sperm DNA. and 1 x 10" dpm probe/ml. Two 1 5 min stringency washes were performed at hO°C in 2 x SSC. 0 . 1 % SDS followed by four 1 5 min washes in 3 x SSC. 0 . 1 % SDS. Autoradiographs were produced by 5-14 d exposures of Kodak XAK-5 X-ray film, processed according to manufac- turers' instructions.

Cellular detection limit. Dilutions of normal mononuclear cells with leukaemic cells were used to discern the cellular detection limit of the biotinylated DNA probe system. Stan- dard ethidium bromide/acridine orange staining under ultra- violet light (Lee et al, 1975) was used to count normal and leukaemic sample mononuclear cell numbers. These cells were then aliquoted and mixed to give a range of leukaemic/ normal cell dilutions each with a total of 2 .7 million cells. These dilutions contained 100% leukaemic cells: 1 0%, leuk- aemic and 90% normal cells: 5% leukaemic and 95% normal cells: 2% leukaemic and 98% normal cells; 1'% leukaemic and 99% normal cells: and 0% leukaemic and 1 OOo/, normal cells. DNA was extracted and digested to completion with Hind111 as described. Aliquots of DNA (10-1 1 j ig ) from each of these samples were probed for J l l gene rearrangement following electrophoresis and Southern blotting as before.

KESULTS

Figs 1 and 2 show representative blots of patient DNA samples probed with biotinylated JII and TCRP probes respectively. All of the 25 precursor B cell ALL patients studied showed rearranged Ig alleles. When sufficient DNA was available, two or more restriction enzymes were used to confirm Jll configuration. Of the 19 patients studied for TCKB gene rearrangements, 12 had non-germline configurations. By comparison. all of the T cell ALL patients had TCRP gene rearrangements and Ig JII in germline configuration. Data relating to numbers of patients, phenotype of their disease and rearrangement patterns are shown in Table 1. DNA dilution experiments indicated that the sensitivity of the biotinylated JII probe system was comparable to that of the radiolabelled J H probe (data not shown). The dilution experi- ments of patient mononuclear cells indicated that a detection sensitivity of 1-2% could be achieved by this assay (data not shown).

Fig 1. JH probed patient DNA. Lane 1 shows biotinylated lambda HindIII-digested size marker DNA: lane 2 shows control DNA: and lanes 3-10 show DNA from patients 1 5 . 11. 13. 32 (AML. -ve control), 26. 14. 19 and 20 respectively. All patient and control DNA lanes show 1 0 pg DNA digested to completion with both Hind111 and BamHI restriction enzymes. Fragment sizes are indicated in kilo- bases. Patients 19 and 20 appear in this blot to have two rearrangements each, but were seen to have one germline and two rearrangements. and three rearrangements respectively by EcoRI and Hind111 digestion (see Table I ) . indicating the need to confirm banding patterns with one or more further enzyme digests.

Fig 2. TCRB probed patient DNA. Lanes 1 and 12 show biotinylated lambda HindIII-digested size marker DNA: lanes 2 and I 1 show control DNA: and lanes 3-10 show DNA from patients 1 5. 1 1.1 3. 32 (AML. -ve control), 26. 14. 19 and 20 respectively. All patient and control DNA lanes show 1 0 p g DNA digested to completion with both Hind111 and BamHI restriction eniymes. Fragment sizes are indicated in kilobases.

5 18 Gerald Martin and Emer Lawlor Table 1. Rearrangement patterns of 2 5 precursor B ALL, and six T ALL patients

directly rather than emitting /?-particles to expose a n X-ray film indirectly via intensifying screens (Leary et al, 1983). Furthermore, since most of the labelled probe in the hybridi- zation solution remains unbound to the target DNA and

Patient Phen Enzymes J1,config Enzymes conFig retains its activity, the hybridization solution can be poured off and stored at - 20°C for re-use.

TCR/i

1 2 3 4 5 6 7 8 9

10 1 1 12 13 14 1 5 I 6 17 18 1 9 20 21 22 23 24 25

26 27

29 3 0 31

32

28

cALL E. H 3R cALL H 1G. 1R cALL E. H 2R cALL E. H. H/B 1G. 1R PreBALL E. H 2R cALL E.H/B 2R cALL E,H 1R. 1D cALL H. H/B 1G. 1R cALL E.H 1G. 1R CALL* E. H 2R CALL' E. H. H/B 2R cALLt H, H/B 1G. 1R cALL H, H/B 1R. I D cALLt H,H/B 1G. 1R PreBALL H.H/aB 2R cALL H 3R nALL E. H 2R nALL E. H 2R Pre B ALL E. H. H/B 1C. 2R nALL E.H. H/B 3R CALL H,H/B 2R cALL E, H 1R. I D nALL E.H 2R CALL' E. H 2R cALL E. H 2R

TALL H.H/B 1C TALL E.H 1 G TALL E 1c TALL E 1 G TALL E. H 1G T ALL E. H 1 G

AML H/B 1 G

E, H 2G H 2G

nd nd

E 2G. 1R nd

H 2G H 2G E IG, 2R H 2G E. H. H/B 2G. 1R

nd H/B 2G. 1R H. H/B 2G H/B 2R

nd nd

E 2G. 1R H. H/B 2R H. H/B 2G H. H/B 2G. 1R E 2R E. H 2R E 2G, 1R E.H 2G(1D)

H. H/B 2R E.H. B 1G. 2R E.H. B 2G. 1R E 2R E,H. B 1G. 2R E. H 2G. 1R

H/B 2G

Phen: phenotype. Enzymes used: E=Eco RI. H=Hin dIII. B=Bam HI, H/B=HindIII and BamHl double digest. JHconfig/ TCRBconfig: rearrangement pattern at Ig Jt1I and TCRP gene. G = germline. R = rearrnaged. D =deleted. Values refer to the number of bands present. = C p not done. t = ALL blast crisis of Ph+ CML. nd=not determined.

DISCUSSION

We have used biotinylated 11, and TCR/? DNA probes to examine rearrangements of the immunoglobulin and T cell receptor genes in precursor B cell ALL and T cell ALL patients. We found a range of rearrangement patterns compatible with both lineage specific and cross-lineage Ig and TCRB gene rearrangement. The pattern of rearrangements appears to be similar to that shown by other authors who have used radiolabelled probes. The biotin system, however. has several advantages in terms of routine use. The probes and detection chemicals are stable for a t least 12 and 6 months respectively at 4°C and the staining reaction requires only 4-16 h to produce adequate clarity. In addition, band resolution is improved since the hybridized probe DNA itself is stained

Initial problems were encountered in setting up this system. Non-specific probe hybridization was eliminated once the optimal blocking, hybridization and stringency wash conditions in the materials and methods were determined empirically. We found that the blocking times and tempera- tures described in the Material and Methods section also reduced non-specific and background staining. Finally. back- ground staining, both in the form of mild stain precipitation covering the entire membrane, and background spots pre- sumably formed by the aggregation of NBT and BCIP around contaminants in the staining solution, can be reduced further or completely eliminated by passing the staining solution through a 0.2 pn filter immediately prior to use. Once staining is complete and the membranes have been rinsed with water, they can be photographed or photocopied and stored in the dark to prevent fading of the stained bands.

Southern blot analysis using radiolabelled DNA probes for the detection of minimal residual disease has been shown to be more successful than morphology in indicating persistence of leukaemia. with a sensitivity of 1-5x (Zehnbauer et nl. 1986: Wright et al, 1987: Katz et al, 1989). Our results show that the biotin/streptavidin method has a comparable sensiti- vity. and can detect leukaemic cells when they comprise 1- 2% of the total mononuclear cell population.

ACKNOWLEDGMENTS

We thank Professor S h a m McCann, Professor Ian Temperley and Dr Fred Jackson for patient samples, Aine McGirl for immunological phenotyping. and Linda Hansen (Oncor Technical Services, Gaithersburg), Professor O'D. McGee and David Flannery (John Radcliffe Hospital, Oxford) for technical assistance. This work was supported by grants from the Health Research Board and the Children's Leukaemia Research Project of Ireland.

REFERENCES

Arnold, A.. Cosman. 1.. Bakhshi. A.. Jaffe. E.S.. Waldmann, T.A. 81 Korsmeyer. S.J. (1 983) Immunoglobulin gene rearrangements as unique clonal markers in human lymphoid neoplasms. New England journal of Medicine. 309, 1593-1 599.

Bennett. J.M.. Catovsky. D.. Daniel. M-T.. Flandrin. G.. Galton. D.A.. Gralnick. H.R. & Sultan. C. (1976) Proposals for the classification of the acute leukaemias. British journal of Haernatologg. 33 , 4 5 I - 458.

Brigati, DJ., Myerson. D., Leary, J.J.. Spalholz, B.. Travis, S.Z.. Fong, C.F.Y.. Hsiung, G.D. & Ward, D.C. (1983) Detection of viral genomes in cultured cells and paraffin-embedded tissue sections using biotin-labelled hybridization probes. Virology, 126, 32-50.

British Society of Haematology (1991) Recommendations for the application of molecular genetic and cytogenetic techniques in Haematology Departments in the U.K.

Chan. V.T.-W.. Fleming, K.A. & McGee. J. O'D. (1985) Detection of sub-picogram quantities of specific DNA sequences on blot hybridi-

Non-isotopic Detection of Gene Rearrangements 5 19 Leary. J .J . . Brigati. D.J. & Ward, D.C. (1983) Rapid and sensitive

colorimetric method for visualizing biotin-labelled DNA probes hybridized to DNA or RNA immobilized on nitrocellulose: Bio- blots. Proceedings of the National Academy oJ Sciences of the United States oJAmerica. 80, 4045-4049.

Lee. S.K.. Singh. J . & Taylor. R.B. (19751 Subclasses of T-cells with different sensitivities to cytotoxic antibody in the presence of anesthetic. European Journal oJ Immunology, 5 , 2 59-262.

Miller, S.A.. Dykes. D.D. & Polesky. H.F. (1 988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acid Research. 16, 121 5.

Reed. K.C. & Mann, D.A. (1 98 5) Rapid transfer of DNA from agarose gels to nylon membranes. Nucleic Acid Research. 13. 7207-722 1.

Worwood. M. & Wagstaff, M. (1990) Molecular biology and leukaemia diagnosis. Bailliere's Clinical Haematology, 3, 949-976.

Wright, J . J . . Poplack. D.G., Bakhshi. A.. Reaman. G.. Cole, D.. Jensen. J.P. & Korsmeyer. S. (1987) Gene rearrangements as markers of clonal variation and minimal residual disease in acute lympho- blastic leukaemia. /ournu1 oJ Clinical Oncology. 5 , 7 3 5-74 1 .

Zehnbauer. B.A.. Pardoll. D.M.. Burke, P.J.. Graham. M.L. & Vogelstein. B. ( 1986) Immunoglobulin gene rearrangements in remission bone marrow specimens from patients with acute lymphoblastic leukemia. Blood. 67, 8 35-838.

zation with biotinylated probes. Nucleic Acid Research. 13, 8083- 8091.

Dykes. D.D. ( 198X) The use of biotinylated DNA probes in parentage testing: non-isotopic labelling and non-toxic extraction. Electro- phoresis, 9. 359-368.

Felix, C.A.. Poplack. D.G.. Reaman. G.H.. Steinberg. S.M.. Cole, D.E.. Taylor, B.J.. Begley. G . & Kirsch. I.R. ( 1 990) Characterization of immunoglobulin and T-cell receptor gene patterns in B-cell precursor acute lymphoblastic leukaemia of childhood. Journal of Clinical Oncology. 8, 431-442.

Felix, C.A.. Reaman. G.H., Korsmeyer. S.J.. Hollis, G.F.. Dinndorf. P.A.. Wright, 1.1. & Kirsch, I.R. ( 1 987) Immunoglobulin and T cell receptor gene configuration in acute lymphoblastic leukemia in infancy. Blood. 70, 536-541.

Gebeyehu. G . . Rao. P.Y., Soolham. P.. Simrns. D.A. & Kleven. L. ( 1989) Novel biotinylated nucleotides-analogs for labelling and colorimetric detection of DNA. Nucleic Acid Research. 1 5 , 45 1 3- 45 34.

Janossy. C. & Campana. 1). ( 1 989) Monoclonal antibodies in the diagnosis of acute leukaemia. The 1,eukaemic Cell (ed. by D. Catovsky). pp. 1-20. Churchill Livingstone. Edinburgh.

Katz. F.. Ball, L.. Gibbons. B. & Chessells. 1. ( I 989) The use of DNA probes to monitor minimal residual disease in childhood acute lyrnphoblastic leukaemia. British Journal oJHuematology. 73, 17 3- 180.