British Journal of HuematologH, 1995, 90, 8-14

Modulation of cell kinetics and cell cycle status by treating CD34+ chronic myeloid leukaemia cells with p5 3 antisense phosphorothioate oligonucleotides

FRANCESCO LANZA,1'3 SUCAI BI,2'3 SABRINA MORETTI,' GIANLUIGI CASTOLDI' AND JOHN M. GOLDMAN3

'Institute of Haematology, University of Ferrara, Italy, 2Department of Molecular Pharmacology, St Jude Children's Research Hospital, Memphis, Tennessee, U.S.A., and 3LRF Centre for Adult Leukaemia, Department of Haematology, Royal Postgraduate Medical School, Hammersmith Hospital, London, U.K.

Received 19 October 1994; accepted for publication 24 January 1995

Summary. Mutations of the p53 tumour suppressor gene occur in 20% of chronic myeloid leukaemia (CML) patients in blastic crisis, but it is still uncertain whether this inactivation plays a role in the pathogenesis of blastic transformation or in maintaining the leukaemic prolifera- tion in CML, as it does in several solid tumours. We have previously shown that more than 50% of both normal and CML CD34+ cells express the p53 protein. However, haemopoietic cells at different phases of the cell cycle express p5 3 with different conformations, suggesting that the function of p53 may be closely regulated during the cell cycle. In order to elucidate the mechanism by which p53 suppresses cell proliferation, we evaluated the effects of inhibiting p53 expression on cell cycle and cell kinetics of chronic phase CML (n = 12) and normal (n = 7) bone marrow light-density cells and purified CD34+ progenitors by using an 18-mer modified antisense oligonucleotide which targets the region covering the six base pairs immediately before the first codon and the first four coding codons of p53. We found that the number of cells positive for the cell cycle-specific nuclear antigen Ki67 and for the BrdU monoclonal antibody (McAb) was significantly increased

after p53 antisense oligonucleotide treatment. At the same time, p53 protein expression was completely abrogated in both light-density and CD34' cells. In addition, DNA analysis by flow cytometry demonstrated that the number of cells in quiescent phases of the cell cycle (GO-G1) was significantly decreased after exposure of light-density cells to p53 antisense oligomers, whereas the number of cells in S or G2-M phases was increased. Furthermore, the longer the incubation time the higher the increase in cell proliferation. Treatment of CML cells with p53 antisense oligomers also resulted in significantly increased numbers of CFU-GM colonies. Our data suggest that p53 is a negative regulator of cell proliferation and its action is mediated through changes in cell cycle kinetics, mainly before the S phase. We can further speculate that the loss of p53 function, at the time of blastic crisis of CML, may play a role, in combination with other genetic changes (p210 BCRIABL, Rb gene abnorm- ality, others to be defined), in inducing disturbances in cell proliferation, differentiation, and apoptosis.

Keywords: p53, chronic myeloid leukaemia, blast crisis, antisense oligonucleotides, cell kinetics, cell cycle.

The p53 protein, a 393 aminoacid nuclear phosphoprotein, is the product of a 20 kb gene located on the short arm of chromosome 17. Wild-type p53 has been defined as a tumour suppressor gene because of its ability to inhibit transformation (Finlay et al, 1989; Levine et al, 1991; Harris & Hollstein, 1993). It also plays an important role in cell differentiation, proliferation, and apoptosis (Vogelstein & Kinzler. 1992; Yonish-Rouach et al, 1991; Shaulsky et al.

Correspondence: Dr Francesco Lama. Institute of Haematology. St Anna Hospital, Corso Giovecca n. 203,I-44100, Ferrara. Italy.

1991; Lane, 1993, Williams, 1991). Although the precise role played by p53 in cell cycle control is not yet well understood, it is thought to affect the GO to G 1 transition and the entry into S-phase (Danova et al, 1990; Yin et al, 1YY2; Zhan et aJ, 1994; Lowe et aI, 1993). The introduction of p53 antibodies blocks the entry of cells into S-phase with a consequent accumulation of cells in the G 1 phase of the cell cycle. Furthermore, p53 has been found to be hypo- phosphorylated in GO-G1, and hyperphosphorylated in S-G2-M phases (Levine et al, 1991).

More recently, the action of p53 mediating G 1 arrest has

8 0 1995 Blackwell Science Ltd

Treatment of CMI, Cells with p53 Antisense Oligonucleotides 9 been shown to involve induction of expression of cyclin- dependent kinase (Cdk) inhibitor p2 1 WAF/CIPl (El-Deiry et al. 1993; Hunter, 1993). p21 WAF/CPl is a strong inhibitor of Cdks in vitro (Harper et al. 1993). Over- expression of WAF/CIPl leads to inhibition of cell growth. All these observations support the notion that wild-type p53 has a strong influence on cell proliferation, probably via its effects on the cell cycle (Hunter, 1993: Zhu et al, 1993).

The p53 protein is present in minute concentrations in haemopoietic mature cells such as neutrophils, lymphocytes and neutrophils, whereas CD34- bone marrow precursors do not contain p53 (Kastan et al. 1991; Lanza et al, 1992). However, a subset of CD34’ progenitor cells and activated T lymphocytes express detectable level of p53 protein (Lanza et al. 1992: Rivas et al, 1992). In contrast, cells from several types of leukaemias contain large amounts of the p53 protein, which is, however, often in a mutant or truncated form (Kastan et d, 1991; Bi et al, 1992. 1994a; Morkve 81 Laerum. 1991).

Mutations of the p53 gene have been found in several types of solid and haematological tumours, and are the commonest genetic alterations in human malignancies (Harris & Hollstein, 1993). 15-25% of patients with chronic myeloid leukaemia (CML) in blastic crisis show p53 gene alterations, and it has been suggested that these abnormalities may be responsible for the progression of CML into the blastic stage (Sugimoto et al, 1992; Mashal et al, 1990: Kelman et al, 1989; Neubauer et al, 1993; Feinstein et al, 1991; Gaidano et al, 1994).

The function of p53 can be inactivated by gene deletion, chromosome deletion, loss of expression or by point mutation which changes the wild-type protein into a mutant protein with a much longer half-life (Levine et al, 1991). The majority of p53 alterations occur between exons 4 and 8 (‘hot regions’) and within these regions there are four locations where mutations are found with the highest frequency (‘hot spots’). The most frequent mechanism of p53 inactivation in patients with CML in blastic crisis is complete deletion of one allele in association with a point mutation in the remaining allele (Feinstein et al, 1991).

By using an antisense oligonucleotide approach, we have previously demonstrated that the mutant or truncated p53 proteins expressed in blast cells from many CML cell lines have no growth-promoting effect and are not required for cell survival, proliferation and colony production (Bi et al, 1993). As a consequence, we speculated that the loss of the tumour- suppressor function of p53 might be the only mechanism by which p53 could be involved in the transition from chronic phase to blast crisis CML. In two subsequent studies we have showed that treatment of chronic phase CML cells with p53 antisense oligonucleotides resulted in a significantly increased formation of CFU-GM colonies in 80% of the cases, and therefore we hypothesized that the expression of the wild-type p53 is involved in the regulation of the proliferation pathway of both normal and CML haemopoietic cells (Lanza et al, 1992; Bi et al, 1994a). In these studies we also demonstrated that 50% of CD34’ cells express levels of p53 detectable by flow cytometry.

In order to evaluate further the involvement of p53 in regulating haemopoietic maturation and proliferation and its mechanisms of action, we studied the expression of the cell cycle-specific nuclear antigen Ki6 7 and other cell kinetic parameters in CML and normal bone marrow cells treated with a 18-mer modified antisense oligonucleotide.

MATERIALS AND METHODS

Cells. Bone marrow aspirates from 12 patients with Ph (Philadelphia) chromosome positive CML at chronic phase, aged 29-58 years (mean 47 years), and seven healthy subjects, aged 26-52 years (mean 43 years), were used in this study. Each subject gave informed consent in accor- dance with local Ethics Committee requirements. All the CML patients but one were studied at the onset of their disease and were previously untreated by chemotherapy or irradiation. Only one chronic phase Ph+ CML patient was treated with hydroxyurea before our study. At the time of the investigation his WBC count was 39 x 109/l. The average WBC of the remaining 11 CML patients was 68 x 109/1 (range 35-102 x 109/l). Light-density mononuclear cells were prepared by centrifuging bone marrow cells over a layer of Percoll of 1.067 g/ml density. The isolated light- density cells were washed twice in PBS before being resuspended in SFM with or without p53 antisense oligonucleotides and then analysed for p5 3 expression, DNA content, and Ki67 antibody estimation. The cells were purified by negative depletion of non CD34’ cells with a mixture of monoclonal antibodies (including CD2, CD19, CD16, CD15, CD41, CD61, glycophorin A) and Dynabeads (Dynal) (Lanza & Castoldi, 1994). Purity was >51% in all cases (CML: range 51-72%. average 54%: normal bone marrow: range 3-82%, average 69%). Within the mono- nuclear cell window, as evaluated by flow cytometry, the purity was >90% in all samples. A negative selection strategy was preferred to a positive one, because in a series of preliminary experiments we noted that the former technique gave rise to less interference with p53 expression and pattern of CD34 growth as compared with that observed following a positive CD34 selection (unpublished data).

Anti-sense oligonucleotides. The 1 8-mer oligonucleotides were synthesized on a DNA synthesizer (Applied Biosystems 380B DNA synthesizer, Foster City, Calif., U.S.A.) by an automated process. The antisense sequences target the region covering the six base pairs immediately before the first codon and the first four coding codons of p53. In order to increase the nuclease resistance of the oligonucleotides, the 18-mer oligonucleotides were made up of 12-mer unmodi- fied sequences flanked by 3-mer phosphorothioate linkages at both ends. The synthesized oligonucleotides were ethanol precipitated followed by multiple washings in 70% ethanol (more than five t i e s ) for purification. Purified oligonucleo- tides were then dissolved in Wh4I 1640 medium (Flow laboratories). The sequences of the three oligonucleotides were as follows: 5’-CGG CTC CTC CAT GGC AGT-3’ (antisense): 5’-ACT GCC ATG GAG GAG CCG-3’ (sense): and 5’-AGT GGC CTC CAT CTC CTC CGG-3‘ (scrambled antisense). The affinity for the target sequences and the

0 1995 Blackwell Science Ltd, British Journal of Huemutology 90: 8-14

10 Pruncesco Lunza et al effectiveness of inhibition of p5 3 translation by anti-sense oligonucleotides have been reported (Bi et ul, 1993).

Serum-free medium and cell growth analysis. Serum-free medium (SFM) was used for incubation of haemopoietic cells with the oligonucleotides. SFM was prepared accord- ing to the method of Salem. The basic ingredients were: dialysed bovine serum albumin (BSA), cholesterol, insulin, 2-mercaptoethanol, transferrin (Sigma) mixed in Dulbecco's modified Eagle's minimum essential medium (Gibco). The haemopoietic cells were incubated with oligonucleotides at a concentration of 5 p ~ at 37°C for 24-48h. The optimal concentration for p53 antisense ologonucleotides has been tested and described elsewhere (Bi et al, 1994a).

Colonu assag. CFU-GM colony assay was performed according to a previously established method (Lanza et al. 1993). In brief, 1 x lo5 light-density cells were pre- incubated with ~ F M oligonucleotides at 37°C for 12 h and then plated in a standard CFU-GM medium (containing 1% methylcellulose, 30% fetal calf serum, and supernatant from the 5637 bladder carcinoma cell line as a source of growth factors) supplemented with p53 phosphorothioate oligonu- cleotides at a concentration of 5 p ~ . Cultures were incubated for 14 d at 3 7°C in 5% C02 in the air. Colonies composed of more than 50 cells were identified and scored using an inverted microscope. All experiments were done in triplicate.

Flow cytornetrg analysis. Bone marrow cells were analysed with a Facscan flow cytometer equipped with a single argon ion laser (Becton Dickinson, Mountain View, Calif., U.S.A.). The instrument was calibrated with fluorescein isothiocya- nate (FITC) and R-phycoerythrin (PE) beads provided by Becton Dickinson. Data were analysed with appropriate negative (isotype-matched non relevant McAbs) controls using Lysis 2 and Paint-a-Gate Research softwares. 10 000 cells were analysed for each sample. Human AB serum was added prior to antibody incubation to avoid non-specific binding of McAbs to Fc receptors (Lanza et al, 1991). Antibody reactivity for the various markers was registered as a logarithmic fluorescence and evaluated on ungated cells. The CD34 determination was performed on the entire cell population, and the fluorescence expression evaluated on a cytogram generated by combining side scatter and fluores- cence 1 or 2 (as pertinent) parameters (Lanza & Castoldi, 1994). Cells showing fluorescence intensities above the upper limit of the negative control distribution were regarded as positive. In some cases the expression for p53 McAbs was very weak. In these cases we used the Kolmogorov-Smirnov (K-S) statistical test in order to quantify the difference between the two flow-cytometry histograms (negative versus sample). As previously outlined (Lanza et al, 1991, 1993), the K-S statistics offers the possibility to assess whether or not the two experimentally observed histograms are an example of different probability functions (D/s( n)). The D value is defined as the maximum absolute difference between the two curves analysed. The diversity or identity of flow cytometry histograms can therefore be evaluated. K-S statistics is provided by Becton Dickinson software.

p53 antibodies. Two McAbs. PAb 1801 and PAb 240 (Oncogene Science, Cambridge, U.K.) were used to detect

p53 expression. The PAb 1801 McAb recognizes an epitope near the NH2-terminal end of the p53 peptide and reacts preferentially with the human wild-type p53. The PAb 240 McAb was originally reported to recognize a common conformational epitope closely associated with mutations in the p53 gene, but recent findings suggest that it can also react with undenatured wild-type p53 protein in haemo- poietic cells, such as CD34' progenitors (Lanza et al, 1992) and activated lymphocytes (Rivas et al, 1992). For the detection of p53 expression by either flow cytometry or a confocal microscope, we used the PLP (paraformaldehyde- lysine periodate) fixation reagent, according to a published method (Bi et ul, 1994a). Double-staining experiments using p53 and CD34 (8G12-FITC or PE, Becton Dickinson) McAbs were performed in order to evaluate the selective effects of inhibiting p53 expression on the CD34' cell population by using p5 3 phosphorothioate oligonucleotides. A detailed description of this dual-colour flow-cytometry analysis is reported elsewhere (Bi et al, 1994a).

Cell cycle analysis. The cellular DNA content was assessed using a flow cytometer (Becton Dickinson, Mountain View, Calif.). The cell cycle distribution data were provided by the Rectangle-Fit (R-fit) mathematical algorithm of Facscan/ Cellfit software program. A double-discrimination module (DDM) was used in order to detect doublets and higher cell aggregates. To determine DNA content, cells were treated with 0.5% Triton X-100 for 3 min and then stained with a propidium iodide solution (50 pg/ml). RNAase (Sigma) at a final concentration of 0.1% was added to the cell prepara- tion. Bromodeoxyuridine (BrdU) incorporation and a monoclonal antibody against BrdU (Becton Dickinson) in conjunction with flow cytometry were used to calculate the proportion of cells in S-phase of the cell cycle. The manufacturer's recommendations were followed for label- ling cells and detection.

Statistical analysis. The significance of differences in the absolute numbers of cells stained for Ki67 and BrdU McAbs before and after treatment with p53 oligonucleotides was evaluated using the paired Student t test, or, when appropriate, by the non-parametric Wilcoxon test.

RESULTS

Chronic myeloid leukaemia Purified CD34' cell population. Table I shows the effects of

p5 3 sense and antisense phosphorothioate oligonucleotides on kinetic characteristics of CML bone marrow CD34' cells, as assessed by Ki67 nuclear antigen estimation and DNA content analysis by flow cytometry. In brief, treatment of CML cells with p53 antisense oligonucleotides abrogated p53 expression by CD34+ cells, accompanied by a significant increase in the number of cells positive for Ki67 (P < 0.001) and BrdU (P < 0.01) McAbs (Fig 1). In contrast, CML CD34' cells (which were purified using a depletion technique to remove non 0 3 4 ' cells) exposed to p53 oligonucleotides in the sense configuration or incubated with a medium containing no oligomers did not show any significant increase in Ki67 and BrdU positivity. DNA analysis by flow cytometry demonstrated that the number of CD34' cells in

0 1995 Blackwell Science Ltd, British Journal ojHaernatology 90: 8-14

Treutment of CML Cells with p53 Antisense Oligonucleotides 11 Table I. Effects of p53 sense and antisense oligonucleotides on kinetic characteristics of CD34' cells derived from 12 CML patients. Data are expressed as mean percentage of positivity (f SD) of 12 experiments done in duplicate.

Baseline values

24 h incubation 48 h incubation

AS-treated

CD34' 54.1 f 10.1 p5 3-Ab2 34.5 + 6.4 p53-Ab3 30.9 f 5 . 3 Ki6 7+ 5.6 + 2.1 BrdU incorporation 3-3 f 1.2 GO/Gl (%) (DNA analysis) 92.9 f 3-1 S/G2/M (DNA analysis) 7-1 f 2.0

50.9 f 9.5 2.6 f 1.1*

10.9 i 2,2* 20.5 f 3Q 10.2 f 2.5* 80.1 f 3.4* 19.9 f 3.2*

Sense-treated ____ ~

52.5 f 9.4 31.2 * 2 5 21.8 f 3.7 11.8 i 3.1 6.9 f 2.2

88.6 f 3.0 11.4 f 1.8

Controls ~ .. ~

54.3 f 8.9 34.0 f 2.9 19.3 zk 2.9 9.9 & 2.5 6.0 f 2.3

90.4 f 2.7 9.6 f 2.0

AS-treated Sense-treated Controls

53.4 f 8.1' 2.0 f 2.0'

12.1 f 1,6* 22.2 f 2.4* 12.3 f 1.4* 77.3 f 3.8* 22.7 f 2.0*

51.9 f 9.2 32.2 f 2.8 29.7 f 4.1 12.3 * 3.0 7.5 f 2.5

90.9 f 3-1 9.1 f 2.9

55.2 i 10.8 32.9 f 4.1 30.4 f 3 . 3 12.2 f 2.7 7.4 f 2.2

91.2 f 3.4 8.8 f 2.2

*Statistically significant difference (see text for further details).

quiescent phases of the cell cycle (GO-G1) was significantly decreased (P < 0.01) after treatment of CD34' cells with p5 3 antisense oligonucleotides, whereas an increase in the number of cells in either S or G2-M phases of the cell cycle (P < 0.02) was observed following treatment with the antisense oligomers. In addition, after treatment with antisense oligomers the increased number of cells positive for Ki67 and BrdU McAbs was inversely correlated with the extent of p53 protein expression, consistent with the concept that the stronger the p53 protein inhibition the higher the cell proliferation activity of CML cells. The minimum period

and BrdU positivity and inhibit p53 protein expression was 6 h. However, exposure of cells for periods longer than h h resulted in more effective p53 inhibition and a higher Kih7 and BrdU cell positivity, showing the greatest changes following 48 h incubation time.

Detailed analysis of the changes in p53 expression after treatment with antisense oligonucleotides showed that the number of cells positive for the immunological subclass recognized by the PAb 1801 McAb was significant decreased (P < 0.01), whereas the reduction in the number of cells positive for the PAb 240 McAb was less evident (P < 0.05)

of antisense oligomer incubation necessary to increase Ki6 7 when compared

'1 24;HOURS INCUBRTION

'1 24yHOURS I NCUBRT I ON

'1 24;HOURS INCUBATION

with that of cells treated with sense

Fig 1. Flow cytometry measurement of the cell-cycle-specific nuclear antigen Ki6 7 before (a) and alter treatment of CML cells with p53 oligonucleotides in either the sense (c) or antisense (d) configuration. Ki67 fluorescence expression on cells cultured in the absence of oligonucleotides is showed in panel (b). Representative example of a CML patient in chronic phase. Percentage of cell positivity is 3.8% (before incubation), 7.5% (no oligonucleotide), 10.5% (sense-treated cells), and 20.1% (antisense-treated cells). (1) Isotypic control; (2) sample. Cells showing fluorescence intensities above the upper limit of the negative (isotype-matched non-relevant McAb) control distribution were considered Ki67 positive.

0 1995 Blackwell Science Ltd, British Journal of Huematology 9 0 8-14

12 Francesco Lanza et a1 Table II. Changes in cell cycle and cell kinetics parameters before and after the in vitro treatment of CD34+ cells (obtained from the bone marrow of seven healthy subjects) with p53 antisense oligonucleotides. Data are expressed as mean percentage f SD. All the experiments were done in duplicate.

24 h incubation 48 h incubation

Baseline values AS-treated Sense-treated Controls AS-treated Sense-treated Controls

CD34' 69.7 f 8.9 66.9 f 7.7 67.5 f 8.2 68.5 f 6.9 63.4 f 8.1 66.7 f 6.7 65.8 f 7.4 p53-Ab2 48.6 f 3.9 3.9 f 1.8* 39.5 f 4.4 41.2 f 5.1 2.3 f 1.6* 36.2 f 3.9 38.4 f 5 .3 p53-Ab3 43.2 f 5.0 12.7 f 3.0: 36.3 f 5.2 35.8 f 4 .4 16.4 f 2.9: 35.5 f 4.6 34.6 f 4.9 Ki67' 4.8 f 2.1 14.6 f 3.2* 9.8 f 4.7 9.5 f 4.3 17.1 f 3.3* 12.1 f 3.7 11.8 f 2.0 BrdU incorporation 1.9 f 1.0 6.3 f 1.3* 4.5 f 1.8 4.9 f 1.9 7.0 f 1.7* 5.5 f 2.1 5.0 f 1.9 GO/G1 (%) (DNA analysis) 95.0 f 4.3 88.6 f 4.7* 90.8 f 5.0 91.1 f 4.6 85.7 f 3.5* 88.8 f 4.4 89.2 f 4.7 S/G2/M (DNA analysis) 5.0 f 2.2 11.4 f 2.2: 9.2 f 2.9 8.9 f 2.5 14.3 f 2.9* 11.2 f 3.4 10.8 f 3.1

* Statistically significant difference (see text for further details).

oligomers or without oligonucleotide. The percentage of cells positive for the CD34 progenitor-associated cell antigen remained stable throughout the entire in vitro incubation period.

Furthermore, exposure of CML cells to p53 antisense oligonucleotides caused a marked increase (P < 0.01) in the number of CFU-GM colonies, as described elsewhere. In addition to the changes in CFU-GM colony numbers, we noted that the colonies formed by antisense oligonucleotide- treated cells were larger than those formed by either sense oligonucleotide-treated cells or by cells that had received no oligonucleotides. Furthermore, the colonies formed by the sense oligonucleotide-treated cells were smaller than those formed by cells that had received no oligonucleotides (Bi et al, 1994b).

In three CML patients we evaluated Ki67 and p53 antibody positivity on a highly enriched CD34+ cell population (mean purity 94.4%, range 92*1-98*0%). We noted a 7-fold mean increase in the percentage of Ki67+ cells after treatment with antisense oligonucleotides (mean value before treatment 4.2%; antisense-treated cells 32.4%: sense- treated cells 9.7%; controls 8.2%). However, the small number of CD34' cells did not allow us to perform further investigations (DNA analysis, BrdU incorporation, and p5 3 expression).

Normal bone marrow. Inhibition of p53 expression by antisense p53 oligomers was associated with a significant increase in the number of cells positive for Ki67 (P < 0.01) and BrdU (P < 0.03) McAbs (Table 11). In addition, DNA analysis by flow cytometry showed that the number of cells in quiescent phases (GO-G1) was significantly decreased (Pc0 .05) after treatment of CD34+ cells with p53 antisense oligonucleotides, whereas an increase in the number of cells in either S or G2-M ( P < 0.05) was observed following treatment with antisense oligonucleotides.

Light density cells. The pattern of positivity for Ki67, BrdU, p53 and CD34 McAbs following exposure of CML and normal light-density cells (<1*068 g/ml) to p53 antisense phosphorothioate oligonucleotides was very similar to that shown in Table I (data not reported). In fact, the exposure of

light-density cells to antisense p53 oligomers resulted in a 2-fold increase in the numbers of cells positive for Ki67 and on average 60% augmentation in anti-BrdU antibody positivity.

DISCUSSION

Mutations of the p53 gene are the commonest genetic alterations in human malignancies. 15-2 5% of the patients with CML in blastic crisis show p53 gene aberrations, which are usually associated with abnormalities of the short arm of chromosome 17, which is the location of the p53 gene. However, whether the mutant p53 protein, which is usually expressed at an elevated level, piays any role in the pathogenesis of blastic transformation of CML patients, as it does in several solid tumours, is still debated (Harris & Holstein, 1993; Feinsteii et al, 1991; Kirkland et al, 1994).

In a previous study, using an antisense oligonucleotide strategy, we demonstrated that inhibition of p5 3 expression by antisense phosphorothioate oligomers had no effect on cell proliferation, cell viability and colony production formed by four CML cell lines that exhibit abnormalities in p53 gene configuration and in protein expression (Bi et al, 1993). We therefore speculated that the loss of the tumour suppressor function of p53 might be the only mechanism by which p53 is involved in the regulation of maturation and proliferation of CML cells.

In two subsequent investigations (Lanza et al, 1992; Bi et al, 1994a). we have first demonstrated that CD34' haemopoietic cells from both CML and normal bone marrow express the p53 protein in either the immunological sub- class recognized by PAb 1801 McAb or in the mutant conformation-associated PAb240 McAb. Furthermore, treatment of chronic-phase CML cells with antisense phosphorothioate oligonucleotides resulted in a significant increase in the numbers of CFU-GM colonies, accompanied by a complete inhibition of p53 expression by CD34' cells. However, in the various CD34' cell subsets (namely CD34+/ HLA-DR'; CD34'/HLA-DRP) the expression of different p53 conformations, as defined by the two p53 McAbs, was

0 1995 Blackwell Science Ltd, British Journal of Haernatologg 90: 8-14

Treatment of CML Cells with p53 Antisense Oligonucleotides 13 related to the cell cycle position, thus supporting the concept that the conformational change of p53 protein may be closely associated with its functional stage (Bi et al, 1994a).

More recently it has been shown, by using retroviral gene transfer, that the introduction of a mutant human p53 cDNA into haemopoietic progenitor cells from CML patients in chronic phase (which already contain p2 10BCR/ABL) could promote growth factor independent cell proliferation in vitro, but was unable to generate continuous cell lines, allowing us to postulate that additional genetic aberrations, other than that involving p53 gene, are required for the transformation of CML into the acute, rapidly fatal, stage of the disease (Bi et al, 1994b).

In order to elucidate the mechanism by which p53 suppresses proliferation of CML and normal bone marrow cells, we studied changes in cell cycle status and cell kinetics before and after treatment with p53 antisense sequences for incubation times ranging from 6 to 48 h. The treatment of CML cells with p53 oligonucleotides in the antisense configuration abrogated p53 expression by light-density cells and purified CD34+ cells, accompanied by a significant increase in the number of cells positive for the cell-cycle specific Ki6 7 and bromodeoxyuridine (BrdU) McAbs. In addition, DNA analysis by flow cytometry demonstrated that the number of cells in quiescent phases of the cell cycle (GO-G1) was significantly decreased, and an increase in the number of cells in either S or G2-M phases of the cell cycle was observed. Furthermore, the longer the incubation time the greater the increase in cell proliferation and the stronger the inhibition of p53 expression. Therefore we have now evidence that p53 is directly involved in the regulation of proliferation of CML and normal progenitor cells.

As previously demonstrated, treatment of CML cells with p53 antisense oligomers also resulted in significantly increased numbers of CFU-GM colonies, thus confirming the inhibitory role played by wild-type p53 on cell proliferation activity and in vitro cell growth of bone marrow cells (Bi et al, 1994a). Some years ago, Hatzfeld et aZ(1991) reported that antisense oligonucleotides against p53 had no effect on colony formation by haemopoietic cells derived Gom the bone marrow of healthy subjects: in contrast, we did show an effect on normal progenitor cells but much less than that seen with CML cells. The finding that inhibiting p53 expression by p53 antisense oligo- nucleotides increased CFU-GM colony production in the majority of CML samples in consistent with the observation that inhibiting expression of the retinoblastoma gene, another tumour suppressor gene and cell cycle regulator, could release early haemopoietic cells from quiescence and stimulate fibroblast cell division and focus formation (Strauss et al, 1992). Since p53 can act as a mediator of apoptosis and a cell cycle regulator, its expression in progenitor cells could be critical in this context. As a consequence, the suppression of 'p53 function could lead to significant changes in the process of apoptosis and in the regulation of the cell cycle machinery.

These data also seem to indicate that the conformational status of p53 protein is closely associated with its functional stage and is strictly regulated during the cell cycle (Zhu et al,

0 1995 Blackwell Science Ltd, British Journal of Haemutology 90: 8-14

1993), because the ratio of the two p53 McAbs was significantly different before and after incubation of CD34' cells with p53 antisense oligomers. In fact, following exposure to antisense oligomers, the number of cells showing positivity for the p5 3 immunological subclass recognized by the PAb 240 McAb (which is associated with active phases of the cell cycle) was significantly higher than that of the PAb 1801 McAb, which identifies a different part of the p53 molecule. We can therefore speculate that the conformational changes of p53 may reflect the functional and kinetics characteristics of the various CD34' cell subpopulations, making it also possible that not all p5 3-positive CD34' cells are capable of undergoing apoptosis. It can also be postulated that CD34+ cells expressing the wild-type p53. which are unreactive with PAb 240 McAb. will enter active cell cycle when inhibition of p53 function is achievable.

In conclusion, our data support the concept that p53 is a negative regulator of cell proliferation and its action is mediated through changes in cell cycle kinetics, mainly before the S phase. On the basis of these data, we can further speculate that in CML patients the loss of p53 function is not sufficient per se to induce blastic transformation, but acts synergistically with other genetic changes (p2 10 BCR/ABL, Rb gene abnormality, p16 deletion, others to be defined) (Szczylik e t al, 1991; Strauss et al, 1992; Daley et al, 1990). However, the disturbances in ceIl proliferation, maturation, and apoptosis secondary to p53 gene abnormality could represent one critical step in the development of the blastic stage in a substantial number of CML patients (Ferrari et al, 1994; Kirkland et al, 1994).

ACKNOWLEDGMENT

This work was supported by Regional Funds, MURST (40 and 60%), and AIRC (Italian Association for Cancer Research).

REFERENCES

Bi, S.. Barton. C.M., Lemoine. N.R.. Cross, N.C.P. & Goldman. J.M. (1 994b) Retroviral transduction of Philadelphia-positive chronic myeloid leukemia cells with a human mutant p53 cDNA and its effect on in vitro proliferation. Experimental Hematology. 22, 95-99.

Bi, S., Hughes, T., Bungey. J., Chase, A., de Fabritiis, P. & Goldman, J.M. (19 92) p5 3 in chronic myeloid leukemia cell lines. Leukemia,

Bi, S., Lanza, F. & Goldman, J.M. (1993) The abnormal p53 proteins expressed in CML cell lines are non-functional. Leukemia, 7,

Bi. S., Lanza, F. & Goldman, J.M. (1994a) The involvement of 'tumour suppressor' p53 in normal and chronic myelogenous leukemia hemopoiesis. Cancer Research, 54, 582-586.

Daley. G.Q.. Van Etten, R.A. & Baltimore, D. (1990) induction of chronic myelogenous leukemia in mice by the p210 BCR/ABL gene of the Philadelphia chromosome. Science, 247, 824-830.

Danova. M.. Giordano, M.. Mazzini. G. & Riccardi, A. (1990) Expression of p53 protein during the cell cycle measured by flow cytometry in human leukaemia. Leukaemia Research, 14,

6, 839-842.

1840- 1845.

41 7-422.

14 Francesco Lanza et al El-Deiry, W.S., Tokino, T., Velculesscu, V.E.. Levy, D.B.. Parsons, R.,

Trent. J.M., Lin, D., Mercer, W.E., Kinzler, K.W. & Vogelstein. B. (1993) WAF-1 a potential mediator of p53 tumour suppression. Cell, 75, 817-825.

Feinstein, E., Cimino, G.. Gale, R.P., Alimena, G., Berthier, R., Kishi, K., Goldman, J.M., Zaccaria, A., Berrebi, A. & Canaani. E. (1991) p53 in chronic myelogenous leukemia in acute phase. Proceedings of the National Academy of Sciences of the United States of America,

Ferrari. S., Manfredini, R. & Torelli, U. (1994) Antisense strategies in leukemia. Hematologica, 79, 107-110.

Finlay. C.A., Hinds, P.W. & Levine. A.J. (1989) The p53 proto- oncogene can act as a suppressor of transformation. Cell, 37,

Gaidano. G., Serra, A., Guerrasio, A., Rege-Cambrin, G.. Mazza, U. & Saglio, G. (1994) Genetic analysis of p53 and RB1 tumor- suppressor genes in blast crisis of chronic myeloid leukemia. Annals of Hematology, 68, 3-7.

Harper, J.W., A d d . G.R., Wei, N.. Keyomarsi, K. & Elledge, S.J. (1993) The p2 1 Cdk-interacting protein cip-1 is a potent inhibitor of G 1 cyclin-dependent kinases. Cell, 75, 805-816.

Harris, C.C. & Hollstein, M. (1993) Clinical implication of the p53 tumor-suppressor gene., New England Journal of Medicine, 329,

Hatzfeld. J., Li, M.L., Brown, E.L., Sookdeo, H., Levesque, J.P., O'Toole, T., Gurney, C., Clark, S.C. & Hatzfeld, A. (1991) Release of early human hemopoietic progenitors from quiescence by antisense transforming growth factor pl or RB oligonucleotides. Journal of Experimental Medicine, 174, 925-929.

88,6293-6297.

1083-1093.

1318-1 32 7.

Hunter, T. (1993) Breaking the cell cycle. Cell, 75, 839-841. Kastan, M.B., Radin, A.I., Kuerbitz. S.J., Onyekwere. 0.. Wolkow,

C.A., Civin, C.I., Stone, K.D., Woo, T., Ravindranath. Y. & Craig, R.W. (1991) Levels of p53 protein increase with maturation in human hematopoietic cells. Cancer Research, 51, 4279-4286.

Kelman, Z., Prokocimer, M., Peller, S., Khan, Y., Rechavi, G.. Manor, Y., Cohen. A. & Rotter, V. (1989) Rearrangements in the p53 gene in Philadelphia chromosome positive chronic myelogenous leukemia. Blood, 74, 2318-2324.

Kirkland, M.A., O'Brien. S.G. & Goldman, J.M. (1994) Antisense therapeutics in haematological malignancies. British Journal of Haematology. 8 7, 44 7-45 2.

Lane. D.P. (1993) A death in the life of p53. Nature. 362, 786-787. Lanza, F., Latorraca. A., Musto, P., Ferrari, L.. Moretti, S., Zabucchi,

G., Carotenuto. M. & Castoldi, G.L. (1991) Cytochemically unreactive neutrophils from subjects with myeloperoxidase deficiency show a complex pattern of immunoreactivity with anti-MPO monoclonal antibodies: a flow cytometric and irnmuno- cytochemical study. Annals of Hematology, 63, 94-100.

Lanza, F. & Castoldi, G.L. (1992) Complement receptor 1 (CR1) expression in chronic myeloid leukemia. Leukemia and Lymphoma.

Lanza, F., Bi, S., Castoldi, G.L. & Goldman, J.M. (1993) Abnormal expression of N-CAM (CD56) adhesion molecule on myeloid and progenitor cells from chronic myeloid leukemia. Leukemia, 7,

Lanza, F.. Bi, S. & Goldman, J.M. (1992) Similarity of p53 expression by CD34' cells in chronic myeloid leukemia and normal progenitors detected by flow cytometry. Blood, 80, 1857-1858.

Lanza, F. & Castoldi, G.L. (1994) Large-scale enrichment of CD34'

8, 35-41.

15 70-1 5 75.

cells by Percoll density gradients: a CML-based study design. Hematopoietic Stem Cells (ed. by E. Wunder, H. Sovalat, P. H. Henon and S. Serke), pp. 255-270. AlphaMed Press, Dayton.

Levine, A.J., Momand, J. & Finlay, C.A. (1991) The p53 turnour suppressor gene. Nature, 351, 453-455.

Lowe, S.W., Schmitt. E.M., Smith, S.W., Osborne, B.A. &Jacks, 1'. (1993) p53 is required for radiation-induced apoptosis in mouse thymocytes. Nature, 362, 847-852.

Mashal, R., Shtalrid, M., Talpaz, M., Kantarjian, H.. Smith, L.. Beran, M.. Cork, A., Trujillo. J., Gutterman, J. & Deisseroth, A. (1990) Rearrangement and expression of p53 gene in the chronic phase and blast crisis of chronic myelogenous leukemia. Blood, 75,180- 189.

Morkve, 0. & Laerum. O.D. (1991) Flow cytometric measurement of p5 3 protein expression and DNA content in paraffin-embedded tissue from bronchial carcinomas. Cytometry. 12, 438-444.

Neubauer, A.. He, M., Schmidt, C.A., Huhn, D. & Liu, E.T. (1993) Genetic alterations in the p53 gene in the blast crisis of chronic myelogenous leukemia: analysis by polymerase chain reaction- based techniques. Leukemia, 7, 593-600.

Rivas. C.I., Wisniewski, D., Strife, A., Perez, A., Lambek, C.. Bruno, S., Darzynkiewicz, S. & Clarkson. B. (1992) Constitutive expression of p53 protein in enriched normal human marrow blast cell populations. Blood, 79, 1982-1986.

Shaulsky, G., Goldfinger, N., Peled, A. & Rotter, V. (1991) Involvement of wild-type p53 in pre-B cell differentiation in vitro. Proceedings of the National Academy of Sciences of the United States of America, 88, 8982-8986.

Strauss, M.. Hering, S., Lieber, A,. Herrmann. G., Griffin, B.E. & Arnold, W. (1992) Stimulation of cell division and fibroblast focused formation by antisense repression of retinobiastoma protein synthesis. Oncogene, 7, 769-773.

Sugimoto. K.. Toyoshima, H., Sakai. R., Miyagawa. K., Hagivara, K.. Ishikawa, F.. Takaku, F.. Yazaki, Y. & Hirai, H. (1992) Frequent mutations in the p53 gene in human rnyeloid leukemia cell lines.

Szczyli. C., Skorski, T., Nicolaides, N.C. Manzella, L., Malaguamera, L., Venturelli, D., Gewirtz, A.M. & Calabretta, B. (1991) Selective inhibition of leukemia cell proliferation by BCR/BAL antisense oligodeoxynucleotides. Science, 253, 562-565.

Vogelstein, B. &Kinder, K.W. (1992) p53 function and dysfunction. Cell, 70, 532-526.

Williams, G.T. (1991) Programmed cell death: apoptosis and oncogenesis. Cell, 65, 1097-1098.

Yin, Y., Tainsky, M.A., Bischoff, P.Z., Strong, L.C. & Wahl, G.M. (1992) Wild-type p53 restores cell cycle control and inhibits gene amplification in cells with mutant p53 alleles. Cell, 70, 937-948.

Yonish-Rouach, E., Resnitzky, D., Lotem, J., Sachs. L., Kimchi, A. & Oren, M. (1991) Wild-type p53 induces apoptosis of rnyeloid leukaemic cells that is inhibited by interleukin-6. Nature. 352, 345-347.

Zhan, Q.. Bae, I., Kastan, M.B. & Fornace, A.J. (1994) The p53- dependent y-ray response of GADD45. Cancer Research. 54,

Zhu. Y.M., Bradbury, D. & Russell, N. (1993) Expression of different conformations of p53 in the blast cells of acute myeloblastic leukaemia is related to in vitro growth characteristics. British Journal of Cancer, 68, 851-855.

Blood. 79, 2378-2383.

2755-2 760.

0 1995 Blackwell Science Ltd. British Journal of Haematology 90: 8-14