ELSEVIER

Combination of SCF, IL-6, IL-3, and GM-CSF Increases the Mitotic Index in Short Term Bone Marrow Cultures from Acute Promyelocytic Leukemia (APL) Patients

Brett Will iams and David J. Allan

ABSTRACT: Acute promyelocytic leukemia (APL) is characterized cytogenetically by the presence of a reciprocal translocation between chromosomes 15 and 17 [t(15;17)(q22-q24;qll-q21)] in the bone mar- row cells in the majority of patients. Cytogenetic evaluation of bone marrow cultures from patients with APL is often technically difficult, due to frequent difficult marrow aspiration and the suboptimal quality of cytogenetic p;reparations. This has important implications for the cytogenetic detection of residual disease. This stL!dy examined the proliferative ability of the recombinant human growth factors--stem cell factor (SCFf intedeukin-6 (IL-6), intedeukin-3 (IL-3), and granulocyte macrophage-colony stimulat- ing factor (GM-CSF)--to determine if they would provide a consistent improvement over the standard cytogenetic culturing techniques in terms of mitotic index (MI). In all cases, the MI of the growth factor stimulated cultures showed a considerably higher (3.5-198 fold) and statistically significant (p < 0.01) increase compared to the unstimulated cultures. We conclude that the use of recombinant human growth factors is potentially an effective way of increasing the MI in bone marrow cultures from APL patients for the purposes of diagnosis and residual disease detection. © Elsevier Science Inc., 1996

INTRODUCTION

APL is characterized by a reciprocal translocation that involves chromosomes 1_5 and 17. The breakpoints have been variously mapped to 15q22-q24 and 17q11-q21 [1-3]. Cytogenetic evaluation of bone marrow cultures from APL patients can be technicaiLly difficult, due to frequent diffi- cult marrow aspiration and the suboptimal quality and quantity of metaphases present in the cytogenetic prepara- tions obtained from the bone marrow cultures. The pres- ence of poor quality metaphases impacts directly on the ability to detect residual disease in these patients when using conventional cytogenetic techniques. Fluorescence in situ hybridization (FISH) has been used by a number of investigators for the detection of residual disease in patients with hematological malignancies [4-8]. Depend- ing on the probe type, F![SH can be used in the analysis of both metaphase and interphase cells and is not con- strained by the cell quality for the identification of the

From the Cytogeneties Laboratory (B. W.), Mater Adult Hospi- - tal, South Brisbane, and School of Life Science (D. J. A.), Queen- sland University of Technology, Brisbane, Australia.

Address reprint requests to: Brett Williams, Cytogenetics Department, Mater Adult I3rospital; South Brisbane 4101, Queens- land, Australia.

Received November 14, 1995; accepted May 11, 1996.

Cancer Genet Cytogenet 91:77-~31 (1996) © Elsevier Science Inc., 1996 655 Avenue of the Americas, N,~w York, NY 10010

t(15;17). Two recent papers have demonstrated the useful- ness of metaphase FISH as a method for the detection of residual disease in patients with APL [7, 8]. The major obstacle to this approach is the requirement of metaphases for analysis.

Much has been written about hematopoietic growth factors since the simultaneous discovery of the colony as- say for hematopoietic progenitor cells by Pluznik and Sachs (1965) and Bradley and Metcalf (1966). Many of the stud- ies have focused on measuring the proliferative ability of the cytokines by using 3H-thymidine incorporation assays or by observing their ability to promote the growth and formation of colony-forming units from both normal and leukemic progenitor cells. Very little work has been pub- lished on the effect of the cytokines on the karyotypes of the ceils being studied [9, 10].

This study examined the use of a cocktail of recombi- nant human growth factors--stem cell factor (SCF), inter- lenkin-6 (IL-6), interleukin-3 (IL-3), and granulocyte macrophage-colony stimulating factor (GM-CSF)--to de- termine if they would provide a consistent improvement over the standard cytogenetic culturing techniques in terms of mitotic index (MI). The bone marrow cells from the di- agnostic sample of six patients with APL were examined, and in all cases, the MI of the growth factor stimulated cultures showed a considerably higher and statistically significant increase compared to the unstimulated cul- tures. We conclude from our studies that the use of recom-

0165-4608/96/$15.00 PII S0165-4608(96)00155-0

78 B. Williams and D. J. Allan

binant human growth factors is an effective way of increasing the MI in bone marrow cultures from APL pa- tients for the purposes of diagnosis and residual disease detection.

MATERIALS AND METHODS

Patients This was a prospective study on six patients suspected of having acute promyelocytic leukemia according to the FAB criteria. All of the patients were found to have the t(15;17) at diagnosis. To date, follow-up samples have been received from three patients (1,5,6). One patient (2) died during induction therapy, and patients 3 and 4 were lost to follow-up. This study has received ethics approval from the University Research Ethics Committee of the Queensland University of Technology.

Cell Culture and Cytogenetic Analysis Bone marrow cells were cultured at 1 × 1 0 6 cells/ml in RPMI 1640 medium supplemented with 10% fetal calf serum, either in the presence or absence of recombinant human cytokines. Where possible, three cultures were set up: one culture was incubated at 37°C for 24 hrs, and two cultures were incubated for 48 hrs with fluorodeoxyuri- dine (FUdR) synchronization, as described elsewhere [11]. One of the 48 hr cultures had growth factors added and the other culture acted as the normal control. The harvest procedure of the 24 hr culture commenced with the addi- tion of colchicine (0.06 ug/ml) for 50 mins, followed by the addition of a hypotonic KC1 solution (0.075 M) and fix- ation in 3:1 methanol/acetic acid. The harvest procedure of the FUdR cultures was commenced with the addition of a hypotonic KC1 solution (0.075 M) containing colchicine at 37.5 ug/ml for 15 mins, followed by fixation in 3:1 methanol/acetic acid. G-banding was performed using a modified Seabright method [12] and karyotyped according to ISCN 1991 recommendations. A minimum of 20 recta- phases were analyzed for each culture.

Cytokines The cytokines used were SCF (100 ng/ml), IL-6 (20 ng/ml), IL-3 (25 ng/ml), and GM-CSF (20 ng/ml); all of the cyto- kines were a generous gift of AMGEN (Thousand Oaks, CA). The cytokines were chosen for their known ability to stimulate the early progenitor cells of the myeloid lineage. All growth factors were used at saturation concentrations as described elsewhere [13-16].

Determination of the Mitotic Index The fixed cell suspension was diluted with 3:1 methanol/ acetic acid and counted using a Technicon H1 automated hematology analyzer. Ten to twenty microliters of cell sus- pension was placed onto a slide, air dried, and stained with giemsa. The metaphases present were counted using a light microscope; the MI was determined by dividing the number of metaphases by the total number of cells on the slide, and expressed as a percentage. The results from the growth factor cultures were compared to the results from the cultures without growth factors at the 48 hour

time point. The range of metaphases counted per slide was 2-696 and the range of interphase nuclei present per slide was 9300-33800. Statistical analysis was performed using the non-parametric Wilcoxon two-sample test.

Fluorescence In Situ Hybridization A chromosome 17 whole chromosome paint probe was obtained from CAMBIO (Cambridge, UK) for the hybrid- ization experiments. Cell pellets of the bone marrow cells remaining from cytogenetic analysis were washed with fresh 3:1 methanol/acetic acid fixative, resuspended in fixative, and dropped onto dry clean slides. The slides were aged 24 hours at room temperature, in a slide box. After treatment with RNAse for 1 hour at 37°C, the slides were dehydrated in a cold ethanol series, then denatured in 70% formamide at 70°C for 2 minutes, followed by dehydration in a cold ethanol series. Probe preparation, hybridization, and post wash conditions were as described by the manufacturer. The hybridized probe was detected by using a mouse monoclonal anti-biotin antibody (DAKO), followed by an FITC conjugated rabbit anti-mouse anti- body (DAKO). The slides were counterstained in a DAPI (0.1 ug/ml)/propridium iodide (0.3 ug/ml) solution for 5 mins, then mounted in an antifade (DABCO) solution. Examination of the hybridization slide was carried out using a Nikon Optiphot II epifluorescent microscope. Photographs were taken using Kodak EPH 800 ASA color slide film.

RESULTS

There was a 3-198 fold increase in the mitotic index of the growth factor-enhanced cultures compared to the normal cultures (Table 1), p < 0.01. Although the rate of the increase varied from case to case, there was an 8.8 fold increase in the mitotic index, on average, for the six patients studied. Metaphases were obtained from all non- growth factor cultures except for the 48 hr culture of patient 3. This was principally due to the sample being 48 hrs old before being received in the laboratory. It was still possi- ble to obtain a 0.5% mitotic index from the growth factor culture even with the advanced age of the sample. In patients 3 and 5, there was insufficient bone marrow sam- ple available to initiate a 24 hr culture without growth fac- tors. The metaphases obtained from the cytokine-enhanced

Table I Mitotic indices (%) of bone marrow cells cultured with (+) and without ( -) growth factors (GF)

Patient 24 hr MI - GF 48 hr MI - GF 48 hr MI + GF

1 0.16 0.04 0.73 2 0.05 0.02 3.96 3 nd 0.0 0.5 4 2.4 0.66 2.3 5 nd 0.33 1.0 6 0.01 0.01 0.6

nd = not done.

Combination of SCF, IL-6, IL-3, and GM-CSF Increase the MI in APL Patients 79

Table 2 Karyotype of bone marrow cells from APL patients cultured with (+) or without ( -) growth factors (GF)

Karyotype

Patient 24 hr - GF 48 hr - GF 48 hr + GF

46,XY,t(15;17)(q24;q21)[20] 46,XX,t(15;17)(q24;q21)[1]/ 47,XX,+ 8,t(15; 17)(q24;q21)[19]

ND 47,XY,+8,t(15; 17)(q24;q21)[16]/ 46,XY[4]

ND U/S

46,XY,t(15;17)(q24;q21)[20]

U/S

U/S 47,XY,+ 8,t(15;17)(q24;q21)[16]/ 46,XY[4] 46,XX,t(15;17)(q24;q21)[15]/ 46,XX,idern,add(22)(p11)[5] 46,XX,t(15;17)(q24;q21)[3]

46,XY,t(15;17)(q24;q21)[20] 46,XX,t(15;17)(q24;q21)[8]/ 47,XX,+8,t(15;17)(q24;q21)[12] 46,XY,der(15)t(15;17)(q24;q21),ider(17)t(15;17) (q24;q21)[34]/46,XY[2] 47,XY,+8,t(15;17)(q24;q21)[17]/ 46,XY[3] 46,XX,t(15;17)(q24;q21)[15]/ 46,XX,idem,add(22)(p11)[5] 46,XX,t(15;17)(q24;q21)[20]

nd = not done. u/s = unsuccessful culture.

cultures are generally superior in quality compared with metaphases obtained from the normal cultures initiated from the APL patients. In every case where there was a comparable normal culture set up, the technical staff in the laboratory preferred to analyze the growth factor- enhanced cultures due to the improved spreading, length, and morphology, compared to the metaphases obtained from the normal cultures.

The karyotypes of the cultures examined after 24 and 48 hrs of culture, with and without cytokines, are shown in Table 2. The t(15;17) was detected in all patients stud- ied. Patients 1, 5, and 6 had a single abnormal clone con- taining the t(15;17). In patients 2, 3, and 4, there were two clones present. Except for patient 2, there was no apparent enrichment for cytogenetically abnormal or normal meta- phases observed after stimulation of the cultures with re- combinant human growth factors. With patient 2, in the 24 hour culture, the trisomy 8 cell line represented 95% of the metaphases analyzed and the cell line counting only the t(15;17) represented 5%. The percentages of the two cell lines changed in the 48 hour stimulated culture to 60% and 40%, respectively.

There was a 1-5 fold increase in the mitotic index of the growth factor-enhanced cultures compared to the nor- mal cultures from the :Follow-up samples received from three patients who were in complete cytogenetic remis- sion by conventional cytogenetic analysis and by FISH (Table 3). Patients 1 and 5 are RT-PCR negative, while PCR results are not available for patient 6. The average mitotic

index increase fell to 2.2 fold in the follow-up samples. The lower response of the cells to the growth factors can, most likely, be attributed to a decrease in the number of cells present that are responsive to the cocktail of growth factors added to the cultures. Using the FISH technique for detection of residual disease in the three follow-up pa- tients resulted in up to 15 times the number of metaphases being examined, compared to the number of metaphases examined by conventional cytogenetic analysis.

DISCUSSION

The marked increase in the MI (3-198 fold) observed in the cytokine-enhanced cultures at the 48 hour time point has practical advantages for the cytogeneticist. First, since the metaphase is the "gold standard" in cytogenetics, the cytokine-enhanced cultures enable conventional cytogenetic analysis to be performed rapidly due to the abundance of good quality metaphases. Second, the characterization of additional clones present in the culture, even if they affect a small percentage of cells, is facilitated.

The cytokine cocktail affected the karyotypes obtained in only one of the three patients (patient 2) with mixed clones. There are three possible reasons for the change in percentages of the clones in the 48 hour cytokine-en- hanced culture from patient 2: First, the cytokines may have favored growth of the clone containing the t(15;17) only. Second, this clone may have simply required the ex- tra time in culture to proliferate. Third, the clone with tri-

Table 3 Karyotype, FISH, and Mitotic index results from the follow-up bone marrow samples received from patients 1, 5, and 6

Patient 24 hr - GF/(MI) 48 hr - GF/(MI) 48 hr + GF/(MI) 48 hr + GF/FISH

1 46,XY[20]/(nd) 46 ,XY[20] / (0 .48) 46,XY[20]/(0.6~1) 0/200 1 46,XY[20]/(nd) 46 ,XY[20] / (0 .34) 46,XY[20]/(0.35) 0/200 5 46,XX[20]/(0.26) 46 ,XX[20] / (1 .2 ) 46,XX[20]/(3.2) 0/300 5 46,XX[20]/(1.()) 46 ,XX[20] / (0 .32) 46,XX[20]/(0.4) 0/300 6 46,XX[20]/(0.12) 46,XX[20]/(0 .13) 46,XX[20]/(0.64} 0/185

nd = not done.

80 B. Williams and D. J. Allan

samy 8 phJs the t(15;17) may have been autocrine in its own right, giving this clone a growth advantage in "nor- mal cultures," while the clone containing the t(15;17) translocation may have required the cytokines to promote its proliferation. It is unfortunate that the metaphases from the 48 hour culture without recombinant human cytokines were unanalyzable because of their extremely poor mor- phology and spreading, as the result from this culture would have confirmed or rejected the second hypothesis for the percentage change in the two clones present. While karyotypes of two of the three mixed clone cultures were unaffected by the cytokine stimulation, individual re- sponses of mixed clone cultures to stimulation may pose a limitation on the use of cytokine stimulation. However, in- creasing the percentage of a minor clone in a stimulated culture may be helpful in conventional cytogenetic analy- sis, as it was in patient 2, by decreasing the amount of time required to characterized the minor clones.

The use of recombinant human cytokines to increase the MI also facilitated the rapid analysis of metaphases by FISH to detect the t(15;17). Both metaphase and inter- phase FISH have been used to detect residual disease in hematological malignancies [4-8]. Interphase FISH does not rely on cycling cells to detect the residual malignant cells, so it is able to measure the constitutive level of malig- nant cells present. However, the ability of interphase FISH to assess the level of residual malignant cells is hampered by the presence of false positive and negative results [17]. Metaphase FISH, on the other hand, is not constrained by the quality of the metaphases or by false positive or false negative results [5, 17]. Metaphase FISH is limited only by the number of metaphases present and the fact that it is as- sessing only the cycling cell compartment. The use of re- combinant human cytokines to increase the mitotic index of the cultures in the study facilitated the rapid analysis by FISH of large numbers of metaphases, thus potentially increasing the sensitivity of detection of cells containing the t(15;17). The metaphase FISH approach to detection of residual disease can be applied to any malignancy where a particular chromosome rearrangement has been found by conventional cytogenetic analysis. An additional advan- tage to the metaphase FISH approach is that a wider range of probes can now be used to assess the level of residual disease present in the cycling cell compartment.

This study shows that the use of cytokines may be of practical advantage in cytogenetic analysis of APML and, potentially, other hematological malignancies. Cytokine stimulation can produce greatly enhanced numbers of metaphases for analysis by conventional cytogenetic tech- niques and by FISH. This may allow for greater accuracy in the characterization of minor clones present, and for greater sensitivity in the detection of residual disease by FISH and conventional cytogenetic methodologies. How- ever, because of the differential response of clones within a mixed clone sample, as in the case of patient 2, some de- gree of caution in interpretation may be required if con- ventional cytogenetic analysis and FISH were performed solely on metaphases obtained from cytokine stimulated cultures. TO maximize the potential benefits obtained

from the use of cytokines in culture, both the culture con- ditions and the cytokines used have to be optimized for each particular hematological malignancy; work is con- tinuing in this area.

The authors would like to thank the technical staff at the Mater Cytogenetics Department for their expert technical assistance, and Colleen Rapkins for supplying bone marrow samples from two of the patients in this study.

REFERENCES

1. Rowley JD, Golomb HM, Dougherty C (1977): 15/17 Translo- cation: A Consistent chromosome change in acute promyelo- cytic leukaemia. Lancet 1:549.

2. Mitelman F (1988): Catalogue Of Chromosome Aberrations In Cancer. Liss. New York, NY.

3. Donti E, Longo L, Pelicci P (1991): Chromosomal localization of the APL t(15;17) breakpoints by molecular cytogenetic analysis. Cancer Genet Cytogenet 54:265.

4. Poddighe P, Moesker O, Smeets D, Awwad B, Ramaekers F, Hopman A (1991): Interphase cytogenetics of hematological cancer: comparison of classical karyotyping and in situ hybridization using a panel of eleven chromosome specific DNA probes. Cancer Research 51:1959.

5. Zhao L, Kantarjian H, Van Oort J, Cork A, Trujillo J, Liang J (1993): Detection of residual proliferating leukemic cells by fluorescence in situ hydridization in CML patients in com- plete remission after interferon treatment. Leukemia 7:168- 171.

6. Jenkins R, Lebeau MM, Kraker W, Borell T, Staboerger P, Davis E, Penland L, Fernald A, Espinosa R, Schaid D, Moel P, Dewald G (1993): Fluorescence in situ hybridization: A sen- sitive method for trisomy 8 detection in bone marrow speci- mens. Blood 79:537.

7. Zhao L, Change K, Estey E, Hayes K, Albert B, Deisserroth A, Liang J (1995): Detection of residual leukaemic cells in patients with acute promyelocyic leukaemia by the fluores- cence in situ hybridization method: potential for predicting relapse. Blood 85:495.

8. Schad C, Hanson C, Paietta E, Casper J, Jalal S, Dewald G (1994): Efficacy of fluorescence in situ hybridization for detecting PML/RARA gene fusion in treated and untreated acute promyelocytic leukemia. Mayo Clinic Proceedings 69:1047-1053.

9. Kerim S, Rege-Cambrin G, Scarauaglio P, Carozzi F, Saglio G, Aglietta M (1991): Effect of recombinant human IL-3 on the mitiotic index and karyotype of hematopoietic cells. Cancer Genet Cytogenet 55:235-241.

10. Goselink H, Williams D, Fibbe W, Wessels H, Beverstock G, Willemze R, Falkenburg F (1992): Effect of mast cell growth factor (c-kit Ligand) on clonogeneic leukaemic precursor cells. Blood 80:750-757.

11. Webber LM, Garson OM (1983): Fluorodeoxyuridine syn- chronization of bone marrow cultures. Cancer Genet Cytoge- net 8:123-132.

12. Seabright M (1971): A rapid banding technique for human chromosomes. Lancet 2:971-972.

13. Brandt J, Briddell R, Leemhuis T, Hoffman R (1992): Role of c-kit Ligand in the expansion of human haematopoietic pro- genitor cells. Blood 79:634-641.

14. Tsuji K, Lyman S, Sudo T, Clark S, Ogawa M (1992): Enhance- ment of murine hematopoiesis by synergistic interactions

Combinat ion of SCF, IL-6, IL-3, and GM-CSF Increase the MI in APL Patients 81

between steel factor (]igand for c-kit), interleukin-11, and other early acting factors in culture. Blood 79:2855-2860.

15. Pietsch T, Kyas U, Stefens U, Yakisan E, Hadam M, Ludwig W, Zsebo K, Welte K (1992): Effects of stern cell factor (c-kit ligand) on the proliferation of myeloid leukaemia cells: het- erogeneity in response and synergy with other hematopoietic growth factors. Blood 80:1199-1206.

16. Migliaccio G, Migliaccio A, Druzin M, Giardina P, Zsebo K,

Adamson J (1992): Long-term generation of colony-forming cells in liquid culture of CD34+ cord blood cells in the pres- ence of recombinant human stem cell factor. Blood 7 9:2620- 2627.

17. Seong D, Kantarjian H, Ro J, Talpaz M, Xu J, Robinson J, Dies- seroth A, Champlin R, Siciliano M (1995): "Hypermetaphase FISH" for quantitative monitoring of Philadelphia chromosome positive cells in chronic myelogenous leukaemia patients dur- ing treatment. Blood 86:2343-2349.