Download PDF

short report
Centrosome aberrations after nilotinib and imatinib treatment
in vitro are associated with mitotic spindle defects and genetic
instability
Alice Fabarius,1 Michelle Giehl,1 Oliver
Frank,1 Birgit Spiess,1 Chun Zheng,1
Martin C. Müller,1 Christel Weiss,2 Peter
Duesberg,1, 3 Rüdiger Hehlmann,1
Andreas Hochhaus1 and Wolfgang
Seifarth1
1
III. Medizinische Universitätsklinik, 2Abteilung
für Medizinische Statistik, Medizinische Fakultät
Mannheim der Universität Heidelberg,
Mannheim, Germany, and 3Department of
Molecular and Cell Biology, Donner Laboratory,
University of California, Berkeley, CA, USA
Received 3 April 2007; accepted for publication
Summary
Centrosomes play fundamental roles in mitotic spindle organisation,
chromosome segregation and maintenance of genetic stability. Recently, we
have demonstrated that the tyrosine kinase inhibitor imatinib induces
centrosome and chromosome aberrations in vitro. Here, we comparatively
investigated the effects of imatinib and the more potent successor drug
nilotinib on centrosome, mitotic spindle and karyotype status in primary
human fibroblasts. Therapeutic doses of imatinib and/or nilotinib
administered separately, consecutively or in combination similarly induced
centrosome, mitotic spindle, and karyotype aberrations. Our data suggest
that distinct tyrosine kinases likewise targeted by both drugs are essential
actuators in maintenance of centrosome and karyotype integrity.
Keywords: chronic myeloid leukaemia, centrosomes, mitotic spindle, nilotinib, imatinib.
11 May 2007
Correspondence: Dr Alice Fabarius, PhD, III.
Medizinische Universitätsklinik, Fakultät für
Klinische Medizin Mannheim der RuprechtKarls-Universität Heidelberg, Wiesbadener
Straße 7-11, 68305 Mannheim, Germany.
E-mail: [email protected]
Imatinib is an ATP-mimicking inhibitor targeting the tyrosine
kinase activity of BCR-ABL-associated human leukaemias.
Despite the success of imatinib in chronic myeloid leukaemia
(CML) first line treatment, response rates of patients in
accelerated or blastic phase are significantly decreased. Thus,
there is a need for novel BCR-ABL tyrosine kinase inhibitors
with greater potency and with the capability to overcome
imatinib-resistance. Computer modelling has led to the
development of nilotinib (AMN107; Novartis Pharma, Basel,
Switzerland), an imatinib derivative with a more selective and
at least 30-fold increased inhibitory potency when compared to
imatinib (Kantarjian et al, 2006). Nilotinib has been described
to act synergistically when combined simultaneously or
sequentially with imatinib, making it highly promising for
future therapy regimens to overcome BCR-ABL-related imatinib resistance in advanced phases of CML (Weisberg et al,
2005, 2006). However, emergence of Ph-negative clones with
aberrant karyotypes distinctly different from the Ph-positive
clone has been reported under continuous imatinib therapy,
raising the controversial issue whether CML genetic instability
might be caused or solely augmented by drug-related c-ABL
inhibition (Bacher et al, 2005).
Recently, we showed that imatinib treatment of normal
human and mammalian cells in vitro caused unexpected
centrosome and chromosome aberrations resembling lesions
observed in CD34+ haematopoietic stem and progenitor cells
concurring with the CML transformation process (Giehl et al,
2005). This supports the hypothesis that imatinib itself may
play a role in the emergence of karyotype aberrations in
Ph-negative cells (Fabarius et al, 2005). Albeit one can surmise
a mechanism of action similar to that of imatinib, there is in
fact a complete lack of comparable data on the effects of
ª 2007 The Authors
Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 138, 369–373 doi:10.1111/j.1365-2141.2007.06678.x
Short Report
nilotinib on normal human primary cells as most in vitro
studies have been performed on aberrant tumour cell lines
(Weisberg et al, 2005). To address the question whether and to
what extent nilotinib might affect genomic stability through
induction of centrosomal and mitotic spindle aberrations, we
comparatively investigated the effects of imatinib and nilotinib
on centrosome, mitotic spindle and karyotype status in
primary human fibroblasts after administration of therapeutic
doses of both drugs in vitro.
Materials and methods
Primary cells
Normal human dermal fibroblasts (NHDF; Promocell GmbH,
Heidelberg, Germany) were grown as described previously
(Fabarius et al, 2005).
spindles were examined. Asymmetric and tri-, tetra- or
multipolar patterns were considered abnormal.
Cytogenetics
Twenty metaphases were analysed by G-banding technique as
described (Schoch et al, 2002).
Statistical analysis
Analyses of variance (anovas) and multiple regression analyses were used to compare control and drug treatment groups
and different concentrations within the same group. P-values
£0.001 were considered significant. All statistical computations
were done with the sas software package, release 8.02 (SAS
Institute Inc., Cary, NC, USA).
Results
In vitro drug exposure
Cells were treated with: (i) 1–20 lmol/l nilotinib (Novartis,
Basel, Switzerland, 3 weeks); (ii) 5–20 lmol/l imatinib (Novartis, 3 weeks); (iii) 5, 10 and 20 lmol/l imatinib (3 weeks)
followed by 0.5, 1 and 2 lmol/l nilotinib respectively
(3 weeks); (iv) 0.5, 1 and 2 lmol/l nilotinib followed by 5,
10 and 20 lmol/l imatinib (3 weeks each); and (v) imatinib
and nilotinib simultaneously (3 weeks, identical concentrations). In vitro concentrations of 1–5 lmol/l represent pharmacological doses used therapeutically in vivo (Cwynarski
et al, 2004).
Centrosome and mitotic spindle staining
Cells were grown on polylysine-coated slides, fixed and
immunostained with an antibody to pericentrin. To confirm
centrosome-specific staining, sample subsets were co-stained
with gamma-tubulin directed antibody according to Giehl et al
(2005). Mitotic spindles were co-stained using polyclonal antialpha-tubulin antibody (No. T6074; Sigma, Deisenhofen,
Germany) under the same conditions. At least 20 mitotic
All nilotinib- and imatinib-treated NHDF cultures were
affected by both drugs and displayed similar centrosome and
sporadic chromosome aberrations as well as spindle defects
(Fig 1) when compared with controls (Fig 2) (P < 0.0001).
Nilotinib induces genetic instability in an imatinibmimicking manner
Cultures treated with nilotinib displayed centrosome alterations in 12–20% and spindle defects in 5.7–53%. Chromosome
alterations could be detected in 30–45% (Fig 2A). However,
cells treated with 20 lmol/l nilotinib showed lower aberration
values. Cells under imatinib treatment displayed centrosome
alterations in 9–19% and spindle defects in 27–32%. Chromosomal changes occurred in 20–45% (Fig 2B).
Sequential drug treatment renders both drugs
interchangeable
After nilotinib treatment and with consecutive administration
of imatinib, cells displayed centrosomal changes, chromosome
(A)
(B)
Fig 1. Mitotic spindles in normal human dermal fibroblasts (NHDF) cells. Representative images showing normal (A) and aberrant (B) mitotic
spindles were obtained by immunofluorescence staining of untreated and drug-treated NHDF cells respectively. Mitotic spindles were targeted with
an antibody directed to alpha-tubulin, followed by a fluorescein isothiocyanate-conjugated secondary antibody.
370
ª 2007 The Authors
Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 138, 369–373
Short Report
(A)
(B)
(C)
(D)
(E)
Fig 2. Correlation between centrosome, mitotic spindle and chromosome aberrations in normal human dermal fibroblasts (NHDF) cultures treated
with nilotinib (A), imatinib (B), sequential treatment with nilotinib followed by imatinib (C) and vice versa (D), and with a simultaneous
combination of both drugs (E). Incubation times for NHDF cultures under mono- and combinatorial drug treatment were 3 weeks, for sequential
drug regimen 6 weeks (3 weeks per drug). Various drug concentrations (x-axis) indicate a dose dependency of the observed aberrations (P < 0.0001
for all experiments).
aberrations and spindle defects (Fig 2C). Inversely, imatinib
treatment followed by nilotinib showed comparable results
(Fig 2D). No differences in alteration patterns could be
described after sequential treatment with both drugs compared
to mono-treatment irrespective of the order of drug administration.
Simultaneous drug treatment induces alterations and is
toxic at high doses
Simultaneous treatment with imatinib and nilotinib also led to
centrosome and chromosome aberrations. Spindle defects
could be observed in 9% of the 1 lmol/l/10 lmol/l imatinib/
nilotinib treated culture, but not in 0.5 lmol/l/5 lmol/l
treated cells and in controls. Chromosome changes occurred
in 15–35% (Fig 2E). Simultaneous high-dose treatment with
2 lmol/l/20 lmol/l nilotinib/imatinib proved deleterious for
cells, suggesting synergistic effects in vitro.
In all treated cultures, sporadic numerical chromosomal
aberrations prevailed over sporadic structural alterations.
Neither clonal changes nor a prevalence of specific chromosomes in aberrations were observed. Numerical and structural
centrosome alterations occurred in an equal incidence.
Our data demonstrated destabilising effects of both drugs in
therapeutic doses (400 mg/800 mg equivalents) on centrosomes, chromosomes and mitotic spindle fidelity in vitro
(P < 0.0001). The same outcome was achieved after culture
treatment with imatinib, nilotinib or both (sequential or
simultaneous drug treatment) (P ¼ 0.89, P ¼ 0.66 or
P ¼ 0.63 respectively). Sequential drug treatment revealed
that both drugs were interchangeable in terms of their resulting
effect.
Discussion
Emergence of Ph-negative clones with aberrant karyotypes
under continuous imatinib therapy gives rise to the speculation that tyrosine kinase inhibitors themselves may trigger
induction of genetic instability. As this adverse effect may be
directly proportional to the drug’s therapeutic potency, this
could have serious impact for patients possibly switching drugs
upon emergence of imatinib resistance and who are advised to
receive such medication lifelong.
Applying therapeutic doses of both drugs in vitro, we found
similar dose-dependent centrosome and karyotype alterations in
NHDF cells. These alterations correlated with spindle defects
(Figs 1 and 2) functionally linking tyrosine kinase inhibitors
with loss of centrosomal integrity and karyotype stability.
Our findings can help to explain why 2–17% of imatinibreceiving patients display karyotype changes in Ph-negative
cells under long-term treatment (Bacher et al, 2005). In vitro
emergence of tyrosine kinase inhibitor-related alterations in
normal human primary cells points to drug-associated mechanisms of de novo induction of aberrations rather than to
selection of pre-existing clonal aberrations of the Ph-negative
haematopoiesis being uncovered by a simultaneous gradual
ª 2007 The Authors
Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 138, 369–373
371
Short Report
elimination of the Ph-positive clone under therapy (Cortes &
O’Dwyer, 2004). The possibility that both inhibitors may
simply augment genetic instability in cells already carrying
hidden genetic defects, thus favouring the acquisition of
further chromosome defects, including the BCR-ABL gene
rearrangement, seems, albeit not impossible, fairly unlikely, as
it would imply that normal human cells harbour mutations at
considerable frequency with capability of centrosome and
gross genomic destabilisation.
To assess potential in vivo side effects of tyrosine kinase
inhibitors on haematopoietic cells, nine patients (imatinib,
n ¼ 5; nilotinib, n ¼ 4) were analysed and spindle defects,
however, at much lower frequencies (range, 5–10%) were
observed (data not shown). However, due to the lack of proper
controls (patients lacking prodromal chemotherapy) these data
have to be considered preliminary. Nevertheless, the discrepancy
between the in vitro and in vivo situation, i.e. higher in vitro
aberration rates in NHDF cells than in haematopoietic cells
in vivo, could be explained by better drug availability in vitro or
differing drug sensitivities. The complete lack of effective
mechanism to eliminate or suppress aberrant clonal phenotypes
in vitro is also conceivable. Imatinib has been reported to affect
the function of normal non-malignant cells and concurs with
altered gene expression and myelosuppression in treated
patients (Mattiuzzi et al, 2003; Balabanov et al, 2005a,b). Thus,
a resulting altered immunosurveillance could explain higher
incidences of secondary malignancies and infections under
imatinib (Mattiuzzi et al, 2003; Bacher et al, 2005).
The molecular mechanisms for the observed detrimental
effects of imatinib and nilotinib on centrosome and karyotype
stability seem to involve one or more ABL-related tyrosine
kinases operating on regulation of centrosome replication,
DNA repair and cell cycle. Potential mechanisms include the
RAD51 protein that plays a fundamental role in DNA double
strand break repair and is essentially regulated by c-ABL
(Bertrand et al, 2003). Moreover, inhibition of c-ABL involved
in p53-dependent G1 arrest response could cause defects in
DNA repair and G1 arrest, thus predisposing cells to aberrant
centrosome and genome duplication (Kharbanda et al, 1998).
Finally, drug-related inhibition of c-ABL may directly affect
ubiquitination of centrosomal components leading to extra
centrosomal duplication and mitotic spindle catastrophy
(Parvin & Sankaran, 2006).
In conclusion, our data demonstrate that therapeutic doses
of nilotinib and imatinib in vitro similarly trigger the
emergence of centrosomal, mitotic spindle and chromosomal
abnormalities. As both drugs are ATP-competitive inhibitors
targeting closely related tyrosine kinases, the same modes of
action on molecular levels interfering with proper centrosome
reduplication can be suggested.
Acknowledgements
The study was supported by the Albert und Anneliese KonanzStiftung, Heidelberg, Germany and the Competence Network
372
‘Acute and chronic leukemias’, sponsored by the German
Bundesministerium für Bildung und Forschung (Projektträger
Gesundheitsforschung; DLR e.V.- 01 GI9980/6). Nilotinib and
imatinib were kindly provided by Dr Paul Manley and Dr
Elisabeth Buchdunger, Novartis Pharma, Basel, Switzerland.
Author contributions
A.F., W.S., R.H. and A.H., designed the research; A.F., M.G.,
O.F., M.C.M., B.S., C.Z., C.W., P.D., and W.S. performed the
research; A.F., M.G., C.W., W.S. and A.H. analysed the data;
A.F., W.S., and A.H. wrote the paper and all authors checked
the final version of the manuscript.
References
Bacher, U., Hochhaus, A., Berger, U., Hiddemann, W., Hehlmann, R.,
Haferlach, T. & Schoch, C. (2005) Clonal aberrations in Philadelphia
chromosome negative hematopoiesis in patients with chronic
myeloid leukemia treated with imatinib or interferon alpha.
Leukemia, 19, 460–463.
Balabanov, S., Appel, S., Kanz, L., Brossart, P. & Brümmendorf, T.H.
(2005a) Effect of tyrosine kinase inhibition using imatinib on normal lymphohematopoietic cells. Annals of the New York Academy of
Sciences, 1044, 168–177.
Balabanov, S., Bartolovic, K., Komor, M., Kanz, L., Hofmann, W.K. &
Brümmendorf, T.H. (2005b) Gene expression profiling of normal
hematopoietic progenitor cells under treatment with imatinib
in vitro. Leukemia, 19, 1483–1485.
Bertrand, P., Lambert, S., Joubert, C. & Lopez, B.S. (2003) Overexpression of mammalian Rad51 does not stimulate tumorigenesis
while a dominant-negative Rad51 affects centrosome fragmentation,
ploidy and stimulates tumorigenesis, in p53-defective CHO cells.
Oncogene, 22, 7587–7592.
Cortes, J. & O’Dwyer, M.E. (2004) Clonal evolution in chronic
myelogenous leukemia. Hematology/Oncology Clinics of North
America, 18, 671–684.
Cwynarski, K., Laylor, R., Macchiarlo, E., Goldman, J., Lombardi, G.,
Melo, J.V. & Dazzi, F. (2004) Imatinib inhibits the activation and
proliferation of normal T lymphocytes in vitro. Leukemia, 18, 1332–
1339.
Fabarius, A., Giehl, M., Frank, O., Duesberg, P., Hochhaus, A.,
Hehlmann, R. & Seifarth, W. (2005) Induction of centrosome and
chromosome aberrations by imatinib in vitro. Leukemia, 19,
1573–1578.
Giehl, M., Fabarius, A., Frank, O., Hochhaus, A., Hafner, M., Hehlmann, R. & Seifarth, W. (2005) Centrosome aberrations in chronic
myeloid leukemia correlate with disease progression and chromosomal instability. Leukemia, 19, 1192–1197.
Kantarjian, H., Giles, F., Wunderle, L., Bhalla, K., O’Brien, S., Wassmann, B., Tanaka, C., Manley, P., Rae, P., Mietlowski, W., Bochinski, K., Hochhaus, A., Griffin, J.D., Hoelzer, D., Albitar, M.,
Dugan, M., Cortes, J., Alland, L. & Ottmann, O.G. (2006) Nilotinib
in imatinib-resistant CML and Philadelphia chromosome-positive
ALL. New England Journal of Medicine, 354, 2542–2551.
Kharbanda, S., Yuan, Z.M., Weichselbaum, R. & Kufe, D. (1998)
Determination of cell fate by c-Abl activation in the response to
DNA damage. Oncogene, 17, 3309–3318.
ª 2007 The Authors
Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 138, 369–373
Short Report
Mattiuzzi, G.N., Cortes, J.E., Talpaz, M., Reuben, J., Rios, M.B., Shan,
J., Kontoyiannis, D., Giles, F.J., Raad, I., Verstovsek, S., Ferrajoli, A.
& Kantarjian, H.M. (2003) Development of Varicella-Zoster virus
infection in patients with chronic myelogenous leukemia treated
with imatinib mesylate. Clinical Cancer Research, 9, 976–980.
Parvin, J.D. & Sankaran, S. (2006) The BRCA1 E3 ubiquitin ligase
controls centrosome dynamics. Cell Cycle, 5, 1946–1950.
Schoch, C., Schnittger, S., Bursch, S., Gerstner, D., Hochhaus, A.,
Berger, U., Hehlmann, R., Hiddemann, W. & Haferlach, T. (2002)
Comparison of chromosome banding analyses, interphase- and
hypermetaphase-FISH, qualitative and quantitative PCR for diag-
nosis and follow-up in chronic myeloid leukemia: a study on 350
cases. Leukemia, 16, 53–59.
Weisberg, E., Manley, P.W., Breitenstein, W., Bruggen, J., Cowan-Jacob, S.W., Ray, A., Huntly, B., Fabbro, D., Fendrich, G., Hall-Meyers, E., Kung, A.L., Mestan, J., Daley, G.Q., Callahan, L., Catley, L.,
Cavazza, C., Azam, M., Neuberg, D., Wright, R.D., Gilliland, D.G. &
Griffin, J.D. (2005) Characterization of AMN107, a selective inhibitor of native and mutant Bcr-Abl. Cancer Cell, 7, 129–141.
Weisberg, E., Manley, P., Mestan, J., Cowan-Jacob, S., Ray, A. &
Griffin, J.D. (2006) AMN107 (nilotinib): a novel and selective
inhibitor of BCR-ABL. British Journal of Cancer, 94, 1765–1769.
ª 2007 The Authors
Journal Compilation ª 2007 Blackwell Publishing Ltd, British Journal of Haematology, 138, 369–373
373