Common Polymorphism G(-248)A in the Promoter Region of the bax Gene Results in Significantly Shorter Survival in Patients With Chronic Lymph
http://www.100md.com
《临床肿瘤学》
the Department of Haematology, Heartlands and Solihull National Health Service Trust
Institute for Cancer Research, University of Birmingham, Birmingham
Department of Haematology, University of Wales College of Medicine, Heath Park, Cardiff
Department of Haematology, Llandough Hospital, Penarth, Vale of Glamorgan, United Kingdom
ABSTRACT
PATIENTS AND METHODS: The frequency of the novel polymorphism, G(–248)A, in the promoter region of the bax gene and bax protein expression was assessed in 203 CLL patients. The results were correlated with clinical outcome.
RESULTS: The polymorphism was found in 23% of the CLL cohort and 15% of normal controls with no significant difference in allele frequency between the two groups (P = .15). It was associated with lower Bax protein expression and a shorter overall survival, especially in the treated patient group (P = .03). Furthermore, the adverse impact of the polymorphism was accentuated when comparing survival from the date of first treatment rather than diagnosis (P = .012). No significant difference in age at diagnosis, stage of disease at presentation, lymphocyte doubling time, time to first treatment, or progression-free survival were observed.
CONCLUSION: The presence of this single nucleotide polymorphism in CLL critically influences the response to treatment and overall survival. Given the relatively high prevalence of this polymorphism in the normal population, further prospective studies in CLL and other human malignancies are indicated.
INTRODUCTION
Over the last decade, experimental evidence has pointed to a key role for the Bcl-2 family not only in disease progression but also in determining response to therapy and clinical outcome in CLL.7-12 Indeed, Bcl-2 expression has been shown to be an independent prognostic factor and the Bcl-2/Bax ratio and the Mcl-1/Bax ratio seem to be important in determining both in vitro and in vivo response to chemotherapeutic drugs in CLL.3-15 Bax has recently been shown to undergo conformational change in response to various chemotherapy agents and seems to play a pivotal role in drug-induced apoptosis in CLL cells.8,16 For many years, mutations and polymorphisms in the bax gene were not investigated for their potential role in the pathogenesis of CLL. However, a recent study of 34 CLL patients identified the presence of a single nucleotide polymorphism G(–248)A in the 5'-untranslated (promoter) region (UTR) of the bax gene.17 This polymorphism was found in 69% of stage I to IV patients, compared with only 5.5% of stage A0 patients and only 4% of healthy controls and was associated with a failure to achieve complete response to therapy. We therefore sought to extend these preliminary findings in a much larger single-center study to further define the role of this polymorphism in CLL.
PATIENTS AND METHODS
Indications for treatment were consistent with published guidelines, notably troublesome lymphadenopathy or hepato-splenomegaly, constitutional symptoms, or developing cytopenias.19 Patients requiring treatment were entered in the prevailing Medical Research Council (MRC) trial and hence received differing initial therapies. In the MRC CLL III trial, patients were randomly assigned to chlorambucil versus epirubicin/chlorambucil. The MRC CLL IV trial randomly assigned patients between chlorambucil versus fludarabine versus fludarabine/cyclophosphamide. In the MRC CLL pilot study, patients received fludarabine to maximal response followed by an autologous stem-cell transplantation. Salvage therapies depended on the initial therapy given but typically included fludarabine (with and without cyclophosphamide); cyclophosphamide, doxorubicin, vincristine, and prednisolone; alemtuzumab; and allogeneic stem-cell transplantation. The bax polymorphism analyses were largely performed retrospectively, and in no patients was the therapeutic decision dependent on the result. Patients who required treatment received a median of three different chemotherapy regimens.
DNA Isolation and Bax 5'-Untranslated (promoter) Region G(–248)A Polymerase Chain Reaction Analysis
DNA was extracted from peripheral blood taken from CLL patients and healthy volunteers using a standard salting-out method.20 Analysis of the 5'-untranslated (promoter) region polymorphism was undertaken using the following method. A 280–base pair fragment of the Bax 5'-untranslated (promoter) region was amplified using the following primer pair: 5'-TTAGAGACAAGCCTGGGCGT-3' and 5'-CAATGAGCATCTCCCGATAA-3'. The polymerase chain reaction (PCR) mix contained 22.5 μL of Reddy Mix (ABgene, Epsom, United Kingdom), with 3 mmol/L MgCl2 and 5 ng of each primer. After initial denaturation at 94°C for 5 minutes, 35 cycles of amplification were performed, consisting of denaturation at 94°C for 30 seconds, annealing at 48°C for 30 seconds, and extension at 72°C for 30 seconds, with a final extension step of 72°C for 7 minutes. PCR was performed on a GeneAmp PCR System 9700 (Applied Biosystems, Warrington, United Kingdom). A 12-μL aliquot of each PCR product was digested with 1 unit of Tau 1 restriction enzyme (MBI Fermentas, Helena BioSciences, Sunderland, UK) at 55°C for 4 hours. The Tau 1 enzyme recognizes the sequence GCSGC, and the presence of the A allele results in the sequence ACSGC, thus abolishing the restriction site. The digested products were electrophoresed on a 2% agarose gel for 40 minutes at 100 V, stained with ethidium bromide, and visualized using ultraviolet. The Tau 1 restriction enzyme digest produced three possible band patterns (Fig 1A). The normal allele was completely digested to give two bands, 234bp and 46bp, the mutant allele was uncut and gave a single band of 280bp, and all three bands were present in patients showing heterozygosity of the G(–248)A polymorphism.
To confirm whether this was a single nucleotide polymorphism or a point mutation, we also performed the analysis on DNA freshly extracted from buccal mucosa and compared the results with those obtained from the CLL samples. Buccal mouthwashes were obtained from 23 CLL patients. DNA was extracted from the mouthwashes using the QIAmp DNA Mini kit (Qiagen Ltd, Crawley, West Sussex, United Kingdom).
Bax Sequence Analysis
PCR products were directly sequenced using Big Dye Terminators version 3.0 (Applied Biosystems) on a 310 DNA sequencer (Applied Biosystems) according to the manufacturer's instructions. The Bax sequences were compared with available germline sequences for the bax gene, accession number U17193. The sequence analysis is shown in Figure 1B.
VH Gene Mutation Analysis
The mutational status of the immunoglobulin VH is an important prognostic indicator.21,22 RNA was extracted from the patient samples using Qiagen RNeasy mini kit (Qiagen). Complementary DNA (cDNA) was synthesized using the Reverse-iT kit (ABgene). The VH gene mutational status of 141 CLL patients was analyzed according to the method previously described.23 The resulting PCR products were sequenced using BIG dye terminator sequencing kit version 3 (Applied Biosystems), and the sequences were analyzed using Immunoglobulin BLAST (National Center for Biotechnology Information, Bethesda, MD). The sequences with a germline homology of 98% or higher were regarded as unmutated, and those with less than 98% were regarded as mutated.
Genetic Analysis
Peripheral blood samples from patients with advanced-stage disease underwent fluorescent in situ hybridization analyses looking for trisomy 12, 6q, 11q, 13q, and 17 p deletions (Vysis, London, United Kingdom) and/or karyotype analysis after 3 to 5 days of incubation with TPA (12-myristate 13-acetate) stimulation.
Analysis of Bcl-2, Bax, and Mcl-1 Protein Expression
Fresh peripheral blood samples from 122 consecutive patients attending the CLL clinic over a 9-month period were analyzed by triple immunofluorescent staining using a Bax antibody in conjunction with CD5 and CD19 antibodies. Briefly, 1 x 106 cells were incubated with 10 μL of both anti-CD5 phycoerythrin-conjugated and anti-CD19 phycoerythrin cyanine 5-conjugated antibodies or isotype-matched controls (DAKO, Ely, United Kingdom). The cells were fixed using a commercially available kit (DAKO) and resuspended in a permeabilization solution together with one of the following antibodies: 7 μL of anti-Bcl-2 fluorescein isothiocyanate (FITC)–conjugated antibody (DAKO), 8 μL of anti-Bax FITC-conjugated antibody (Santa Cruz Biotechnology, Santa Cruz, CA), 8 μL of anti-Mcl-1 antibody (Santa Cruz Biotechnology), or an isotype-matched control (DAKO). The Mcl-1 labeled cells were subsequently subjected to secondary labeling with a FITC-conjugated antibody (DAKO). After a final washing step, the cells were resuspended in 0.5 mL of 1% paraformaldehyde before flow cytometric analysis using a FACScan flow cytometer (Becton Dickinson, San Jose, CA). From each sample, at least 10,000 cells were analyzed, and nonspecific binding was excluded by gating using the isotype control antibody. Gating of the CD5+/CD19+ cells was performed in the analyses in order to the specific protein expression in leukemic CLL cells. The mean fluorescent intensity was calculated for each individual protein using WINMDI software (J. Trotter, Scripps Research Institute, La Jolla, CA) and these values were then converted to molecules of equivalent soluble fluorochrome using a calibration curve to standardize the data.
Immunoblotting
A total of 3 x 107 CLL cells derived from previously untreated patients with either the wild-type (GG) or polymorphic (GA) bax genotype were lysed in 1 mL of lysis buffer containing 1% Triton X-100. Protein concentration was determined by Bradford assay (Bio-Rad, Hemel Hempsted, United Kingdom). Each sample (12 μg of protein) was separated on a 4% to 12% Bis-Tris acrylamide gel and transferred to polyvinylidene fluoride membrane (Bio-Rad). Western blots were incubated with a mouse monoclonal antibody against Bax (clone 4F11; Beckman Coulter, High Wycombe, United Kingdom), in phosphate-buffered saline with 4% milk. The secondary antibody was a horseradish peroxidase–conjugated sheep antimouse immunoglobulin G (Amersham Biosciences, Chalfont St Giles, United Kingdom), and detection was performed by an enhanced chemiluminescence method (Amersham).
Clinical/Laboratory Prognostic Parameters
To assess the role of the G(–248)A 5'-untranslated (promoter) region polymorphism of the bax gene, comparison was made between various clinical/laboratory parameters including expression of Bax protein, VH gene mutational status, Binet staging, age at presentation, lymphocyte doubling time (> or < 12 months), time to first treatment, progression-free survival, and overall survival.
Statistical Analysis
Patients were grouped into two subsets depending on genotype (GG v GA/AA), and statistical analysis was performed using parametric and nonparametric methods depending on whether the data obeyed a Gaussian distribution. Age at diagnosis and Bax protein expression were analyzed using the unpaired Student's t test. Comparison of the allele frequency of the G(–248)A polymorphism, VH mutational status, lymphocyte doubling time, and patient sex were analyzed using the Fisher's exact test, and all other parameters were analyzed using the Mann-Whitney test, except for the stage of disease at presentation, where the {chi}2 test was used. Survival analysis was calculated according to the Kaplan-Meier method. All statistical analyses were performed using the Graphpad Prism 3.0 software (Graphpad Software Inc, San Diego, CA).
RESULTS
Studies over recent years have identified that in up to 5% of CLL patients, there is a familial tendency.24-26 Given the central role of Bax in the pathogenesis of CLL and the identification of this potentially etiologic hereditary factor, we sought to study further this particular bax polymorphism in a large single-center study. Within the CLL cohort, there were six patients who were part of a familial CLL pedigree. All of these patients had the GG genotype, indicating that the G(–248)A 5'-untranslated (promoter) region polymorphism is not the hereditary factor responsible for familial CLL.
Clinical Features and Prognostic Markers
In contrast with the previously published data,17 we found no increase in frequency of the A allele in CLL cases with more advanced disease as defined by the Binet classification system (P = .62). There was no significant difference in mean age at diagnosis or median follow-up time between the GA/AA and GG genotypic groups (P = .62 and .39, respectively). Neither was there a significant difference in lymphocyte doubling time (> or < 12 months), the time to first treatment, or progression-free survival (P = .32, .38, and .42, respectively). However, the overall survival of the GA/AA genotypic group was significantly shorter than the GG genotypic group when analyzing only CLL-related deaths (P = .05). The data are summarized in Table 2, and the overlaid Kaplan-Meier survival curves for the GG patient group and the combined GA/AA patient groups are shown in Figure 2.
The VH gene mutational status of CLL patients is an important prognostic marker with those patients with unmutated VH genes having significantly shorter survival times. This was confirmed in this study with the unmutated patient subset having a significantly shorter survival time than the mutated subset (P = .002; Fig 3A). However, we found no bias in distribution in terms of VH gene mutational status between the bax genotypic subgroups (GG v GA/AA); 28 (28%) of 101 CLL patients with the GG genotype and eight (32%) of 25 patients with the GA/AA genotypes had unmutated VH genes (P = .81). Furthermore, subdivision of the mutated and unmutated groups by genotype (GG v GA/AA) revealed no significant difference in overall survival between the wild-type and polymorphic groups in either the mutated subset (P = .24) or the unmutated subset (P = .17; Fig 3B).
Genetic analysis was available in 66 patients with the GG genotype and 28 patients with the GA/AA genotype. Comparisons between the GG and GA/AA showed no significant difference in the frequency of a normal karyotype (32 of 66 v 13 of 28 patients), 11q deletions (15 of 66 v six of 28 patients), 17p deletions (four of 66 v two of 28 patients), and 13q deletions; trisomy 12 and 6q deletions were present in 15 of 66 and seven of 28 patients, respectively.
Ex Vivo Bax Protein Expression
Previous studies have shown the importance of Bax protein expression, high Bcl-2/Bax ratios, and Mcl-1/Bax ratios in determining both in vitro and in vivo response to chemotherapy.14,15 In this study, CLL patient samples (122 of 203) were assessed for ex vivo expression of Bax to determine whether any of these parameters were significantly different between the bax genotypic subsets (GA/AA v GG). Bax protein expression was significantly lower in those patients with a GA/AA genotype (P = .05; Fig 4A). Because we have previously shown that drug resistance may be, at least in part, mediated by selection of subclones expressing low levels of Bax, we assessed whether there was a difference in Bax expression between the previously treated and untreated patients.12 As anticipated, the mean Bax expression was significantly lower in the previously treated patient group (P = .002; Fig 4B). Therefore, we excluded all patients who had been previously treated and reanalyzed our data set to more accurately delineate the true effect of the polymorphism on Bax protein expression (Fig 4C); this accentuated the difference between the two genotypic subsets (P = .0001). The lower constitutive expression of Bax in the GA polymorphic group was confirmed by Western blot experiments (Fig 5). Interestingly, this lower mean Bax expression was associated with significantly higher Mcl-1/Bax and Bcl-2/Bax ratios in those patients with the GA polymorphism (Fig 6).
Overall Survival in the Treated Patient Cohort
In this present study, we have demonstrated that those patients that possess the G(–248)A polymorphism have lower constitutive Bax protein expression and a shorter overall survival than those patients with a wild-type genotype. Given that our data confirm that Bax expression is modulated by exposure to chemotherapeutic drugs and is involved in apoptotic signaling, we postulated that the shorter median survival was likely to be caused by a poorer response to treatment rather than more aggressive disease. Therefore, we reanalyzed our data set including only those patients who had received previous treatment. The overall survival was significantly shorter in the GA/AA subset (P = .03) with a hazard ratio of 2.0 (Fig 7A). Furthermore, when we analyzed the survival data using date of first treatment rather than date of diagnosis, it was even more significantly different between the two genotypic groups (P = .012; Fig 7B).
DISCUSSION
Bax is a death-promoting protein that has been shown in some animal models and tumor cell lines to be a tumor suppressor that stimulates apoptosis in vivo.28 Recent studies have demonstrated a transcription-independent role for Bax in drug-induced apoptosis mediated by conformational changes in Bax protein before loss of mitochondrial membrane potential and downstream caspase activation.8,16 However, a clear transcription-dependent mechanism of apoptosis induction has also been demonstrated in some cell types.29
In this context, a recent study in 34 CLL patients17 identified a novel single nucleotide polymorphism G(–248)A in the 5'-untranslated (promoter) region of the bax gene that was associated with low Bax protein expression, disease progression beyond Rai stage 0, and a failure to achieve complete response to therapy. In addition, the polymorphism was found at a much higher frequency in cases of CLL than in healthy controls, implicating it in the etiology of the disease. In contrast, our results from a single-center study of 203 patients failed to find a significantly higher frequency of the G(–248)A polymorphism in cases of CLL when compared with healthy controls (P = .15). Neither did we find a bias in the distribution of the polymorphism among CLL cases with more advanced disease at presentation, indicating that it was not involved in promoting a more aggressive clinical course. Our results are in keeping with animal studies in which bax –/– mice, although developing lymphoid hyperplasia, did not develop lymphoid malignancies.28,30 The lack of correlation between the G(–248)A polymorphism and faster lymphocyte doubling time, shorter time to first treatment, or reduced progression-free survival further corroborates this. In addition, the polymorphism was not associated with VH gene mutational status, having a similar incidence in both mutated and unmutated subsets (28 of 73 v eight of 17). When analyzed separately, the presence of the polymorphism resulted in reduced survival in both the VH unmutated and mutated subsets, although this did not reach statistical significance. Similarly, there was no difference in the frequency of the common cytogenetic abnormalities between the two polymorphic groups. Our data did confirm, however, that the polymorphism was associated with significantly reduced Bax protein expression and a significantly shorter overall survival.
We have previously shown that chemoresistance to chlorambucil is mediated, at least in part, by clonal selection of sub-clones with low Bax expression.12 In keeping with this hypothesis, we found that the mean Bax expression was significantly lower in the treated patient group (P = .002). We therefore reanalyzed the data, excluding all previously treated patients. The removal of the confounding effects of prior treatment increased the significance of the lower Bax expression in the polymorphic group (P = .0001). This confirmed the significance of the haplo-insufficiency in determining constitutive Bax protein expression in our CLL cohort.
It is now well established that a small percentage of CLL cases have a familial linkage characterized by anticipation in subsequent generations.24,25 It has been suggested that by investigating these familial cases, it might be possible to elucidate the primary genetic lesion responsible for the development of CLL.26 None of the six patients in our series with a familial pedigree had the G(–248)A polymorphism, indicating that this polymorphism is not an etiologic factor in familial CLL, a finding in keeping with the rest of our data. Moreover, bax seems to exert its effect once the malignant transformation has taken place by playing a critical role in modulating the response to chemotherapy.
In the original study by Saxena et al,17 the presence of the G(–248)A polymorphism seemed to be associated with a reduced response to therapy because 10 patients with the polymorphism subtype failed to achieve a complete response, and in contrast, the two complete responders in their patient cohort had a wild-type genotype. Our study confirms and extends this observation. We found that CLL patients with the polymorphism had a shorter overall survival time, and this was particularly marked in the treated patient subset. The adverse impact of the polymorphism seems to be treatment-related, because when we compared survival times from the date of first treatment rather than the date of diagnosis between the two genotypic groups, there was an even more significant difference.
Identification of this particular bax gene polymorphism should ultimately direct chemotherapeutic intervention toward drugs whose mode of action does not involve the Bcl-2 family of proteins in general and Bax in particular.31,32 It should be noted that this particular polymorphism accounted for only 29% of our CLL patient cohort defined as having low Bax expression. Although the effects of previous treatment can account for some of the residual cases, it seems likely that there are other genetic changes within the bax gene (inherited or acquired) that result in reduced Bax expression, and further studies to identify such changes are presently ongoing in our laboratory.
Although Bax has a transcription-independent role in drug-induced apoptosis, the bax gene is also a primary transcriptional target after p53 activation.33-35 Indeed, some studies have shown that Bax plays a critical role in p53-dependent chemotherapy-induced apoptotic pathways.29 It is therefore possible that the poor prognosis of the G(–248)A polymorphic group is the result of both transcription-dependent and independent factors that contribute toward a blunting of the cellular response to p53 activation and low constitutive expression of Bax, leading to the failure of mitochondrial disruption associated with conformational changes.
Although this single-center study clearly indicates a role for the G(–245)A bax polymorphism, the results should be interpreted with caution. First, the data are retrospective and patients did not receive identical initial chemotherapy. Second, some of the data with regard to important prognostic parameters (ie, cytogenetic abnormalities and VH gene mutation status) are incomplete, so a meaningful multivariate analysis to precisely define the importance of this polymorphism was not possible.
In conclusion, this study demonstrates that the common G(–248)A bax polymorphism does not seem to lead to the development of CLL or affect disease progression but reduces survival by blunting the response to chemotherapy in patients who require treatment. Given the relatively high prevalence of this polymorphism in the normal population and the central role of Bax in mediating chemotherapy-induced cell death in many tumors, genotypic studies of this polymorphism in other human malignancies are strongly indicated. Larger prospective studies using standardized chemotherapy allowing multivariate analysis of all the known important prognostic markers are required to further clarify the importance of this polymorphism.
Authors' Disclosures of Potential Conflicts of Interest
NOTES
J.S. and C.P. jointly share first authorship.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Reed JC: Molecular biology of chronic lymphocytic leukaemia. Semin Oncol 25:11-18, 1998
2. Jurgensmeier JM, Xie Z, Deveraux, et al: Bax directly induces release of cytochrome c from isolated mitochondria. Proc Natl Acad Sci U S A 95:4997-5002, 1998
3. Zamzami N, Brenner C, Marzo I, et al: Subcellular and submitochondrial mode of action of Bcl-2-like proteins. Oncogene 16:2265-2282, 1998
4. Martinou JC, Green DR: Breaking the mitochondrial barrier. Nat Rev Mol Cell Biol 2:63-67, 2001
5. van Loo G, Saelens X, Matthijssens F, et al: Caspases are not localized in mitochondria during life or death. Cell Death Differ 9:1207-1211, 2002
6. Janumayan YM, Sansam CG, Chattopadhyay A, et al: Bcl-x(L)/ Bcl-2 coordinately regulates apoptosis, cell arrest and cell cycle entry. EMBO J 22:5459-5470, 2003
7. Pepper C, Thomas A, Hoy T, et al: Antisense-mediated suppression of Bcl-2 highlights its pivotal role in failed apoptosis in B-cell chronic lymphocytic leukaemia. Br J Haematol 107:611-615, 1999
8. Bellosillo B, Villamor N, Lopez-Guillermo A, et al: Spontaneous and drug-induced apoptosis is mediated by confirmational changes of Bax and Bak in B-cell chronic lymphocytic leukemia. Blood 100:1810-1816, 2002
9. Kitada S, Andersen J, Akar S, et al: Expression of apoptosis regulating proteins in chronic lymphocytic leukemia: Correlations with in vitro and in vivo chemoresponses. Blood 91:3379-3389, 1998
10. Robertson LE, Plunkett W, McConnell K, et al: Bcl-2 expression in chronic lymphocytic leukaemia and its correlation with the induction of apoptosis and clinical outcome. Leukemia 10:456-459, 1996
11. Pepper C, Thomas A, Hoy T, et al: Antisense oligonucleotides complementary to Bax transcripts reduce the susceptibility of B-cell chronic lymphocytic leukaemia cells to apoptosis in a Bcl-2 independent manner. Leuk Lymphoma 43:2003-2009, 2002
12. Pepper C, Hoy T, Thomas A, et al: Chlorambucil resistance in B-cell chronic lymphocytic leukaemia is mediated through failed Bax induction and selection of high Bcl-2 expressing subclones. Br J Haematol 104:581-588, 1999
13. Faderl S, Keating MJ, Do KA, et al: Expression profile of 11 proteins and their prognostic significance in patients with chronic lymphocytic leukaemia (CLL). Leukemia 16:1045-1052, 2002
14. Pepper C, Hoy T, Bentley P: Elevated Bcl-2/Bax are a consistent feature of apoptosis resistance in B-cell chronic lymphocytic leukaemia and are correlated with in vivo chemoresistance. Leuk Lymphoma 28:355-361, 1998
15. Bannerji R, Kitada S, Flinn IW, et al: Apoptotic-regulatory and complement-protecting protein expression in chronic lymphocytic leukaemia: Relationship to in vivo Rituximab resistance. J Clin Oncol 21:1466-1471, 2003
16. Dewson G, Snowden RT, Almond JB, et al: Conformational change and mitochondrial translocation of Bax accompany proteasome inhibitor-induced apoptosis of chronic lymphocytic leukaemia cells. Oncogene 22:2643-2654, 2003
17. Saxena A, Moshynska O, Sankaran K, et al: Association of a novel single nucleotide polymorphism, G(-248)A, in the 5'-UTR of Bax gene in chronic lymphocytic leukemia with disease progression and treatment resistance. Cancer Lett 187:199-205, 2002
18. Binet JL, Augier A, Dighiero G, et al: A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer 48:198-206, 1981
19. Cheson BD, Bennett JM, Grever M, et al: National cancer Institute-sponsored Working Group guidelines for chronic lymphocytic leukaemia: Revised guidelines for diagnosis and treatment. Blood 87:4990-4997
20. Miller SA, Dykes DD, Polesky HF: A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 16:1215-1222, 1988
21. Hamblin TJ, Davis Z, Gardiner A, et al: Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood 94:1848-1854, 1999
22. Damle RN, Wasil T, Fais F, et al: Ig V gene mutation status and CD38 expression as a novel prognostic indicators in chronic lymphocytic leukemia. Blood 94:1840-1847, 1999
23. Stankovic T, Stewart GS, Fegan C, et al: Ataxia telangiectasia mutated deficient B-cell chronic lymphocytic leukaemia occurs in the pre-germinal centre cells and results in defective damage response and unrepaired chromosome damage. Blood 99:300-309, 2002
24. Horowitz M, Goode L, Jarvik GP: Anticipation in familial chronic lymphocytic leukemia. Am J Hum Genet 59:990-998, 1996
25. Wiernik PH, Ashwin M, Hu X-P, et al: Anticipation in familial chronic lymphocytic leukaemia. Br J Haematol 113:407-414, 2001
26. Rawstron AC, Yuille MR, Fuller J, et al: Inherited predisposition to CLL is detectable as a subclinical monoclonal B-lymphocyte expansion. Blood 100:2289-2291, 2002
27. Parker SL, Tong T, Bolden S, et al: Cancer statistics. CA Cancer J Clin. 47:5-27, 1997
28. Yin C, Knudson CM, Korsmeyer SJ, et al: Bax suppresses tumorigenesis and stimulates apoptosis in vivo. Nature 637-640, 1997
29. Zhang L, Yu J, Park BH, et al: Role of Bax in the apoptotic response to anticancer agents. Science 290:989-992, 2000
30. Knudson CM, Tung KSK, Tourtellotte WG, et al: Bax–deficient mice with lymphoid hyperplasia and male germ cell death. Science 270:96-99, 1995
31. Pedersen IM, Kitada S, Schimmer A, et al: The triterpenoid CDDO induces apoptosis in refractory CLL B cells. Blood 100:2965-2972, 2002
32. Pepper C, Thomas A, Hoy T, et al: The vitamin D3 analog EB1089 induces apoptosis via a p53-independent mechanism involving p38 MAP kinase activation and suppression of ERK activity in B-cell chronic lymphocytic leukaemia cells in vitro. Blood 101:2454-2460, 2003
33. Miyashita T, Reed JC: Tumor suppressor p53 is a direct transcriptional activator of human BAX gene. Cell 80:293-299, 1995
34. Oltvai ZN, Milliman CL, Korsmeyer SJ: Bcl-2 heterodimerizes in vivo with a conserved homolog, Bax, that accelerates programmed cell death. Cell 74:609-619, 1993
35. Selvakumaran M, Lin H-K, Miyashita T, et al: Immediate early up-regulation of bax expression by p53 but not TGF beta-1: A paradigm for distinct apoptotic pathways. Oncogene 9:1791-1798, 1994(Jane Starczynski, Chris P)
Institute for Cancer Research, University of Birmingham, Birmingham
Department of Haematology, University of Wales College of Medicine, Heath Park, Cardiff
Department of Haematology, Llandough Hospital, Penarth, Vale of Glamorgan, United Kingdom
ABSTRACT
PATIENTS AND METHODS: The frequency of the novel polymorphism, G(–248)A, in the promoter region of the bax gene and bax protein expression was assessed in 203 CLL patients. The results were correlated with clinical outcome.
RESULTS: The polymorphism was found in 23% of the CLL cohort and 15% of normal controls with no significant difference in allele frequency between the two groups (P = .15). It was associated with lower Bax protein expression and a shorter overall survival, especially in the treated patient group (P = .03). Furthermore, the adverse impact of the polymorphism was accentuated when comparing survival from the date of first treatment rather than diagnosis (P = .012). No significant difference in age at diagnosis, stage of disease at presentation, lymphocyte doubling time, time to first treatment, or progression-free survival were observed.
CONCLUSION: The presence of this single nucleotide polymorphism in CLL critically influences the response to treatment and overall survival. Given the relatively high prevalence of this polymorphism in the normal population, further prospective studies in CLL and other human malignancies are indicated.
INTRODUCTION
Over the last decade, experimental evidence has pointed to a key role for the Bcl-2 family not only in disease progression but also in determining response to therapy and clinical outcome in CLL.7-12 Indeed, Bcl-2 expression has been shown to be an independent prognostic factor and the Bcl-2/Bax ratio and the Mcl-1/Bax ratio seem to be important in determining both in vitro and in vivo response to chemotherapeutic drugs in CLL.3-15 Bax has recently been shown to undergo conformational change in response to various chemotherapy agents and seems to play a pivotal role in drug-induced apoptosis in CLL cells.8,16 For many years, mutations and polymorphisms in the bax gene were not investigated for their potential role in the pathogenesis of CLL. However, a recent study of 34 CLL patients identified the presence of a single nucleotide polymorphism G(–248)A in the 5'-untranslated (promoter) region (UTR) of the bax gene.17 This polymorphism was found in 69% of stage I to IV patients, compared with only 5.5% of stage A0 patients and only 4% of healthy controls and was associated with a failure to achieve complete response to therapy. We therefore sought to extend these preliminary findings in a much larger single-center study to further define the role of this polymorphism in CLL.
PATIENTS AND METHODS
Indications for treatment were consistent with published guidelines, notably troublesome lymphadenopathy or hepato-splenomegaly, constitutional symptoms, or developing cytopenias.19 Patients requiring treatment were entered in the prevailing Medical Research Council (MRC) trial and hence received differing initial therapies. In the MRC CLL III trial, patients were randomly assigned to chlorambucil versus epirubicin/chlorambucil. The MRC CLL IV trial randomly assigned patients between chlorambucil versus fludarabine versus fludarabine/cyclophosphamide. In the MRC CLL pilot study, patients received fludarabine to maximal response followed by an autologous stem-cell transplantation. Salvage therapies depended on the initial therapy given but typically included fludarabine (with and without cyclophosphamide); cyclophosphamide, doxorubicin, vincristine, and prednisolone; alemtuzumab; and allogeneic stem-cell transplantation. The bax polymorphism analyses were largely performed retrospectively, and in no patients was the therapeutic decision dependent on the result. Patients who required treatment received a median of three different chemotherapy regimens.
DNA Isolation and Bax 5'-Untranslated (promoter) Region G(–248)A Polymerase Chain Reaction Analysis
DNA was extracted from peripheral blood taken from CLL patients and healthy volunteers using a standard salting-out method.20 Analysis of the 5'-untranslated (promoter) region polymorphism was undertaken using the following method. A 280–base pair fragment of the Bax 5'-untranslated (promoter) region was amplified using the following primer pair: 5'-TTAGAGACAAGCCTGGGCGT-3' and 5'-CAATGAGCATCTCCCGATAA-3'. The polymerase chain reaction (PCR) mix contained 22.5 μL of Reddy Mix (ABgene, Epsom, United Kingdom), with 3 mmol/L MgCl2 and 5 ng of each primer. After initial denaturation at 94°C for 5 minutes, 35 cycles of amplification were performed, consisting of denaturation at 94°C for 30 seconds, annealing at 48°C for 30 seconds, and extension at 72°C for 30 seconds, with a final extension step of 72°C for 7 minutes. PCR was performed on a GeneAmp PCR System 9700 (Applied Biosystems, Warrington, United Kingdom). A 12-μL aliquot of each PCR product was digested with 1 unit of Tau 1 restriction enzyme (MBI Fermentas, Helena BioSciences, Sunderland, UK) at 55°C for 4 hours. The Tau 1 enzyme recognizes the sequence GCSGC, and the presence of the A allele results in the sequence ACSGC, thus abolishing the restriction site. The digested products were electrophoresed on a 2% agarose gel for 40 minutes at 100 V, stained with ethidium bromide, and visualized using ultraviolet. The Tau 1 restriction enzyme digest produced three possible band patterns (Fig 1A). The normal allele was completely digested to give two bands, 234bp and 46bp, the mutant allele was uncut and gave a single band of 280bp, and all three bands were present in patients showing heterozygosity of the G(–248)A polymorphism.
To confirm whether this was a single nucleotide polymorphism or a point mutation, we also performed the analysis on DNA freshly extracted from buccal mucosa and compared the results with those obtained from the CLL samples. Buccal mouthwashes were obtained from 23 CLL patients. DNA was extracted from the mouthwashes using the QIAmp DNA Mini kit (Qiagen Ltd, Crawley, West Sussex, United Kingdom).
Bax Sequence Analysis
PCR products were directly sequenced using Big Dye Terminators version 3.0 (Applied Biosystems) on a 310 DNA sequencer (Applied Biosystems) according to the manufacturer's instructions. The Bax sequences were compared with available germline sequences for the bax gene, accession number U17193. The sequence analysis is shown in Figure 1B.
VH Gene Mutation Analysis
The mutational status of the immunoglobulin VH is an important prognostic indicator.21,22 RNA was extracted from the patient samples using Qiagen RNeasy mini kit (Qiagen). Complementary DNA (cDNA) was synthesized using the Reverse-iT kit (ABgene). The VH gene mutational status of 141 CLL patients was analyzed according to the method previously described.23 The resulting PCR products were sequenced using BIG dye terminator sequencing kit version 3 (Applied Biosystems), and the sequences were analyzed using Immunoglobulin BLAST (National Center for Biotechnology Information, Bethesda, MD). The sequences with a germline homology of 98% or higher were regarded as unmutated, and those with less than 98% were regarded as mutated.
Genetic Analysis
Peripheral blood samples from patients with advanced-stage disease underwent fluorescent in situ hybridization analyses looking for trisomy 12, 6q, 11q, 13q, and 17 p deletions (Vysis, London, United Kingdom) and/or karyotype analysis after 3 to 5 days of incubation with TPA (12-myristate 13-acetate) stimulation.
Analysis of Bcl-2, Bax, and Mcl-1 Protein Expression
Fresh peripheral blood samples from 122 consecutive patients attending the CLL clinic over a 9-month period were analyzed by triple immunofluorescent staining using a Bax antibody in conjunction with CD5 and CD19 antibodies. Briefly, 1 x 106 cells were incubated with 10 μL of both anti-CD5 phycoerythrin-conjugated and anti-CD19 phycoerythrin cyanine 5-conjugated antibodies or isotype-matched controls (DAKO, Ely, United Kingdom). The cells were fixed using a commercially available kit (DAKO) and resuspended in a permeabilization solution together with one of the following antibodies: 7 μL of anti-Bcl-2 fluorescein isothiocyanate (FITC)–conjugated antibody (DAKO), 8 μL of anti-Bax FITC-conjugated antibody (Santa Cruz Biotechnology, Santa Cruz, CA), 8 μL of anti-Mcl-1 antibody (Santa Cruz Biotechnology), or an isotype-matched control (DAKO). The Mcl-1 labeled cells were subsequently subjected to secondary labeling with a FITC-conjugated antibody (DAKO). After a final washing step, the cells were resuspended in 0.5 mL of 1% paraformaldehyde before flow cytometric analysis using a FACScan flow cytometer (Becton Dickinson, San Jose, CA). From each sample, at least 10,000 cells were analyzed, and nonspecific binding was excluded by gating using the isotype control antibody. Gating of the CD5+/CD19+ cells was performed in the analyses in order to the specific protein expression in leukemic CLL cells. The mean fluorescent intensity was calculated for each individual protein using WINMDI software (J. Trotter, Scripps Research Institute, La Jolla, CA) and these values were then converted to molecules of equivalent soluble fluorochrome using a calibration curve to standardize the data.
Immunoblotting
A total of 3 x 107 CLL cells derived from previously untreated patients with either the wild-type (GG) or polymorphic (GA) bax genotype were lysed in 1 mL of lysis buffer containing 1% Triton X-100. Protein concentration was determined by Bradford assay (Bio-Rad, Hemel Hempsted, United Kingdom). Each sample (12 μg of protein) was separated on a 4% to 12% Bis-Tris acrylamide gel and transferred to polyvinylidene fluoride membrane (Bio-Rad). Western blots were incubated with a mouse monoclonal antibody against Bax (clone 4F11; Beckman Coulter, High Wycombe, United Kingdom), in phosphate-buffered saline with 4% milk. The secondary antibody was a horseradish peroxidase–conjugated sheep antimouse immunoglobulin G (Amersham Biosciences, Chalfont St Giles, United Kingdom), and detection was performed by an enhanced chemiluminescence method (Amersham).
Clinical/Laboratory Prognostic Parameters
To assess the role of the G(–248)A 5'-untranslated (promoter) region polymorphism of the bax gene, comparison was made between various clinical/laboratory parameters including expression of Bax protein, VH gene mutational status, Binet staging, age at presentation, lymphocyte doubling time (> or < 12 months), time to first treatment, progression-free survival, and overall survival.
Statistical Analysis
Patients were grouped into two subsets depending on genotype (GG v GA/AA), and statistical analysis was performed using parametric and nonparametric methods depending on whether the data obeyed a Gaussian distribution. Age at diagnosis and Bax protein expression were analyzed using the unpaired Student's t test. Comparison of the allele frequency of the G(–248)A polymorphism, VH mutational status, lymphocyte doubling time, and patient sex were analyzed using the Fisher's exact test, and all other parameters were analyzed using the Mann-Whitney test, except for the stage of disease at presentation, where the {chi}2 test was used. Survival analysis was calculated according to the Kaplan-Meier method. All statistical analyses were performed using the Graphpad Prism 3.0 software (Graphpad Software Inc, San Diego, CA).
RESULTS
Studies over recent years have identified that in up to 5% of CLL patients, there is a familial tendency.24-26 Given the central role of Bax in the pathogenesis of CLL and the identification of this potentially etiologic hereditary factor, we sought to study further this particular bax polymorphism in a large single-center study. Within the CLL cohort, there were six patients who were part of a familial CLL pedigree. All of these patients had the GG genotype, indicating that the G(–248)A 5'-untranslated (promoter) region polymorphism is not the hereditary factor responsible for familial CLL.
Clinical Features and Prognostic Markers
In contrast with the previously published data,17 we found no increase in frequency of the A allele in CLL cases with more advanced disease as defined by the Binet classification system (P = .62). There was no significant difference in mean age at diagnosis or median follow-up time between the GA/AA and GG genotypic groups (P = .62 and .39, respectively). Neither was there a significant difference in lymphocyte doubling time (> or < 12 months), the time to first treatment, or progression-free survival (P = .32, .38, and .42, respectively). However, the overall survival of the GA/AA genotypic group was significantly shorter than the GG genotypic group when analyzing only CLL-related deaths (P = .05). The data are summarized in Table 2, and the overlaid Kaplan-Meier survival curves for the GG patient group and the combined GA/AA patient groups are shown in Figure 2.
The VH gene mutational status of CLL patients is an important prognostic marker with those patients with unmutated VH genes having significantly shorter survival times. This was confirmed in this study with the unmutated patient subset having a significantly shorter survival time than the mutated subset (P = .002; Fig 3A). However, we found no bias in distribution in terms of VH gene mutational status between the bax genotypic subgroups (GG v GA/AA); 28 (28%) of 101 CLL patients with the GG genotype and eight (32%) of 25 patients with the GA/AA genotypes had unmutated VH genes (P = .81). Furthermore, subdivision of the mutated and unmutated groups by genotype (GG v GA/AA) revealed no significant difference in overall survival between the wild-type and polymorphic groups in either the mutated subset (P = .24) or the unmutated subset (P = .17; Fig 3B).
Genetic analysis was available in 66 patients with the GG genotype and 28 patients with the GA/AA genotype. Comparisons between the GG and GA/AA showed no significant difference in the frequency of a normal karyotype (32 of 66 v 13 of 28 patients), 11q deletions (15 of 66 v six of 28 patients), 17p deletions (four of 66 v two of 28 patients), and 13q deletions; trisomy 12 and 6q deletions were present in 15 of 66 and seven of 28 patients, respectively.
Ex Vivo Bax Protein Expression
Previous studies have shown the importance of Bax protein expression, high Bcl-2/Bax ratios, and Mcl-1/Bax ratios in determining both in vitro and in vivo response to chemotherapy.14,15 In this study, CLL patient samples (122 of 203) were assessed for ex vivo expression of Bax to determine whether any of these parameters were significantly different between the bax genotypic subsets (GA/AA v GG). Bax protein expression was significantly lower in those patients with a GA/AA genotype (P = .05; Fig 4A). Because we have previously shown that drug resistance may be, at least in part, mediated by selection of subclones expressing low levels of Bax, we assessed whether there was a difference in Bax expression between the previously treated and untreated patients.12 As anticipated, the mean Bax expression was significantly lower in the previously treated patient group (P = .002; Fig 4B). Therefore, we excluded all patients who had been previously treated and reanalyzed our data set to more accurately delineate the true effect of the polymorphism on Bax protein expression (Fig 4C); this accentuated the difference between the two genotypic subsets (P = .0001). The lower constitutive expression of Bax in the GA polymorphic group was confirmed by Western blot experiments (Fig 5). Interestingly, this lower mean Bax expression was associated with significantly higher Mcl-1/Bax and Bcl-2/Bax ratios in those patients with the GA polymorphism (Fig 6).
Overall Survival in the Treated Patient Cohort
In this present study, we have demonstrated that those patients that possess the G(–248)A polymorphism have lower constitutive Bax protein expression and a shorter overall survival than those patients with a wild-type genotype. Given that our data confirm that Bax expression is modulated by exposure to chemotherapeutic drugs and is involved in apoptotic signaling, we postulated that the shorter median survival was likely to be caused by a poorer response to treatment rather than more aggressive disease. Therefore, we reanalyzed our data set including only those patients who had received previous treatment. The overall survival was significantly shorter in the GA/AA subset (P = .03) with a hazard ratio of 2.0 (Fig 7A). Furthermore, when we analyzed the survival data using date of first treatment rather than date of diagnosis, it was even more significantly different between the two genotypic groups (P = .012; Fig 7B).
DISCUSSION
Bax is a death-promoting protein that has been shown in some animal models and tumor cell lines to be a tumor suppressor that stimulates apoptosis in vivo.28 Recent studies have demonstrated a transcription-independent role for Bax in drug-induced apoptosis mediated by conformational changes in Bax protein before loss of mitochondrial membrane potential and downstream caspase activation.8,16 However, a clear transcription-dependent mechanism of apoptosis induction has also been demonstrated in some cell types.29
In this context, a recent study in 34 CLL patients17 identified a novel single nucleotide polymorphism G(–248)A in the 5'-untranslated (promoter) region of the bax gene that was associated with low Bax protein expression, disease progression beyond Rai stage 0, and a failure to achieve complete response to therapy. In addition, the polymorphism was found at a much higher frequency in cases of CLL than in healthy controls, implicating it in the etiology of the disease. In contrast, our results from a single-center study of 203 patients failed to find a significantly higher frequency of the G(–248)A polymorphism in cases of CLL when compared with healthy controls (P = .15). Neither did we find a bias in the distribution of the polymorphism among CLL cases with more advanced disease at presentation, indicating that it was not involved in promoting a more aggressive clinical course. Our results are in keeping with animal studies in which bax –/– mice, although developing lymphoid hyperplasia, did not develop lymphoid malignancies.28,30 The lack of correlation between the G(–248)A polymorphism and faster lymphocyte doubling time, shorter time to first treatment, or reduced progression-free survival further corroborates this. In addition, the polymorphism was not associated with VH gene mutational status, having a similar incidence in both mutated and unmutated subsets (28 of 73 v eight of 17). When analyzed separately, the presence of the polymorphism resulted in reduced survival in both the VH unmutated and mutated subsets, although this did not reach statistical significance. Similarly, there was no difference in the frequency of the common cytogenetic abnormalities between the two polymorphic groups. Our data did confirm, however, that the polymorphism was associated with significantly reduced Bax protein expression and a significantly shorter overall survival.
We have previously shown that chemoresistance to chlorambucil is mediated, at least in part, by clonal selection of sub-clones with low Bax expression.12 In keeping with this hypothesis, we found that the mean Bax expression was significantly lower in the treated patient group (P = .002). We therefore reanalyzed the data, excluding all previously treated patients. The removal of the confounding effects of prior treatment increased the significance of the lower Bax expression in the polymorphic group (P = .0001). This confirmed the significance of the haplo-insufficiency in determining constitutive Bax protein expression in our CLL cohort.
It is now well established that a small percentage of CLL cases have a familial linkage characterized by anticipation in subsequent generations.24,25 It has been suggested that by investigating these familial cases, it might be possible to elucidate the primary genetic lesion responsible for the development of CLL.26 None of the six patients in our series with a familial pedigree had the G(–248)A polymorphism, indicating that this polymorphism is not an etiologic factor in familial CLL, a finding in keeping with the rest of our data. Moreover, bax seems to exert its effect once the malignant transformation has taken place by playing a critical role in modulating the response to chemotherapy.
In the original study by Saxena et al,17 the presence of the G(–248)A polymorphism seemed to be associated with a reduced response to therapy because 10 patients with the polymorphism subtype failed to achieve a complete response, and in contrast, the two complete responders in their patient cohort had a wild-type genotype. Our study confirms and extends this observation. We found that CLL patients with the polymorphism had a shorter overall survival time, and this was particularly marked in the treated patient subset. The adverse impact of the polymorphism seems to be treatment-related, because when we compared survival times from the date of first treatment rather than the date of diagnosis between the two genotypic groups, there was an even more significant difference.
Identification of this particular bax gene polymorphism should ultimately direct chemotherapeutic intervention toward drugs whose mode of action does not involve the Bcl-2 family of proteins in general and Bax in particular.31,32 It should be noted that this particular polymorphism accounted for only 29% of our CLL patient cohort defined as having low Bax expression. Although the effects of previous treatment can account for some of the residual cases, it seems likely that there are other genetic changes within the bax gene (inherited or acquired) that result in reduced Bax expression, and further studies to identify such changes are presently ongoing in our laboratory.
Although Bax has a transcription-independent role in drug-induced apoptosis, the bax gene is also a primary transcriptional target after p53 activation.33-35 Indeed, some studies have shown that Bax plays a critical role in p53-dependent chemotherapy-induced apoptotic pathways.29 It is therefore possible that the poor prognosis of the G(–248)A polymorphic group is the result of both transcription-dependent and independent factors that contribute toward a blunting of the cellular response to p53 activation and low constitutive expression of Bax, leading to the failure of mitochondrial disruption associated with conformational changes.
Although this single-center study clearly indicates a role for the G(–245)A bax polymorphism, the results should be interpreted with caution. First, the data are retrospective and patients did not receive identical initial chemotherapy. Second, some of the data with regard to important prognostic parameters (ie, cytogenetic abnormalities and VH gene mutation status) are incomplete, so a meaningful multivariate analysis to precisely define the importance of this polymorphism was not possible.
In conclusion, this study demonstrates that the common G(–248)A bax polymorphism does not seem to lead to the development of CLL or affect disease progression but reduces survival by blunting the response to chemotherapy in patients who require treatment. Given the relatively high prevalence of this polymorphism in the normal population and the central role of Bax in mediating chemotherapy-induced cell death in many tumors, genotypic studies of this polymorphism in other human malignancies are strongly indicated. Larger prospective studies using standardized chemotherapy allowing multivariate analysis of all the known important prognostic markers are required to further clarify the importance of this polymorphism.
Authors' Disclosures of Potential Conflicts of Interest
NOTES
J.S. and C.P. jointly share first authorship.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Reed JC: Molecular biology of chronic lymphocytic leukaemia. Semin Oncol 25:11-18, 1998
2. Jurgensmeier JM, Xie Z, Deveraux, et al: Bax directly induces release of cytochrome c from isolated mitochondria. Proc Natl Acad Sci U S A 95:4997-5002, 1998
3. Zamzami N, Brenner C, Marzo I, et al: Subcellular and submitochondrial mode of action of Bcl-2-like proteins. Oncogene 16:2265-2282, 1998
4. Martinou JC, Green DR: Breaking the mitochondrial barrier. Nat Rev Mol Cell Biol 2:63-67, 2001
5. van Loo G, Saelens X, Matthijssens F, et al: Caspases are not localized in mitochondria during life or death. Cell Death Differ 9:1207-1211, 2002
6. Janumayan YM, Sansam CG, Chattopadhyay A, et al: Bcl-x(L)/ Bcl-2 coordinately regulates apoptosis, cell arrest and cell cycle entry. EMBO J 22:5459-5470, 2003
7. Pepper C, Thomas A, Hoy T, et al: Antisense-mediated suppression of Bcl-2 highlights its pivotal role in failed apoptosis in B-cell chronic lymphocytic leukaemia. Br J Haematol 107:611-615, 1999
8. Bellosillo B, Villamor N, Lopez-Guillermo A, et al: Spontaneous and drug-induced apoptosis is mediated by confirmational changes of Bax and Bak in B-cell chronic lymphocytic leukemia. Blood 100:1810-1816, 2002
9. Kitada S, Andersen J, Akar S, et al: Expression of apoptosis regulating proteins in chronic lymphocytic leukemia: Correlations with in vitro and in vivo chemoresponses. Blood 91:3379-3389, 1998
10. Robertson LE, Plunkett W, McConnell K, et al: Bcl-2 expression in chronic lymphocytic leukaemia and its correlation with the induction of apoptosis and clinical outcome. Leukemia 10:456-459, 1996
11. Pepper C, Thomas A, Hoy T, et al: Antisense oligonucleotides complementary to Bax transcripts reduce the susceptibility of B-cell chronic lymphocytic leukaemia cells to apoptosis in a Bcl-2 independent manner. Leuk Lymphoma 43:2003-2009, 2002
12. Pepper C, Hoy T, Thomas A, et al: Chlorambucil resistance in B-cell chronic lymphocytic leukaemia is mediated through failed Bax induction and selection of high Bcl-2 expressing subclones. Br J Haematol 104:581-588, 1999
13. Faderl S, Keating MJ, Do KA, et al: Expression profile of 11 proteins and their prognostic significance in patients with chronic lymphocytic leukaemia (CLL). Leukemia 16:1045-1052, 2002
14. Pepper C, Hoy T, Bentley P: Elevated Bcl-2/Bax are a consistent feature of apoptosis resistance in B-cell chronic lymphocytic leukaemia and are correlated with in vivo chemoresistance. Leuk Lymphoma 28:355-361, 1998
15. Bannerji R, Kitada S, Flinn IW, et al: Apoptotic-regulatory and complement-protecting protein expression in chronic lymphocytic leukaemia: Relationship to in vivo Rituximab resistance. J Clin Oncol 21:1466-1471, 2003
16. Dewson G, Snowden RT, Almond JB, et al: Conformational change and mitochondrial translocation of Bax accompany proteasome inhibitor-induced apoptosis of chronic lymphocytic leukaemia cells. Oncogene 22:2643-2654, 2003
17. Saxena A, Moshynska O, Sankaran K, et al: Association of a novel single nucleotide polymorphism, G(-248)A, in the 5'-UTR of Bax gene in chronic lymphocytic leukemia with disease progression and treatment resistance. Cancer Lett 187:199-205, 2002
18. Binet JL, Augier A, Dighiero G, et al: A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer 48:198-206, 1981
19. Cheson BD, Bennett JM, Grever M, et al: National cancer Institute-sponsored Working Group guidelines for chronic lymphocytic leukaemia: Revised guidelines for diagnosis and treatment. Blood 87:4990-4997
20. Miller SA, Dykes DD, Polesky HF: A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 16:1215-1222, 1988
21. Hamblin TJ, Davis Z, Gardiner A, et al: Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood 94:1848-1854, 1999
22. Damle RN, Wasil T, Fais F, et al: Ig V gene mutation status and CD38 expression as a novel prognostic indicators in chronic lymphocytic leukemia. Blood 94:1840-1847, 1999
23. Stankovic T, Stewart GS, Fegan C, et al: Ataxia telangiectasia mutated deficient B-cell chronic lymphocytic leukaemia occurs in the pre-germinal centre cells and results in defective damage response and unrepaired chromosome damage. Blood 99:300-309, 2002
24. Horowitz M, Goode L, Jarvik GP: Anticipation in familial chronic lymphocytic leukemia. Am J Hum Genet 59:990-998, 1996
25. Wiernik PH, Ashwin M, Hu X-P, et al: Anticipation in familial chronic lymphocytic leukaemia. Br J Haematol 113:407-414, 2001
26. Rawstron AC, Yuille MR, Fuller J, et al: Inherited predisposition to CLL is detectable as a subclinical monoclonal B-lymphocyte expansion. Blood 100:2289-2291, 2002
27. Parker SL, Tong T, Bolden S, et al: Cancer statistics. CA Cancer J Clin. 47:5-27, 1997
28. Yin C, Knudson CM, Korsmeyer SJ, et al: Bax suppresses tumorigenesis and stimulates apoptosis in vivo. Nature 637-640, 1997
29. Zhang L, Yu J, Park BH, et al: Role of Bax in the apoptotic response to anticancer agents. Science 290:989-992, 2000
30. Knudson CM, Tung KSK, Tourtellotte WG, et al: Bax–deficient mice with lymphoid hyperplasia and male germ cell death. Science 270:96-99, 1995
31. Pedersen IM, Kitada S, Schimmer A, et al: The triterpenoid CDDO induces apoptosis in refractory CLL B cells. Blood 100:2965-2972, 2002
32. Pepper C, Thomas A, Hoy T, et al: The vitamin D3 analog EB1089 induces apoptosis via a p53-independent mechanism involving p38 MAP kinase activation and suppression of ERK activity in B-cell chronic lymphocytic leukaemia cells in vitro. Blood 101:2454-2460, 2003
33. Miyashita T, Reed JC: Tumor suppressor p53 is a direct transcriptional activator of human BAX gene. Cell 80:293-299, 1995
34. Oltvai ZN, Milliman CL, Korsmeyer SJ: Bcl-2 heterodimerizes in vivo with a conserved homolog, Bax, that accelerates programmed cell death. Cell 74:609-619, 1993
35. Selvakumaran M, Lin H-K, Miyashita T, et al: Immediate early up-regulation of bax expression by p53 but not TGF beta-1: A paradigm for distinct apoptotic pathways. Oncogene 9:1791-1798, 1994(Jane Starczynski, Chris P)