Welcome international healthcare professionals

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site.

You are here

A comprehensive study of TP53 mutations in chronic lymphocytic leukemia: Analysis of 1287 diagnostic and 1148 follow-up CLL samples

Leukemia Research, 7, 35, pages 889 - 898

Abstract

TP53 plays a pivotal role in the process of DNA repair and apoptosis. In 10–20% of patients with chronic lymphocytic leukemia (CLL), the TP53 pathway is affected. In this study, we analyzed the TP53 mutation status in 2435 consecutive CLL samples, including 1287 diagnostic samples and 1148 samples during follow-up, using FASAY (Functional Analysis of Separated Alleles in Yeast) and direct sequencing. In a cohort of 1287 diagnostic CLL samples, we identified 237 cases with TP53 variants, including mutations, temperature-sensitive variants, deletions, insertions and aberrant splicing variants (18.4%). In 1148 follow-up samples, no TP53 clonal evolution was observed.

Keywords: TP53, CLL, FASAY.

1. Introduction

TP53 is a transcription factor that plays a pivotal role in the process of DNA repair and apoptosis [1] and [2]. In 10–20% of patients with chronic lymphocytic leukemia (CLL), the TP53 pathway is affected by genomic TP53 mutations, aberrant mRNA splicing or by the dysfunction of cooperating protein partners [3], [4], [5], [6], [7], [8], [9], [10], and [11]. TP53 mutations/variants are diverse and affect most of the coding sequence of TP53 [12] . Though there are a few known hot-spot mutations with proven biological functions, there are still a number of TP53 mutations, for which biological functions have not yet been assessed. The majority of TP53 mutations/variants are inactivating, but some of them have unusual characteristics (e.g. temperature sensitivity) [13] , or might represent aberrant splicing isoforms [6], [7], [9], and [10] expressed at variable levels over time. Moreover, it has been reported that in CLL, chemotherapy induces the evolution of TP53 mutated subclones that might lead to progression of the disease [14] .

In our study (from January 2005 to December 2009), we analyzed the TP53 mutation status in 2435 consecutive CLL samples. The aim of the study was to address the mutational status of the TP53 gene, modes of inactivation/modulation of the TP53 gene and the possibility of TP53 clonal evolution driven by chemotherapy in a large cohort of CLL patients, including 1287 diagnostic samples and 1148 samples during the follow-up. The follow-up samples comprised peripheral blood and/or bone marrow samples taken at regular intervals of 1–5 months (median 3 months) from 181 CLL patients, for whom clone-specific assays to monitor minimal residual disease (MRD) had been prepared.

In a cohort of 1287 diagnostic CLL samples, we identified 237 cases harboring TP53 variants (18.4%), including point mutations, insertions, deletions, temperature-sensitive mutations and aberrant splicing variants. The majority of TP53 variants have fully penetrant transactivation-defective mutant phenotypes in vitro, as assessed by FASAY (functional analysis of separated alleles in yeast) [15] . Nine mutants displayed temperature-sensitive properties. A novel TP53 splicing variant delta ex6 [16] was molecularly characterized. In 1148 follow-up samples analyzed for MRD levels in 181 CLL patients undergoing chemo/immunotherapy or bone marrow transplant, no TP53 clonal evolution was observed, contrary to the data published [14] . The following text summarizes the TP53 data analyzed during this five-year project.

2. Materials and methods

2.1. Patients

Beginning in January 2005, 1287 consecutive diagnostic CLL samples and 1148 follow-up CLL samples were enrolled in the study. All patients conformed to the NCIWG criteria for the diagnosis of CLL and were treated according to the institutional protocols of the referring hospitals. The treatment modalities were diverse and involved intensive chemo/immunotherapy (FC = Fludarabine 25 mg/m2 iv D1–3 + cyclophosphamide 250 mg/m2 iv D1–3 or Fludarabine 24 mg/m2 po D1–5 + cyclophosphamide 150 mg/m2 po D1–5 every 28 days, 6 cycles; FCR = FC + Rituximab 375 mg/m2 iv D0–1, 6 cycles; CHOP = Cyclophosphamide 750 mg/m2 iv D1 + Vincristine 1.4 mg/m2 iv D1 + Doxorubicin 50 mg/m2 iv D1 + Prednisolone 50 mg/m2 po D1–4 every 21 days, 6–8 cycles; R-CHOP = CHOP + Rituximab 375 mg/m2 iv D0–1, 6–8 cycles; COP = Cyclophosphamide 750 mg/m2 iv D1 + Vincristine 1.4 mg/m2 iv D1 + Prednisolone 60 mg/m2 po D1–5, 6 cycles, R-COP = COP + Rituximab 375 mg/m2 iv D0–1, 6 cycles; CAM = Alemtuzumab 30 mg sc 3× week, 12 weeks; F = Fludarabine monotherapy 40 mg/m2 po D1–5 every 28 days, 3–6 cycles; R = Rituximab monotherapy 375 mg/m2 iv; CPA = Cyclophosphamide monotherapy 10–15 mg/kg iv every 14 days; MP = Methylprednisolone pulses: 500 mg iv D1–4; CLB = Chlorambucil 0.4–0.8 mg/kg D every 14 days; Cladribine monotherapy 0.1 mg/kg iv D1–7 every 28 days, 6 cycles) or allogeneic bone marrow transplant.

For 181 individuals, clone-specific assays to monitor MRD were prepared as described elsewhere [16] . During the duration of the project, we tested 1148 successive follow-up samples (median 6.3 samples). The 181 patients had been subjected to different treatment regimens, comprising chemo/immunotherapy (153 individuals) and allogeneic bone marrow transplant (28 individuals). The median follow-up was 439 days (87–1965 days; mean 773 days). Peripheral blood and/or bone marrow were obtained upon written informed consent in accordance with the Declaration of Helsinki. During the MRD follow-up, peripheral blood and/or bone marrow were sampled at 3-month intervals. The study was approved by the local ethical committees.

2.2. Isolation of RNA and preparation of cDNA

To isolate RNA, 10 million Ficoll-separated (Sigma, USA) mononuclear cells (lymphocytosis 70–90%) were subjected to RNA extraction by TriReagent (Sigma, USA) according to the manufacturer's recommendations. After isopropanol precipitation, the RNA pellet was dissolved in 50 μl of RNAase-free PCR-grade water. To synthesize cDNA, a Verso cDNA kit in combination with random hexamers (AB-gene, United Kingdom) was used according to the manufacturer's instructions.

2.3. TP53 PCR and sequencing

cDNA was used as a template for PCR amplification of a TP53 fragment harboring exons 4–9, which includes the DNA-binding domain of TP53, using FailSafe proof-reading DNA polymerase (Epicentre, USA) and FailSafe 2× premix A reaction buffer (Epicentre, USA), with the primers FASAY forward (5′-ccttgccgtcccaagcaatggatgat-3′) and FASAY reverse (5′-accctttttggacttcaggtggctggagt-3′). Both primers were obtained from MWG Biotech, Germany. The PCR profile was as follows: 95 °C for 8 min, followed by 35 cycles of 95 °C for 20 s, 60 °C for 30 s and 72 °C for 45 s. PCR products were separated on a 1.5% agarose gel, cut out and purified using a QIAquick Gel Extraction Kit (Qiagen, Germany). Purified PCR products were directly sequenced using an ABI3130 automatic sequencer (Applied Biosystems, Foster City, CA, USA). Obtained sequences were analyzed using the program Chromas 1.45 (Queensland, Australia) and aligned with the TP53 reference sequence (NM_000546).

2.4. FASAY (functional analysis of separated alleles in yeast)

As described elsewhere [15] , the ADE2 yeast strain yIG397 was grown in YPDA++ medium overnight, reinoculated into fresh YPDA++ and then transformed using the standard LiAC based protocol. Fragments of the TP53 gene covering codons 34–374 (from the end of the transactivation domain through the DNA-binding domain to the nuclear localization signal) were PCR amplified with the primers FASAY forward and FASAY reverse (see above), with cDNA as template. PCR products were directly recombined into the HindIII and StuI (New England Biolabs, UK) linearized and calf intestine phosphatase dephosphorylated vector pSS16 [15] using the recipient yeast strain yIG397 (a kind gift from Drs. Richard Iggo and Jana Smardova). Transformants were plated onto selection media with low adenine. Transformants harboring wild type TP53 were distinguished from mutant TP53 variants on the basis of color: mutant TP53 yielded small red colonies (due to a by-product of an aberrant adenine metabolism), while colonies with wild type TP53 grew white and large.

2.5. Yeast colony PCR and sequencing

Twenty single red colonies were picked from each plate and pooled in a lysis solution consisting of 2 mg/ml Lyticase (Sigma, USA) and 1.2 M sorbitol, and incubated at 37 °C for 3 h. The pSS16 plasmids with recombined TP53 fragments were rescued from yeast colonies using a Charge Switch Plasmid Yeast Mini Kit (Invitrogen, UK). PCR primers, conditions and the PCR profile were as described in Section 2.3 . PCR products were separated by 1.5% agarose gel electrophoresis, cut out and gel-purified using a QIAquick Gel Extraction Kit. Purified PCR products were used as a template for direct sequencing on the ABI3130 sequencing analyzer. In those cases when a clonal mutation was present, an aberrant peak or a frame shift was clearly detectable.

2.6. Cloning the Delta Ex6 splicing variant

We recently identified a novel splicing variant of the TP53 gene, termed Delta Ex6 [16] , lacking the whole coding sequence of exon 6 and harboring alternative exon 9B. The pSS16 vector harboring the whole Delta Ex6 coding sequence was isolated from the respective red yeast colony and re-transformed into chemically competent TOP10 E. coli using the standard heat-shock technique. The pSS16 plasmid × Delta Ex6 was rescued from E. coli, and the Delta Ex6 coding sequence was molecularly subcloned into a pcDNA4/TO vector (Invitrogen, UK) via engineered EcoRI (5′ end) and EcoRV (3′ end) restriction sites. The 5′ end of the Delta Ex6 was further modified by a sequence coding for a 3× FLAG Tag peptide (Sigma, USA) to enable anti-FLAG antibody detection of the transgene by Western Blot. The correctness of this construct was verified by restriction analysis and direct sequencing.

2.7. Site-directed mutagenesis

To construct in vitro the nine temperature-sensitive (t.s.) TP53 variants identified among the 1287 diagnostic CLL samples (V157F, A161T, S215I, V216 M, Y234C, N235S, R283C, K320E, K320T), we used standard megaprimer site-directed mutagenesis [18] . The restriction sites EcoRI (5′ end), Bsu36I and EcoRV (3′ end) were then used to create and subclone the mutated templates into the pcDNA4/TO vector. The correctness of all constructs was verified by restriction analysis and direct sequencing.

2.8. Preparing stable H1299 cell lines harboring the TP53 variants

The constructs pcDNT4/TO, pcDNA4/TO harboring Delta Ex6 TP53, pcDNA4/TO harboring R175H hot spot TP53 mutant, pcDNA4/TO harboring wild type TP53 and pcDNA4/TO harboring each of the nine t.s. TP53 variants (V157F, A161T, S215I, V216M, Y234C, N235S, R283C, K320E, K320T) were transfected into the H1299 cell line (ATCC-CRL-5803) using Lipofectamine 2000 (Invitrogen, UK) according to the manufacturer's recommendations. To assess the transfection efficiency, pEGPF-C1 (Invitrogen, UK) was co-transfected with the respective TP53 construct at a molar ratio of 5:1 (TP53 variant: EGFP). Transfection efficiency of at least 70% was regarded as successful and the cells were exposed to antibiotic selection. After 14 days of 400 μg/ml Zeocine treatment (Invitrogen, UK), hundreds of stable colonies repopulated each cultivation flask. All experiments were done in duplicates. Cells were grown to confluence and harvested into either TriReagent (for mRNA analysis) or Laemmli sample buffer (for Western blot analysis). Expression of the stably integrated TP53 transgenes was verified by quantitative real-time PCR and Western blot analyses.

2.9. Quantitative real-time PCR analysis of the TP53 variants’ mRNA expression

Using the primers TP53 for: agagaccggcgcacagaggaag, TP53 rev: ggctccttcccagcctgggcatc and Human ProbeLibrary probe Nr. 58 (Roche, Germany), a fragment of TP53 comprising exons 8–10 was PCR amplified and quantified using a RotorGene Q (Qiagen, Germany) platform. Beta2-microglobulin was used as the control gene. Primers and probes to quantify Beta2-microglobulin were as follows: B2-MG for: gaggctatccagcgtactcc; B2-MG rev: gatagaaagaccagtccttgctg; B2-MG probe: FAM-aggtttactcacgtcatccagcag-BHQ1. All primers and probes were obtained from MWG Biotech, Germany.

2.10. Western blot analysis of the TP53 variants

H1299 cell lines harboring stably integrated TP53 variants were seeded at a density of 40,000 cells/ml into adherent cultivation flasks and grown in DMEM cultivation medium (PAA, Austria) supplemented with 10% FCS (PAA, Austria), penicillin, streptomycin and glutamine (PAA, Austria). The cells were split every other day and grown to confluence before harvesting. At the time of harvest, H1299 cell lines were gently washed in 1× PBS (PAA, Austria) and directly lysed in 2× Laemmli Sample Buffer supplemented with b-Mercaptoethanol (BioRad, USA). Lysates were denatured at 95 °C for 10 min, chilled on ice and loaded on 8% SDS-PAGE. After electrophoresis, the gels were blotted onto PVDF membrane (BioRad, USA) overnight and probed with the respective antibody. To identify wild type TP53 protein, TP53 bona fide transactivating mutations or t.s. TP53 variants, mouse anti-human TP53 DO-1 antibody (Santa Cruz Biotechnology, USA) in combination with anti-mouse HRP conjugated secondary antibody AMI4404 (Biosource, USA) were used. Because the DO-1 antibody did not recognize the Delta Ex6 TP53 variant, most probably due to aberrant folding of the protein, Delta Ex6 was N’ fused to the 3xFLAG epitope (see above). Specific HRP-conjugated antibody against the 3× FLAG epitope (ProteoQwest FLAG Colorimetric Western Blotting kit, Sigma, USA) enabled detection of Delta Ex6 at the protein level. To visualize the colorimetric HRP reaction, TBM substrate (Sigma, USA) was used.

2.11. Microarray expression profiling of the Delta EX6 TP53 variant

Four stable delta ex6 producing H1299 cell lines, as well as control cell lines (mock-transfected cells H1299 harboring the pcDNA4/TO cloning vector, and H1299 stably transfected with pcDNA4/TO × wild type TP53) were subjected to the Affymetrix GeneChip Human Exon 1.0 ST whole genome expression analysis (Affymetrix, USA). Microarray analyses were done in duplicates for the control cell lines to filter out possible inter-individual biological variables in the in vitro cultivation conditions. Data analysis and detection of differentially expressed and/or spliced genes were performed using the Expression Console (Affymetrix, CA, USA), Bioconductor packages and the R project software ( http://www.r-project.org , http://www.bioconductor.org ) for statistical calculations.

Transcripts showing significant differences in expression between the controls and the Delta Ex6 harboring H1299 cell lines (evaluated using a modified t-test, statistics from the Limma package) were taken into consideration. Raw data from the Affymetrix microarray expression profiling are available upon request.

For quality assessment and quality control, transcripts showing high expression variability across all arrays were selected. The criterion used was a minimal inter-quartile region (IQR) of 0.5 in log 2 scale. A Euclidean distance matrix for all samples based on the filtered dataset was calculated. Filtered data were tested for differences in expression using the moderated t-test in the Limma package from the Bioconductor repository, taking advantage of empirical Bayes moderation. Multiple-testing correction was performed using the Benjamini and Hochberg method [19] .

3. Results

3.1. Modes of TP53 inactivation

We studied 1287 diagnostic CLL samples for the mutational status of the TP53 gene. In 237 cases (18.4%) a TP53 aberration was identified. These TP53 mutations comprised point mutations (69.2%), insertions/deletions (7.2%), aberrant splicing (19%), temperature-sensitive variants (4.2%) and frame shift mutations detectable on genomic DNA only (0.4%).

3.2. TP53 variants with a proven in vitro FASAY-tested mutant phenotype

Table 1 shows the list of TP53 mutations/variants identified in this study. All TP53 variants in Table 1 have been identified by the functional assay FASAY and were corroborated by direct sequencing of the original RNA (cDNA)/gDNA of the respective CLL patients.

Table 1 List of all TP53 mutations/variants identified in the study. Temperature-sensitive variants are in italics.

Mutation No. Mutation No. Mutation No. Mutation No. Mutation No. Mutation No.
E68X 3 K164E 1 H214R 1 P250L 1 D281V 2 TP53 splicing variant Delta Ex6 16
W91X 1 Q167X 1 S215I 1 P250S 1 R282W 1 c.780_781insAGT 1
K101E 1 V172A 1 V216M 1 I251V 1 R283C 7 c.559_560insG 1
P109I 1 V173A 2 R219I 1 I255T 2 T284S 1 c.782_783insTAT 1
R110L 2 R175H 2 Y220C 4 T256P 1 E286G 1 c.102_103insC 1
F113S 2 C176R 2 Y220H 1 E258G 1 K305M 1    
Y126C 2 C176W 1 P222L 1 E258K 1 K320E 1    
Y126H 1 P177L 1 S227P 1 D259G 2 K320T 1    
K132N 2 H178R 1 H233R 4 D259S 1 L330P 1    
Q136E 1 P178S 1 Y234C 5 G266R 1 c.302_del 1    
A138V 1 H179L 1 Y234D 3 R267P 1 c.328_339del 1    
K139R 2 H179R 4 Y234H 1 R267W 1 c.503_ 578del 1    
C141Y 1 R181S 1 N235S 2 R273C 2 c.504_578del 1    
Q144X 2 D186N 1 M237I 2 R273H 4 c.515_559del 1    
W146X 1 G187D 1 N239D 1 R273S 1 c. 550_576del 1    
P151S 1 L188P 1 S241C 1 C275G 1 c. 636del 1    
P152L 1 P190H 1 M243T 1 C275Y 1 c.704_709del 1    
P153L 1 P190S 1 G245D 1 A276V 1 c.716_736del 1    
R156P 1 H193L 2 G245S 8 C277F 1 c.724_739del 1    
V157F 2 H193R 2 N247D 1 C277Y 2 c.749_751del 1    
V157G 1 L194P 1 R248Q 1 R280K 1 c.792_794del 1    
A159P 2 R196X 1 R248W 8 R280P 1 c.828del 1    
A161T 3 Y205C 1 R249S 3 R280T 1 c. 532_ 549del 1    
Y163H 2 R213X 1 R249W 1 D281G 1 TP53 splicing variant Beta 30    

3.3. Temperature-sensitive variants of TP53 vary in their in vitro behavior

Interestingly, some TP53 variants show temperature-sensitive (t.s.) properties. As a routine procedure, after transformation the yeast were grown at 35 °C for three days, followed by cultivation at 25 °C for three days. Yeast colonies harboring bona fide TP53 inactivating mutants/variants retained their red color at both cultivation temperatures; yeast colonies harboring wild type 53 remained large and white irrespective of the cultivation temperature. Of note, yeast colonies harboring t.s. TP53 variants displayed a fully penetrant mutant phenotype when cultivated at permissive temperature; when cultivated at non-permissive temperature, they showed standard wild type TP53 growing characteristics. The permissive temperature for individual t.s. TP53 variants differed: TP53 variants V157F, A161T, S215I, V216 M and Y234C displayed the mutant phenotype (red yeast colonies) at 35 °C and the wild type phenotype (white yeast colonies) at 25 °C, whereas TP53 variants N235S, V274A, R283C, K320E and K320T produced white yeast colonies at 35 °C and red colonies at 25 °C ( Fig. 1 ). To address the issue if these t.s. TP53 variants had any biological relevance in vitro, we cloned all the t.s. TP53 variants and expressed them stably in the H1299 TP53−/− recipient cell line.

gr1

Fig. 1 All nine t.s. TP53 variants, wild type TP53 and hot spot mutant R175H analyzed in the study. Yeast were transformed with the pSS16 vector recombined with the respective TP53 variant and grown for 3 days at 35 °C followed by 25 °C for another three days. Note the temperature-sensitive change in color of the colonies harboring the individual t.s. TP53 variants: the top of each colony represents the clonal population of yeast grown at 35 °C, the bottom of each colony represents the clonal population of yeast grown at 25 °C. Wild type TP53 and the hot spot R175H mutant retain their phenotypes (white and red, respectively) despite the temperature changes.

3.4. T.s. TP53 variants V157F, A161T, S215I, V216M, Y234C display mutant-like phenotype in vitro, while t.s. TP53 variants N235S, R283C, K320E and K320T resemble wild type TP53

Using the megaprimer approach, we cloned nine t.s. TP53 variants (V157F, A161T, S215I, V216M, Y234C, N235S, R283C, K320E, K320T), as well as the hot spot TP53 mutation R175H ( Fig. 2 ). H1299 TP53−/− cells were transfected in duplicate with the pCDNA4/TO expression vector harboring the t.s. TP53 variants, the H175R hot spot TP53 mutant and wild type TP53. Mock-transfected cells were used as controls. After 14 days of Zeocine selection, hundreds of colonies repopulated each cultivation flask. Polyclonal populations were harvested and the levels of TP53 mRNA as well as protein levels were measured, using quantitative real-time PCR and Western blot, respectively ( Fig. 3 ). As assessed at the mRNA level and corroborated at the protein level, the expression of t.s. TP53 variants N235S, R283C, K320E and K320T expressed the transgenic mRNA and protein at levels comparable with the wild type TP53. On the other hand, t.s. TP53 variants V157F, A161T, S215I, V216M and Y234C were comparable with the expression of the hot spot TP53 variant R175H. It is known that bona fide TP53 mutants escape physiological regulation and tend to be expressed at much higher levels than wild type TP53 [20] . All t.s. variants resembling the hot spot H175R mutation at 37 °C (when tested in the H1299 cell line) grew red at 35 °C (when tested in yeast). In contrast, all t.s. variants resembling wild type TP53 (when tested at 37 °C in H1299 cells) grew white at 35 °C (when tested in yeast). These findings suggest that t.s. TP53 variants V157F, A161T, S215I, V216M and Y234C have in vitro characteristics of TP53 inactivating mutants and should be considered biologically relevant, while the t.s. TP53 variants N235S, R283C, K320E and K320T have in vitro characteristics mimicking the wild type TP53.

gr2

Fig. 2 Sequences of nine t.s. TP53 variants constructed in vitro using the megaprimer approach. The exact position of the mutated base is highlighted by an amber square (t.s. TP53 variants) or green square (hot spot mutation R175H).

gr3

Fig. 3 Quantitative real-time PCR (A) and Western blot (B) detection of the expression of individual TP53 variants in stable polyclonal H1299 cell lines. (A) Graph showing the relative expression of the TP53 transgene compared to the expression of b2-microglobulin. As controls, mock-transfected stable H1299 cell lines were used, which show no TP53 expression. H1299 stably harboring wild type TP53 and H1299 stably harboring the t.s. variants N235S, R283C, K320E, K320T show comparable, very low TP53 expression, which reflects the situation observed under physiological conditions. T.s. variants V157F, A161T, S215I, V216M and Y234C show TP53 expression levels comparable with the hot spot TP53 mutation R175H. (B) Western blot detection of the TP53 expression of all variants studied, in the stable polyclonal H1299 cell line. T.s. TP53 variants V157F, A16T, S215I, S216M and Y234C have similar protein expression profiles as the hot spot mutant R175H (overexpression), while the t.s. TP53 variants N235S, R283C, K320E and K320T resemble the wild type TP53 (undetectable by Western blot). Lines: 1 – V157F, 2 – A161T, 3 – R175H, 4 – S215I, 5 – V216M, 6 – Y234C, 7 – N235S, 8 – R283C, 9 – K320E, 10 – K320T, 11 – wild type TP53, 12 – mock-transfected H1299 cell line. Black arrows indicate the molecular weights of detected proteins (53 kDa). Beta-actin was used as the loading control.

As the expression of TP53 transgenes and the Zeocine cassette from the pcDNA4/TO vector in H1299 cell lines were driven by separate promoters, stable clones that arose after Zeocine selection were specifically those with a correctly integrated Zeocine resistance cassette. Thus, numerous stable colonies obtained after the trasfection with pcDNA4/TO harboring the wild type TP53, N235S, R283C, K320E and K320T TP53 t.s. TP53 variants suggest that the transfection efficiency was sufficient and the antibiotic selection cassette was functional. The presence of stable colonies does not directly imply the expression of the TP53 transgene, which either might have been high (if conferring a biological advantage) or might have been intrinsically downregulated to levels compatible with cell survival. It may be anticipated that under the influence of the wild type TP53 transgene and its kin, only those stable colonies that were able to modulate the expression of the TP53 transgene to levels resembling wild type TP53, and hence not negatively interacting with the cell cycle and survival, would grow.

3.5. An in vivo correlate of the R283C t.s. TP53 variant

To bring the above in vitro findings to some in vivo perspective, we present here a short case report implicating the in vivo behavior of one of the tested t.s. variants, R283C. The R283C t.s. TP53 variant was revealed in one of our CLL patients after she had been allotransplanted for her aggressive disease. The graft for the allotransplant was collected from her HLA-matched brother. Since one of the project tasks was to evaluate the possibility of evolution of TP53 mutated subclones induced by chemotherapy (see below), we repeatedly FASAY-tested her bone marrow samples during the MRD follow-up. At the time of diagnosis, she presented with wild type TP53, as determined by FASAY. Of utmost interest, after the bone marrow transplant, the lower her MRD level was, the higher the number of t.s. colonies she displayed on FASAY. At the time point when she was fully engrafted and her MRD was nearly negative, her FASAY showed 51.8% t.s. colonies ( Fig. 4 A). Resequencing of the t.s. colonies revealed the R283C TP53 variant, which had not been present at the time of diagnosis. In the search for the source of this mutation, her brother – the donor of her graft, was invited to the hospital and had peripheral blood taken for TP53 direct sequencing and FASAY. Interestingly, he was found to harbor the germinal heterozygous R283C TP53 mutation ( Fig. 4 B), which he donated to his sister. Mr. Brother is a healthy individual, with no signs of any clinical disorder. Though it cannot be excluded that under some conditions the R283C TP53 variant might pose a biological danger to its carrier, so far both siblings, sharing R283C positive hematopoietic cells, enjoy good health.

gr4

Fig. 4 (A) FASAY at the time of diagnosis and during the MRD follow-up in patient MS. The patient presented with wild type TP53, as assessed by FASAY. After allogeneic bone marrow transplant, she was followed-up using a clone-specific IgVH molecular assay and the mutational status of her TP53 was regularly investigated by FASAY. Interestingly, the more the MRD level decreased, the more the number of t.s. TP53 colonies harboring the R283C TP53 variant increased. (B) The mutation was back-tracked to the peripheral blood of her brother, the donor of the graft, and the germinal heterozygous R283C mutation was corroborated.

3.6. Delta EX6 TP53 variant confers an accented proliferative phenotype in vitro

Recently, it has been reported that TP53 mRNA does not exist as a sole splicing isoform, but might entail at least nine different splicing variants [6] . Importantly, depending on the level of the aberrantly spliced TP53 variant, the cell might acquire a mutant phenotype [6], [7], and [8]. In CLL patients, we recently identified a novel TP53 splicing isoform, Delta Ex6, which is differentially expressed in CLL patients as compared to healthy donors16. To investigate its possible biological role in vitro, we cloned the Delta Ex6 TP53 cDNA into the expression vector pCNDA4/TO and stably introduced it into H1299 TP53−/− cells. Expression of the transgene was corroborated using Western blotting and anti-FLAG antibody ( Fig. 5 ). After 14 days of Zeocine selection, ten individual stable colonies were picked and replanted into separate cultivation flasks to obtain monoclonal cultures harboring the Delta Ex6 TP53 variant. Mock-transfected H1299 cells were used as the background control. Interestingly, 8 out of ten stable H1299 Delta Ex6 monoclonal colonies were distinguished by unusual growth properties: H1299 cell lines stably expressing the Delta Ex6 transgene grew in semi-suspended clusters, apparently with a loss of intercellular contacts ( Fig. 6 A), and accented growth properties, as determined using longitudinal cumulative cell number counting ( Fig. 6 B). In stark contrast, the parental H1299 cell line is characterized by strictly adherent cell growth and the firm attachment of individual cells to each other to form a confluent layer ( Fig. 6 C). Given the fact that 8 out of 10 monoclonal stable H1299 cell lines harboring the DeltaEx6 variant displayed an identical in vitro phenotype, we ascribed it to the direct effect of the transgene. To reveal the molecular background of the in vitro observed phenotype, we subjected four Delta Ex6 H1299 monoclonal stable cell lines as well as two stable H1299 cell lines harboring wild type TP53 and two stable H1299 mock-transfected cell lines to Affymetrix GeneChip Human Exon 1.0 ST Array to carry out whole transcriptome RNA profiling ( Fig. 6 D). Interestingly, the observed in vitro phenotype fully correlated with our molecular data: the microarray showed downregulation of genes involved in cell adhesion and production of the intercellular matrix, and upregulation of genes involved in cell cycle regulation, destruction of the intercellular matrix and inhibition of apoptosis ( Table 2 ). Thus, the microarray data are in good agreement with the observed in vitro pnenotype of the Delta Ex6 variant and imply an accented proliferative phenotype, loss of intercellular contacts and defects of apoptosis induced under the influence of the Delta Ex6 TP53 transgene.

gr5

Fig. 5 Western blot detection of the Delta Ex6 TP53 variant. The N-flagged Delta Ex6 TP53 variant as well as N-flagged hot spot TP53 mutant R175H were transiently expressed in H1299 cells, harvested and subjected to Western blot analysis using anti-FLAG antibody. Lines: 1 – N-flagged hot spot variant R175H; 2 – N-flagged Delta Ex6 TP53 variant; HeLa cells transiently transfected with pEGPF-C1. Expression of the hot spot R175H TP53 variant is high, while expression of the Delta Ex6 TP53 variant is far lower, a situation corresponding to the presumed biological stability of both TP53 variants, the latter being truncated after the 189th amino acid.

gr6

Fig. 6 Eight out of ten monoclonal H1299 colonies harboring stably integrated transgene Delta Ex6 showed an apparent loss of intercellular contacts, semi-suspension growth in clusters (A) and accented proliferation, as assessed by cumulative cell number counting (B). In contrast, (C) shows the growth of the parental H1299 cell line, which is strictly adherent, with the tendency to form a confluent monolayer. Four stable Delta Ex6 harboring H1299 cell lines, as well as control cell lines, were subjected to Affymetrix GeneChip Human 1.0 ST Array (D). The “heat map” suggests marked differences in expression profiles between the Delta Ex6 TP53 variant and mock-transfected H1299 (see also Table 2 ).

Table 2 List of differentially expressed genes as a direct effect of the transgene Delta Ex6 stably expressed in H1299 cells. Genes shown in italics are overexpressed in the stable Delta Ex6 H1299 cell lines; genes shown in bold are downregulated in stable Delta Ex6 expressing H1299, as compared to mock-transfected H1299 cell lines.

Gene symbol Gene name log FC AveExpr P value
CCNA1 Cyclin A1 1.987772 8.1292385 2.43E−05
TP53 Tumor protein p53 (Li-Fraumeni syndrome) 1.501532 8.44899825 0.00406574
CCNG1 Cyclin G1 0.94022 10.21566225 9.49E−05
CDCA2 Cell division cycle associated 2 0.809202 9.42270525 0.000400403
ADAM9 ADAM metallopeptidase domain 9 (meltrin gamma) 0.696348 10.42573163 0.010519032
MMP1 Matrix metallopeptidase 1 (interstitial collagenase) 0.682683 8.23727575 0.301052459
CDKL2 Cyclin-dependent kinase-like 2 (CDC2-related kinase) 0.628412 7.319255875 0.001669125
CCNF Cyclin F 0.624909 9.98326 0.002788007
CCNI Cyclin I 0.591225 10.933535 0.001450265
CTSH Cathepsin H 0.584205 6.957025 0.062920966
CCNB2 cyclin B2 0.518028 11.578855 0.001696948
TP53INP2 Tumor protein p53 inducible nuclear protein 2 0.49461 8.461409125 0.01560313
ADAM22 ADAM metallopeptidase domain 22 0.463773 8.7003425 0.01565168
CDK7 Cyclin-dependent kinase 7 0.428272 8.77931525 0.078375718
CCNB1 Cyclin B1 0.420433 12.7160675 0.002545476
MMP16 Matrix metallopeptidase 16 (membrane-inserted) 0.36512 7.94168875 0.28442019
CCNA2 Cyclin A2 0.338685 11.1359775 0.057153719
CCNT2 Cyclin T2 0.325747 9.238263625 0.283231056
PRSS3 Protease, serine, 3 (mesotrypsin) 0.300147 6.2207275 0.052227114
PRSS12 Protease, serine, 12 (neurotrypsin, motopsin) 0.237427 8.081077125 0.140991412
MMP13 Matrix metallopeptidase 13 (collagenase 3) 0.201086 6.243485125 0.172261585
CNNM4 Cyclin M4 0.17842 8.51714175 0.153840707
CCNL2 Cyclin L2 0.159943 9.74342925 0.41308348
CCNE1 Cyclin E1 0.158643 8.831538875 0.622404425
CCNC Cyclin C 0.156538 11.26617 0.373198847
CCNJ Cyclin J 0.140215 8.02908675 0.157150487
CNNM2 Cyclin M2 0.118963 8.18353425 0.347021857
CCNH Cyclin H 0.114179 9.87748675 0.416173288
CCNO Cyclin O 0.074694 8.32527625 0.572699908
MMP15 Matrix metallopeptidase 15 (membrane-inserted) 0.065574 7.1489615 0.500330471
CARD14 Caspase recruitment domain family, member 14 0.05987 7.161872625 0.53410234
DFFB DNA fragmentation factor, 40kDa, beta polypeptide (caspase-activated DNase) 0.06225 7.2329525 0.717893341
CFLAR CASP8 and FADD-like apoptosis regulator 0.07773 8.73841025 0.617069386
CARD9 caspase recruitment domain family, member 9 0.09622 7.747972875 0.407241037
CASP8 Caspase 8, apoptosis-related cysteine peptidase 0.10701 8.25145025 0.434560079
CARD11 Caspase recruitment domain family, member 11 0.11176 6.198165125 0.310073033
CARD10 Caspase recruitment domain family, member 10 0.12869 8.514723 0.259805154
CARD8 Caspase recruitment domain family, member 8 0.2912 8.954539125 0.148542342
CASP4 Caspase 4, apoptosis-related cysteine peptidase 0.42522 9.985524125 0.055122148
COL12A1 Collagen, type XII, alpha 1 1.04472 6.740112125 0.042764989
DSG2 Desmoglein 2 1.24819 6.825398375 0.000420777
AMIGO2 Adhesion molecule with Ig-like domain 2 1.31457 8.831015875 3.46E 05
JAM2 Junctional adhesion molecule 2 1.55456 8.071170875 0.002791424
LAMC2 Laminin, gamma 2 1.80414 8.895991375 0.000249893
CALB2 Calbindin 2, 29 kDa (calretinin) 1.82215 7.7815825 0.000202464
ITGA2 Integrin, alpha 2 (CD49B, alpha 2 subunit of VLA-2 receptor) 2.38844 6.863864125 0.001614398

3.7. Evolution of TP53 mutants following treatment does not seem to play a major role in CLL progression

During the duration of the project, we FASAY-tested 1148 successive follow-up samples from 181 CLL patients for whom clone-specific MRD assays had been prepared [17] . In contrast to previously published data [14] we did not observe any clonal TP53 evolution following chemotherapy treatment. Those CLL individuals who harbored a TP53 mutation at diagnosis either succumbed to their disease/related complications, or the TP53 mutation disappeared during treatment, or the TP53 mutation remained the same during the follow-up despite the chemotherapy. In no analyzed follow-up sample were we able to identify new TP53 variants that would be truly clonal, identifiable not only by FASAY but also verified by sequencing of patient cDNA or gDNA and not be present at diagnosis but emerge at some point during chemotherapy.

4. Discussion

In our study, we investigated the TP53 mutational status in 1287 CLL patients at the time of diagnosis and 1148 successive follow-up samples derived from 181 CLL patients undergoing treatment, for whom clone-specific MRD assays had been prepared. This work has led to several interesting observations, some of them novel.

Recently, it has been shown that there are at least nine different splicing isoforms of TP53 which had not been previously recognized. We recently identified a novel, not-yet described splicing variant, Delta Ex6, lacking the entire coding sequence of exon 6 and harboring a cryptic exon 9B between exons 9 and 10 [16] . Here, we show that Delta Ex6 is differentially expressed in CLL patients as compared to healthy donors, and, in vitro, it confers an accented proliferative phenotype and causes the loss of intercellular contacts. Chiorazzi has shown that CLL cells are more proliferatively active than previously thought [21] . Our in vitro findings on the Delta Ex6 TP53 variant further support their data and suggest that aberrant splicing variants of genes can exert yet unexpected functions and might, depending on the load of the aberrant splicing product, confer a mutated phenotype to the cell. In these instances, genomic DNA remains intact, but distortion of the mRNA splicing machinery leads to negative modulation of the TP53 pathway. The mechanism of aberrant TP53 mRNA splicing needs elucidation, but it clearly implies that cells have several possible ways to subvert their TP53 pathway: genomic mutations, aberrant mRNA splicing, postranslational modifications or modulation of TP53 interacting partners, just to mention the few already known mechanisms [4], [5], [6], [7], [8], [9], [10], and [11].

The temperature-sensitive variants (t.s. variants) are another interesting issue. It is generally assumed that any TP53 mutation, except for known SNPs, is inactivating. We have identified nine t.s. TP53 variants that acquire different phenotypic traits depending on the biological conditions they are exposed to. It is known that in order for proteins to be folded properly, they require the help of specialized proteins (molecular chaperons) during the establishment of their secondary and tertiary structure. It is suggestive that some t.s. TP53 variants cannot fold correctly under non-permissive conditions, while other (permissive) conditions allow the process to be completed successfully. Our in vitro experiments on yeast as well as on H1299 TP53−/− cells suggest that for TP53 variants N235S, R283C, K320E and K320T, physiological temperatures of 35 °C (yeast) or 37 °C (human) are not permissive for the mutated phenotype to develop. Thus, these t.s. TP53 variants might be considered biologically benign. It remains to be established what their actual biological role is, but even our preliminary in vitro data imply that these TP53 variants should be treated with caution in respect to CLL patient prognoses, as they resemble wild type TP53 rather than a hot spot TP53 mutation. The Mr. Brother and Mrs. Sister case might serve as a prima facie example ( Fig. 4 ), where both siblings now carry the heterozygous germinal R283C t.s. TP53 variant, which in this long-term “in vivo experiment” seems to behave benignly.

We provide here a table of biologically relevant transactivation-defective TP53 mutations/variants that have been FASAY-identified and corroborated by direct sequencing on cDNA/gDNA of the individual CLL patients. We believe this data might be useful for those investigators who do not perform functional TP53 tests to assess the biological relevance of TP53 mutations identified in their patients (using direct sequencing or other non-functional analysis).

Last but not least, we would like to discuss the discrepancy between our data and that published [14] on the evolution of TP53 mutated clones induced by chemotherapy. In our study of 1148 consecutive follow-up CLL samples, derived from 181 CLL patients undergoing chemo/immunotherapy or bone marrow transplant according to the respective institutional protocols, we did not observe a single case of clonal TP53 evolution induced by treatment. Thus, we suggest that this discrepancy might be caused by some technical aspects of the way FASAY has been carried out. In our study, to identify TP53 mutations using FASAY we routinely picked and pooled 20 red colonies and looked for clonal mutations on the pooled DNA template. The theory underlying this approach is that if a red colony carries a TP53 with a mutation, the cause might be: (i) the mutation is truly clonal, mirroring the in vivo CLL situation, (ii) it is a result of a reverse transcriptase error (the reverse transcriptase enzyme is prone to mistakes as it does not possess proof-reading activity), or iii) it is a result of a PCR error (though proof-reading DNA polymerases are used). As such, if nucleotide substitutions arise from the assay methodology (reverse transcriptase or DNA polymerase errors), these mutations would be spread randomly over the analyzed sequence. These aberrant signals should reach a maximum of 5% (1 possible mistake in 20 templates) and the chromatogram obtained would show wild type sequence, as a 5% signal is indistinguishable from the sequencing background. Moreover, each FASAY – identified TP53 variant in our study was back-tracked to the patient cDNA or gDNA and only positively corroborated findings were regarded as relevant.

FASAY is a powerful technique, but is prone to artifacts if the quality of RNA is compromised, as is often the case when investigating material from patients undergoing intensive chemotherapy. Cells harvested during this period are often suboptimal for analysis: they are low in number, their RNA might be degraded, and if resequencing is carried out on individual single red colonies, technical artifacts might bias and skew the results.

Taken together, the story of TP53 is far from being complete. The more work is devoted to writing this story, the more challenges appear. The field has been a place of intensive research for more than 25 years, and novel findings and interesting data are still being revealed that, we are happy to observe, can often be translated into clinical practice.

Conflict of interest statement

The authors declare no conflict of interests.

Acknowledgements

We are indebted to Iva Zelezna and Martina Snajdrova for their invaluable technical assistance. The authors wish to thank Dr. Robert Ivanek for the bioinformatics and biostatistics analyses of the Affymetrix microarray data. This work has been supported by the Grant Agency of the Ministry of Health of the Czech Republic, NS 9652-4 and the Research Project MZO 00179906 of the Ministry of Health of the Czech Republic.

Contributions. S.P. and O.M. contributed equally to the work and should be considered as first author. All co-authors contributed to the work significantly, they were responsible for the design of the experiments, individual analyses, and writing and final editing of the manuscript.

References

  • [1] D.P. Lane. Cancer, P53 guardian of the genome. Nature. 1992;358:15-16 Crossref.
  • [2] D.P. Lane, P.M. Fischer. Turning the key on p53. Nature. 2004;427:789-790 Crossref.
  • [3] http://www-p53.iarc.fr .
  • [4] J. Momand, G.P. Zambetti, D.C. Olson, D. George, A.J. Levine. The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell. 1992;69:1237-1245 Crossref.
  • [5] C.J. Sherr. Divorcing ARF and p53: an unsettled case. Nat Rev Cancer. 2006;6:663-673 Crossref.
  • [6] J.C. Bourdon, K. Fernandes, F. Murray-Zmijewski, G. Liu, A. Diot, D.P. Xirodimas, et al. p53 isoforms can regulate p53 transcriptional activity. Genes Dev. 2005;19:2122-2137 Crossref.
  • [7] K.L. Harms, X. Chen. The C terminus of p53 family proteins is a cell fate determinant. Mol Cell Biol. 2005;25:2014-2030 Crossref.
  • [8] M. Nissim-Rafinia, B. Kerem. Splicing regulation as a potential genetic modifier. Trends Genet. 2002;18:123-127 Crossref.
  • [9] G. Rohaly, J. Chemnitz, S. Dehde, A.M. Nunez, J. Heukeshoven, W. Deppert, et al. A novel human p53 isoform is an essential element of the ATR-intra-S phase checkpoint. Cell. 2005;122:21-32 Crossref.
  • [10] M. Nissim-Rafinia, B. Kerem. The splicing machinery is a genetic modifier of disease severity. Trends Genet. 2005;21:480-483 Crossref.
  • [11] S. Kato, S.Y. Han, W. Liu, K. Otsuka, H. Shibata, R. Kanamaru, et al. Understanding the function-structure and function-mutation relationships of p53 tumor suppressor protein by high-resolution missense mutation analysis. Proc Natl Acad Sci USA. 2003;100:8424-8429 Crossref.
  • [12] T. Zenz, D. Vollmer, M. Trbusek, J. Smardova, A. Benner, T. Soussi, et al. TP53 mutation profile in chronic lymphocytic leukemia: evidence for a disease specific profile from a comprehensive analysis of 268 mutations. Leukemia. 2010; [Epub ahead of print]
  • [13] J. Slovackova, D. Grochova, J. Navratilova, J. Smarda, J. Smardova. Transactivation by temperature-dependent p53 mutants in yeast and human cells. Cell Cycle. 2010;9:11 [Epub ahead of print]
  • [14] J. Malcikova, J. Smardova, L. Rocnova, B. Tichy, P. Kuglik, V. Vranova, et al. Monoallelic and biallelic inactivation of TP53 gene in chronic lymphocytic leukemia: selection, impact on survival, and response to DNA damage. Blood. 2009;114:5307-5314 Crossref.
  • [15] J. Smardova. FASAY: a simple functional assay in yeast for identification of p53 mutations in tumors. Neoplasma. 1999;46:80-88
  • [16] S. Pekova, R. Cmejla, L. Smolej, T. Kozak, M. Spacek, M. Prucha. Identification of a novel, transactivation-defective splicing variant of p53 gene in patients with chronic lymphocytic leukemia. Leuk Res. 2008;32:395-400 Crossref.
  • [17] S. Pekova, L. Bezdickova, L. Smolej, T. Kozak, I. Hochova, P. Zak, et al. Quantitation of minimal residual disease in patients with chronic lymphocytic leukemia using locked nucleic acid-modified, fluorescently labeled hybridization probes and real-time PCR technology. Mol Diagn Ther. 2007;11:325-335 Crossref.
  • [18] Sambrok J, Fritsch EF, Maniatis I. Molecular cloning: a laboratory manual; 1989. ISBN-10:0879693096.
  • [19] Y. Hochberg, Y. Benjamini. More powerful procedures for multiple significance testing. Stat Med. 1990;9:811-818 Crossref.
  • [20] D. Reisman, W.T. Loging. Transcriptional regulation of the p53 tumor suppressor gene. Semin Cancer Biol. 1998;8:317-324 Crossref.
  • [21] N. Chiorazzi. Cell proliferation and death: forgotten features of chronic lymphocytic leukemia B cells. Best Pract Res Clin Haematol. 2007;20:399-413 Crossref.

Footnotes

a Laboratory for Molecular Diagnostics, Chambon Inc., Evropska 176/16, Prague 6, 160 00, Czech Republic

b Department of Hematology and Oncology, Teaching Hospital Pilsen, Pilsen, Czech Republic

c Department of Clinical Hematology, Teaching Hospital Hradec Kralove, Hradec Kralove, Czech Republic

d Department of Clinical Hematology, Teaching Hospital Kralovske Vinohrady, Prague, Czech Republic

lowast Corresponding author. Tel.: +420 221 985 475; fax: +420 221 985 475.