SEARCH

SEARCH BY CITATION

Keywords:

  • polycythemia vera;
  • CD34+;
  • gene signature;
  • DNA-PK;
  • KU86

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Polycythemia vera (PV) is a clonal hematopoietic stem cell disease characterized by a trilinear accumulation of blood cells that has been recently associated with a JAK2V617F point mutation. However, this molecular defect represents a rather late event in the disease progression, is not specific for this disease, and is not ascertained in all patients indicating that additional factors contribute to the specific phenotype of PV. Therefore, cDNA microarray analyses were performed on CD34+ peripheral blood stem cells (PBSC) with subsequent evaluation on mRNA and protein level of a larger cohort of PV patients. Microarray analyses revealed a significant dysregulation of 11 genes. KU86, a gene coding for a subunit of the DNA-dependent protein kinase (DNA-PK), displayed the strongest upregulation in all patients under study. This peculiarity was accompanied by downregulation of the catalytic DNA-PK subunit DNA-PKcs. Also Ku86 protein was upregulated and expressed in the vast majority of CD34+ PBSC nuclei while a weak nuclear expression was detected in only one blood donor. Differential expression of several genes, imbalance of the distinct subunits of DNA-PK, and particularly the strong upregulation of Ku86 protein, are new findings in PV CD34+ PBSC. These factors may contribute to the accumulation of chromosomal aberrations, accumulation of hematopoietic cells (especially of erythropoiesis), and prolongation of CD34+ PBSC life span observed in PV. © 2008 Wiley-Liss, Inc.

Chronic myeloproliferative disorders are characterized by abnormal accumulation of more than one hematopoietic lineage associated with relatively normal maturation.1 Amongst these polycythemia vera (PV) is an acquired clonal myeloaccumulative disease (panmyelosis) with increased red cell production independent of mechanisms that normally regulate erythropoiesis.2 An acquired and activating JAK2V617F point mutation was previously described occurring in most patients suffering from this disease.3–7 However, it has to be taken into account that this point mutation is not specific for PV and that 3–36% of PV patients are JAK2V671F negative.3, 6 In addition, Kralovics et al. were able to demonstrate that JAK2V617F represents a rather late event in disease progression.8, 9 Therefore, other molecular defects within the stem cell compartment have to be responsible for the specific phenotype of PV.

On this account, we analyzed gene regulation of CD34+ peripheral blood stem cells (PBSC) from PV patients and healthy blood donors. The most prominent elevated gene was KU86, a subunit of the DNA-dependent protein kinase (DNA-PK). As a multifunctional protein Ku86 has been implicated in the regulation of many pivotal nuclear processes such as DNA double-strand break repair by non-homologous end joining (NHEJ).10–12 Furthermore, in acute and chronic myeloid leukemias an aberrant activity of DNA-PK subunits Ku70/Ku86 has been shown to be a candidate mechanism for chromosomal instability.13, 14 Therefore, taken into account that DNA-PK subunits not only acting as a precise adjusted entity but also each exhibiting their own distinct functions an imbalance may cause severe disorder to cell metabolism.15 Consequently, we further analyzed gene expression of the 3 DNA-PK subunits and Ku86 protein expression in PV CD34+ PBSC. Furthermore, in PV bone marrow biopsies cellular distribution and abundance of Ku86 protein in hematopoiesis was scrutinized.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Study design

After informed consent, bone marrow trephine biopsies and peripheral blood (PB) samples from 9 patients with histopathologically determined chronic phase PV were obtained for diagnostic or therapeutical purpose (phlebotomy). No additional myelosuppressive regimens were administered. The diagnosis of PV was ascertained according to the criteria of the WHO classification of 2001 and all patients also meet the criteria of the new WHO classification of 2008.16–18

At initial diagnosis, all patients presented hemoglobin values of >18.5 g/dl (in men) and >16.5 g/dl (in women), a thrombocytosis of >400 × 109/l, and low Epo levels. Beside biopsies from 5 PV patients in the chronic phase of their disease 5 archived trephine biopsies conducted to evaluate dissemination of the disease where taken from patients with nodal lymphoma (without bone marrow involvement) and 26 blood samples from healthy individuals served as controls. All patients under study carried the JAK2V617F point mutation as evaluated by PCR.

Additionally, to verify the influence of the allelic burden of the JAK2V617F point mutation due to the RNA expression levels of the 6 genes examined by RT-PCR we used allele specific PCR and PCR product sequencing to answer this question. All patients under study revealed heterozygosity in their CD34+ PBSC compartment (Supporting Figure 1).

Selection and isolation of CD34+ and CD14+ cells

Phlebotomy samples of 500 ml peripheral blood underwent centrifugation over Ficoll-Paque (Pharmacia, Uppsala, Sweden) to obtain a mononuclear cell concentrate. Mononuclear cells were labeled with either anti-CD34 micro beads or anti-CD14 micro beads (both Miltenyi Biotec, Bergisch-Gladbach, Germany) and selected by an immunomagnetic separation system according to the manufacturer's instruction (mini-MACS, Miltenyi Biotec). To increase purity of the eluted cells two further passes through separation columns were performed. Purity was ascertained by FACS flow cytometry (Becton Dickinson, San Jose, CA). Cells were washed, immediately frozen in liquid nitrogen, and stored at −80°C.

DNA and RNA extraction

Isolation of genomic DNA and total RNA from 1 × 105 to 5 × 105 CD34+ PBSC was performed with the DNeasy or the RNeasy Micro Kit including DNAse treatment, respectively (Qiagen AG, Hilden, Germany) according to the manufacturer's instructions. DNA and RNA yield, purity and quality were measured photometrically and validated by gel electrophoresis.

cDNA microarray

Atlas Human 1.2 I array (BD Biosciences Clontech, Heidelberg, Germany) containing 1185 gene specific cDNA was used for hybridization experiments. Total RNA was reversely transcribed and radioactively labeled with [alpha-32P] dATP (Amersham Biosciences, Buckinghamshire, UK). The very low RNA content of the samples required the selection of the ominiscript reverse transcriptase (Qiagen AG, Hilden, Germany) according to a modified protocol of the manufacturer's instructions. With this modification samples of minimum 500 ng RNA could be analyzed. Hybridization was performed according to the producer's instructions. For signal detection Phosphor Imager screens (Amersham) were exposed for up to 1 week and scanned with the Storm Imaging System (Amersham). Data were quantified using the ArrayVision V4.0 software (Imaging Research, Ontario, Canada) and normalized against 6 housekeeping genes included on the microarray (glyceraldehyde-3-phosphate dehydrogenase; alpha tubulin; major histocompatibility complex, class I; beta actin; ribosomal protein L13a; ribosomal protein S9).

Allele-specific QPCR and PCR product sequencing

AS-QPCR was performed on an MX3000p (Stratagene, Amsterdam, Netherlands) detection system with typical reactions of a final volume of 20 μl, containing primers (300 nM) and probes (200 nM) of a JAK2V617F specific SNP assay (Applied Biosystems, Foster City, CA) mixed with the appropriate volume of Eppendorf RealMasterMix (Eppendorf AG, Hamburg, Germany) and 2 μl genomic DNA.

Briefly, the results of PCR were validated using sequence specific primers labeled with the IRD 800 fluorescence dye (Metabion, Planegg-Martinsried, Germany). Sequence analysis was carried out on a LI-COR DNA analyzer (Gene Reader 4200; MWG-Biotech). The sequences products were compared with the published data of the NCBI database by BLAST analyses to exclude amplification of a false amplicon.

Real-time RT-PCR

Total RNA was reversely transcribed using random hexamer primers (Invitrogen, Karlsruhe, Germany) and Sensiscript RT Kit (Qiagen AG, Hilden, Germany) in case of CD34+ PBSC and Omniscript RT Kit (Qiagen AG, Hilden, Germany) in case of CD14+ cells. Primers and probes were designed using open source Primer3-web software (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) and were evaluated by Blast searches at NCBI (http://www.ncbi.nlm.nih.gov/blast/) (Table I). All primers and probes were synthesized by Metabion (Planegg-Martinsried, Germany). For detection and quantification of the PCR assays the MX3000p (Stratagene, Amsterdam, Netherlands) was employed. The real-time PCR amplification was performed in a final reaction volume of 20 μl containing primers (300 nM) and probe (200 nM) mixed with the appropriate volume of Eppendorf RealMasterMix (Eppendorf AG, Hamburg, Germany) and 4 μl cDNA. The reaction mixture was preheated at 95°C for 2 minutes, followed by 45 cycles at 95°C for 20 seconds and 60°C for 1 minute. All experiments were evaluated by performing an identical second run. To avoid quantification bias a standard curve of every assay on every single run has been carried out to ascertain the specific amplification efficiency. Briefly, the relative expression mRNA level was determined setting the cycle threshold against the standard curve in each case. Normalization against the internal control Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was followed by calculating the case specific calibrated gene expression. The mean value of normalized gene expression of all healthy controls in one assay served as calibrator.

Table I. Primers and Probes for Real-Time RT-PCR Assays
ProteinGene Primers/probes 5′ [RIGHTWARDS ARROW] 3′Amplicon length
  1. F, indicates forward primer; R, reverse primer′; P, TaqMan probe.

Ku86XRCC5FCAA AGA GGA AGC CTC TGG AA116bp
RCCA CAT CAC CAC CTT CTT CA
PFAM—TTC TGT CAC AGC TGA GGA AGC CAA A—BHQ1
Ku70XRCC6FGAC ATT GCC CAA GGT TGA AG185bp
RTGT GGG TCT TCA GCT CCT CT
PFAM—AGA CTG GGC TCC TTG GTG GAT GAG—BHQ1
DNA-PKcsPRKDCFGCG AAT ACT TCC AGG CTT TG155bp
RTTG TTT CGC AAC CAG TTC AC
PFAM—AAA GAA GTG TAT GCC GCT GCA GCA—BHQ1
Jak3JAK3FTAT CCT TGA CCT GCC AGT CC119bp
RACT CAC CCT GCT CCT TGA GA
PFAM—AGC ACC GCA GTG ACC TGG TGA GT—BHQ1
MAPKK3MAP2K3FGGA GCT CAT GGA CAC ATC CT105bp
RCGC ACG ATA GAC ACA GCA AT
PFAM—ACA AGT TCT ACC GGA AGG TGC TGG A—BHQ1
STAT1STAT1FCAA GTT CGG CAG CAG CTT A144bp
RCAC CAC AAA CGA GCT CTG AA
PFAM—TGT TAT GGG ACC GCA CCT TCA GTC—BHQ1
GAPDHGAPDHFCTC TGC TCC TCC TGT TCG AC112bp
RACG ACC AAA TCC GTT GAC TC
PFAM—AGC CAC ATC GCT CAG ACA CCA TG—BHQ1

Western blot analysis

Extracted CD34+ PBSC were immediately removed from buffer solutions and frozen in liquid nitrogen. For western blot analysis, cells were homogenized in 200 μl lyses buffer consisting of RIPA-buffer, 10 μl Protease Inhibitor Cocktail (Sigma-Aldrich, Munich, Germany), and 5 μl PMSF (100 mM) by multiple aspirations through a 24-gauge needle. Protein concentrations were measured employing BCA-Protein Assay (Pierce Biotechnology, Bonn, Germany). Equal amounts of protein (10 ng) were separated on a NuPAGE 4–12% Bis-Tris Gel (Invitrogen, Paisley, UK) and transferred to a 0.2 μm pore size nitrocellulose membrane. Western blotting was performed with antibodies against Ku86 (B-1) (Santa Cruz Biotechnology, Heidelberg, Germany). Filters were reprobed with anti-actin (Sigma-Aldrich, St. Louis, MO) mouse IgG2a monoclonal antibody (clone AC-40), to normalize the protein levels. After incubation with the respective secondary antibodies specific bands were observed via enhanced chemiluminescence analysis employing the ECL Plus Western Blotting Detection Kit (GE Healthcare, Munich, Germany). Measurement of the relative optical density (OD) was performed using the Scion Image software release alpha 4.0.3.2 (Scion Corporation; Frederick, MD). For every case the obtained relative OD of Ku86 was normalized by the relative OD of actin.

Immunocytochemistry

Enriched CD34+ PBSC were immediately removed from buffer solutions and mixed with sheep erythrocytes not only to obtain a visible pellet but also to avoid cell clustering. After formalin fixation and paraffin embedding 4 μm sections were dewaxed through xylene, rinsed with descending alcohol concentrations, heated in appropriate buffer (100 mM Tris/50 mM EDTA, pH 8.0, 120°C, 5 min), cooled, and rinsed twice in Tris-HCl/0.05% Tween 20 buffer (pH 7.6). Subsequently, immunostaining with Ku86 (B-1) (Santa Cruz Biotechnology, Heidelberg, Germany) was performed by EnVision HRP labeled System (Dako, Hamburg, Germany). AEC Substrate Chromogen (Dako, Hamburg, Germany) was used as chromogenic substance.

Immunohistochemistry on bone marrow biopsies

Four micrometer paraffin-embedded sections from archived bone marrow trephine biopsies (5 PV patients, 5 controls; mean size 17.5 ± 1.8 mm2) were dewaxed through xylene, air-dried, microwave-heated in appropriate buffer (100 mM Tris/50 mM EDTA, pH 8.0) for 12 min, cooled, and rinsed twice in Tris-HCl/0.05% Tween 20 buffer (pH 7.6) at 4°C. Subsequently, immunostaining with CD34 (QBEND10, Dako, Hamburg, Germany) or Ku86 (B-1) (Santa Cruz Biotechnology, Heidelberg, Germany) was performed by EnVision HRP labeled System (Dako, Hamburg, Germany). AEC substrate (Dako, Hamburg, Germany) was used as chromogenic substance. In case of immunophenotypic double staining incubation with CD34 antibody (120 min, 37°C) was followed by addition of a biotinylated secondary antibody (polyclonal rabbit anti-mouse; Dako, Hamburg, Germany) and detection with streptavidin alkaline phosphatase (Lab Vision; Fremont, CA). Fast Red substrate system was used as chromogenic substance. Subsequently, immunodetection with Ku86 antibody (over night, 4°C) was performed by EnVision HRP labeled System (Dako, Hamburg, Germany) with following diaminobenzidine (DAB)-nickel (Vectorlabs, Burlingame, CA) staining.

Photographic documentation

All pictures were performed on a Zeiss Axiophot microscope (Zeiss, Oberkochen, Germany) with the objective lenses 40X/0.70 NA PL Fluotar or 100X/1.32 oil NA PL Fluotar, respectively. Digital photography was employed with a JVC KY-F75 U (JVC Germany, Friedberg, Germany) using the acquisition software Diskus Version 4.60.342 (Diskus, Koenigswinter, Germany). No further image processing was performed.

Statistical analysis

The Student's t-test was performed for statistical analysis of the data achieved by semiquantitative real-time PCR after testing the normal distribution with one-sample Kolmogoroff-Smirnov-Test. p values ≤ 0.05 were considered as statistically significant and values ≤ 0.01 are shown indicating a higher level of significance. Statistical analysis was achieved using SPSS 14.0.1 software (SPSS, Chicago, IL).

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

KU86 is the most significantly upregulated gene in PV CD34+ peripheral blood stem cells

To assess potential differences in gene expression between CD34+ cells derived from PV patients and healthy individuals an initial screen interrogating 1185 genes was performed (Atlas Human 1.21 array). CD34+ PBSC of 3 PV patients and 12 healthy individuals were immunomagnetically enriched (purity > 95%) before analysis. For this initial screen samples from healthy individuals were pooled while the samples from the 3 PV patients were analyzed individually. The experiments revealed that 5 out of 1185 genes were significantly upregulated and 6 out of 1185 genes were downregulated by a more than twofold difference in either PV patient (Table II). The most prominent upregulated gene was the KU86 subunit of DNA-PK which was significantly elevated 4.97 to 6.27 fold in PV CD34+ PBSC. In contrast, no significant difference in KU70 mRNA expression between PV patients and healthy controls could be revealed. The cDNA of the catalytic subunit of the holoenzyme DNA-PKcs was not spotted on the array.

Table II. Gene Expression Data from cDNA Arrays
ClassificationGenesUniGenePV patients
ABC
Cell cycleMAP kinase kinase 3 (MKK3)Hs.514012−2.89−2.91−4.05
DNA replication licensing factor MCM7 (MCM7)Hs.4387203.013.463.11
Intracellular transducers/ effectors/modulatorsActivated CDC42 kinase 1 (TNK2)Hs.518513−3.48−7.17−8.51
Janus kinase 3 (JAK3)Hs.5152473.324.13.84
Non-receptor tyrosine-protein kinase TNK1 (TNK1)Hs.2034203.173.53.75
STAT1Hs.642990−3.28−2.5−3.26
DNA-synthesis, recombination, repairReplication factor C subunit 2 (RFC2)Hs.647062−3.76−4.95−8.06
KU86 (XRCC5)Hs.3887396.275.894.97
TranscriptionC/EBP alpha (CEBPA)Hs.76171−2.72−4.51−11.11
Heat shock proteinsHeat-shock protein beta-1 (HSPB1)Hs.520973−3.55−4.32−7.58
Cell signalingInterleukin 13 (IL13)Hs.8453.342.649.72

Among the other dysregulated genes 3 candidates have been described to be involved in the JAK-STAT signaling pathway.19–21 Representing the class of cell cycle promoting genes the expression level of mitogen-activated protein kinase kinase 3 (MAPKK3) was significantly reduced in the PV CD34+ PBSC as assessed by examination of 138 genes involved in cell cycle regulation. Also the signal transducer and activator of transcription 1 gene (STAT1) was downregulated whereas expression of Janus kinase 3 (JAK3) revealed significantly increased expression. Altered expression of the JAK-STAT signaling pathway associated genes determined by cDNA array was verified by RT-PCR employing CD34+ PBSC from further 9 PV patients and 9 healthy individuals. All 3 genes revealed analogous dysregulation (Figs. 1a1c).

thumbnail image

Figure 1. mRNA expression of JAK3, MAPKK3, STAT1, DNA-PK subunits KU86, KU70, and DNA-PKcs in CD34+ peripheral blood stem cells. mRNA expression of JAK3 (a), MAPKK3 (b), STAT1 (c), DNA-PK subunits KU86 (d), KU70 (e), and DNA-PKcs (f) in CD34+ peripheral blood stem cells of 9 polycythemia vera patients compared with mRNA expression of 9 healthy blood donors (control) employing semiquantitative RT-PCR technique. Vertical axis represents factor of relative mRNA expression. Graph indicating 25th and 75th percentiles as box margins, 10th and 90th percentiles as error bars and the median as a line in the box. Outlying points are displayed as dots. The p value is given for each graph.

Download figure to PowerPoint

As comprised in Table II further upregulated genes were seen in the group of cell cycle regulating genes (i.e., DNA replication licensing factor MCM7), intracellular transducers (i.e., non-receptor tyrosine-protein kinase TNK1), and cell signaling genes (i.e., interleukin 13 IL13). A significant downregulation was seen in the group of intracellular transducers (i.e., activated CDC42 kinase 1 TNK2), DNA synthesis associated genes (i.e., replication factor C subunit 2 RFC2), transcriptional activators (i.e., C/EBP alpha CEBPA), and heat shock proteins (i.e., heat shock protein ß-1 HSPB1).

Dysregulation of DNA-PK subunits in PV blood cells is restricted to CD34+ cells

As the DNA-PK complex has been implicated in DNA replication, regulation of transcription, and DNA double-strand break repair, regulation of gene expression of the KU86, KU70, and DNA-PKcs subunits of the holoenzyme was analyzed performing RT-PCR. In keeping with the cDNA array results again KU86 was significantly higher expressed in PV CD34+ PBSC compared with the controls. Concerning KU70, a congeneric expression in PV CD34+ PBSC and CD34+ PBSC of healthy individuals was seen. In contrast, in PV CD34+ PBSC a significant decrease of DNA-PKcs compared with the controls could be highlighted (Figs. 1d1f).

To determine whether the dysregulation of the 3 DNA-PK subunits is restricted to CD34+ PBSC or still relevant in mature blood cells we assessed expression of DNA-PK subunits in CD14+ monocytes. CD14+ cells were isolated from peripheral blood and mRNA was analyzed by RT-PCR. In contrast to CD34+ cells of the same individuals, expression of the 3 DNA-PK subunits was not significantly altered between healthy individuals and PV patients (Fig. 2).

thumbnail image

Figure 2. mRNA expression of DNAPK subunits KU86, KU70, and DNA-PKcs in CD14+ monocytes. mRNA expression of DNAPK subunits KU86 (a), KU70 (b), and DNA-PKcs (c) in CD14+ monocytes of 9 polycythemia vera (PV) patients compared with mRNA expression of 9 healthy blood donors (control) employing semiquantitative RT-PCR technique. Vertical axis represents factor of relative expression. Graph indicating 25th and 75th percentiles as box margins, 10th and 90th percentiles as error bars, and the median as a line in the box. Outlying points are displayed as dots. The p value is given for each graph. In all assays a congeneric expression was seen in PV patients and healthy blood donors.

Download figure to PowerPoint

Nuclear overexpression of Ku86 protein in PV CD34+ peripheral blood stem cells

To assess whether the increased mRNA expression of KU86 resulted in even increased protein level, immunoblotting on protein lysates of CD34+ PBSC was performed. As demonstrated in Figure 3 the Ku86 protein (83kDa) was detected in CD34+ PBSC of all PV patients under study while only in 1 of 9 healthy volunteers a faint band was observed.

thumbnail image

Figure 3. Western blot analysis of Ku86 protein expression in CD34+ peripheral blood stem cells. Western blot analysis of Ku86 protein expression in CD34+ PBSC of 9 polycythemia vera (PV) patients compared with 9 healthy blood donors. Relative protein expression was determined by setting the optical density (OD) of Ku86 bands against actin OD. Concerning the healthy blood donors (control) a Ku86 expression was seen only in 1 individual (extreme outlying point displayed as an asterisk). In PV patients Ku86 was seen in all individuals. Graph indicating 25th and 75th percentiles as box margins, 10th and 90th percentiles as error bars and the median as a line in the box. Outlying points are displayed as dots.

Download figure to PowerPoint

When examining the subcellular localization of Ku86 in CD34+ PBSC by immunocytochemistry a strong and strictly nuclear Ku86 expression in the vast majority of cells was seen in all PV samples while an absent or at most weak staining was detected in CD34+ PBSC of healthy individuals. As an internal control the few contaminating leukocytes amidst the enriched PV CD34+ PBSC all remained unstained (Fig. 4).

thumbnail image

Figure 4. Ku86 immunocytochemistry of CD34+ peripheral blood stem cell clots. Ku86 immunocytochemistry of CD34+ peripheral blood stem cells (PBSC) cell clots of 5 polycythemia vera patients compared with 5 healthy blood donors. Ku86 is expressed in a strictly nuclear pattern. Only few cells present a weak expression in CD34+ of healthy blood donors (a, b) while in the PV patients almost all PBSC are intensively stained (c, d). Note the Ku86 negative mature leukocytes (arrows). Original magnification ×1000 for all panels. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Download figure to PowerPoint

Nuclear overexpression of Ku86 protein in PV bone marrow biopsies

Considering the constant overexpression of Ku86 protein in CD34+ PBSC we asked whether this phenomenon could also been revealed in bone marrow CD34+ cells. We therefore analyzed paraffin embedded bone marrow biopsies of 5 PV patients and tumor-free staging bone marrow biopsies of 5 patients with nodal ascertained lymphoma as control. Immunhistochemical analysis described Ku86 expression as a rather rare and again strictly nuclear restricted event in the bone marrow biopsies (Fig. 5). However, affecting less than 1% of hematopoietic cells in the control group and up to 10% of the cells in PV it was clearly more frequently seen in the PV biopsies. Thereby, Ku86 was stained in cells of the erythroid and megakaryocytic lineage. In addition, as already presumed in these assays, double immunostaining against Ku86 and CD34 could further clarify that in PV but not in the control biopsies a major subpopulation of CD34+ cells expressed Ku86 (Fig. 5).

thumbnail image

Figure 5. Ku86 and CD34 immunohistochemistry of archived paraffin embedded bone marrow biopsies. Ku86 and CD34 immunohistochemistry of archived paraffin embedded bone marrow biopsies from 5 polycythemia vera (PV) patients (a, c) and 5 controls (b, d). CD34 immunohistochemistry is demonstrated in A and B, Ku86 immunohistochemistry representing the identical area in C and D. The overall expression of Ku86 in PV patients clearly exaggerates Ku86 expression in the controls. In direct comparison with each other a coexpression of Ku86 and CD34 in PV patients seems likely. Immuno-doublestaining (CD34 and Ku86) substantiates strong membranous CD34 (red) and nuclear Ku86 (dark-gray) staining (e). Original magnifications ×400 (a-d) and ×1000 (e). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

In 2005, 5 groups independently published a point mutation of JAK2V617F resulting in an activated kinase activity without the need of ligand binding to hematopoietic receptors, and also effecting cytokine receptor expression and modulating epigenetic cell profile.3–7, 22 This mutation was described to occur in up to 97% of PV patients. However, the mutation is not specific for PV and is also seen in other Ph chronic myeloproliferative disorders (CMPD). In addition, individual cases of JAK2V617F mutations in other patients and even in healthy volunteers have been described.23–26 Finally, data published by others indicate that JAK2V617F represents a rather late event in disease progression.8, 9 Altogether, although detection of the JAK2V617F mutation is a milestone in the understanding of PV, this activating point mutation alone cannot explain the specific picture of this disease and other yet unknown mechanisms must account for the pathogenesis restricted in the stem cell compartment.2 In this context it has to be underscored that circulating CD34+ PBSC as a potential source for stem and progenitor cells can be easily achieved and consist of a higher number of stem and progenitor cells with a lower cell cycling activity compared to CD34+ bone marrow stem cells.27, 28 Therefore, comparative gene signatures of CD34+ PBSC, compared with bone marrow CD34+ stem cells, offers easy access by phlebotomy and reflects cells which are highly representative for examination of this hematological stem cell disease.

Employing this setting, we show for the first time distinct dysregulation of several genes in PV CD34+ PBSC important for regulation of apoptosis, cell cycle and DNA repair. Most notably, the upregulation of KU86 on mRNA and protein level together with downregulation of DNA-PKcs is an exciting finding and leads to further speculation of mechanisms in the pathophysiology of PV.

Performing cDNA microarray analyses we found gene expression of MAPKK3, an activator of p38 mitogen-activated protein kinase modulated apoptosis,29, 30 and STAT1, an activator of Fas and Caspase cascade dependent apoptosis,31 significantly downregulated. Furthermore, altered gene expression of additional factors involved in the fine tune regulation of apoptosis were detected (i.e., IL13 and JAK3).32–35 Apoptosis is affected by many signaling pathways and due to its importance fine adjustment is an essential prerequisite to cell fate. Imbalance of even a few or just one factor can lead to severe disorder. In this regard these experimental findings may explain longer live span of PV CD34+ PBSC.36–38

Downregulation of 6 pro apoptotic genes and upregulation of 1 anti apoptotic gene was already described in PV CD34+ bone marrow stem cells when compared with their healthy counterparts.39 Comparing these data with our results revealed that 2 genes (HSPB1 and IL13) were congruently dysregulated in both compartments (i.e., peripheral blood and bone marrow) and 4 genes (COMT, NUCB1, HSPA1A, and NTF3) exhibited a similar tendency but failed to exceed the twofold threshold in our study. However, there also are clear constitutional varieties of gene expression in both compartments.28 In this context, the finding of a prominent Ku86 protein expression by nearly all CD34+ PBSC but only by a subpopulation of bone marrow CD34+ cells therefore remains unsurprising.

Within the blood cells a significant upregulation of KU86 together with a downregulation of DNA-PKcs and a prominent upregulation of Ku86 protein was restricted to the PV CD34+ PBSC. In PV monocytes this complex was not regulated, although these cells are known to be clonal as well.4, 6, 40 However, in PV bone marrow biopsies a strong upregulation of Ku86 protein was also seen in maturing cells of the erythropoiesis and the megakaryopoiesis. As these cell lineages form the specific phenotype of the disease it is tempting to speculate that ongoing dysregulation of DNA-PK in maturing erythropoiesis and megakaryopoiesis might be responsible for their accumulation or at least reflects a today unknown phenomenon in PV.

Ku86 together with Ku70 builds a heterodimer which represents the regulatory subunit of the DNA-dependent protein kinase. This holoenzyme and also its unbound subunits are known to play a pivotal role within the cell and an imbalance of the subunits leads to several interferences in cell metabolism.10 Beside other functions Ku86 has been implicated to be involved in many nuclear processes such as DNA replication, transcription regulation, and DNA double-strand break repair by NHEJ in a directly or indirectly manner.10, 11 This repair mechanism rejoins free DNA ends within minutes of their occurrence, but this error-prone process may involve the loss or alteration of nucleotides.41 Therefore activated Ku86 is able to induce genetic abnormalities like deletions and end to end ligations in several hematological disorders.13, 14 Hence, the well-known elevated frequency of chromosomal aberrations including the JAK2V617F point mutation observed in chronic phase PV may be a primary effect of increased Ku86 action—even if this phenomenon could also be reactive and caused by previous DNA damage due to unknown effects.13, 42 Indeed, the relation between increased Ku86 activity and the JAK2V617F point mutation must be embedded in a multifactor event due to the specifity of this activating mutation in CMPD.

Beside its mutagenic potential, DNA-PK is known to play a pivotal role in stabilization of p53 protein resulting in G1 phase block and induction of apoptosis.43 Therefore, downregulation of DNA-PKcs together with the above mentioned dysregulation of several apoptosis related genes may be involved in the well-known prolonged cell survival in PV.36–38

Finally, Ku86 protein has been shown to directly affect telomerase activity by telomere capping and recruitment of telomerase.10, 11 An increase in telomerase activity protects cells and prevents apoptosis. In this context a direct involvement of telomerase activity in the hematopoietic progenitor cell fraction and in erythroid differentiation has already been demonstrated.44–46 Both compartments are typically altered and mark impressive hallmarks in PV leading to the assumption that the overexpression of Ku86 contributes to this specific phenotype.

Even it is not clear today whether the above mentioned changes in gene and protein expression are primary or reactive events in the pathophysiology of PV, these results may give new impetus in research of this field. Comprehensively, the differential expression of several genes, imbalance of the distinct subunits of DNA-PK in PV CD34+ PBSC, and particularly the upregulation of Ku86 protein are new findings in PV CD34+ PBSC. These factors may contribute to accumulation of chromosomal aberrations, accumulation of hematopoietic cells (especially of erythropoiesis) and prolonged CD34+ PBSC life span in PV. Our observations of dysregulation in important pathways such as apoptosis, hematopoiesis, cell cycle or DNA repair at the stem cell level open new avenues of research into unknown aspects of PV pathogenesis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

This work was accomplished in the Institute of Pathology, University of Cologne.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_23985_sm_Suppfig1.tif1564KSupporting Figure 1 Allelic distribution of the JAK2V617F point mutation revealed by allele-specific QPCR. PCR product sequences (GATC) of the JAK2V617F locus (complementary sequence) of all PV-patients under study, of the cell line HEL (DSMZ, Braunschweig, Germany) as homozygote control and of one healthy individual. Additionally, results of allele-specific multi-QPCR with minor grove binder probes (Applied Biosystems) of the three allelic situations (A) homozygote JAK2V617F negative. Lower left picture: FAM-labeled amplification of the non-mutated allele with no amplification of mutated allele (linear graph). Lower middle picture: Half logarithmic graph. Lower right picture: linear graph of the VIC-labeled mutated PCR product (no amplification); (B) homozygote JAK2V617F positive; (C) heterozygote for JAK2V617F point mutation. All nine patients under study revealed a heterozygote mutational status of the highly purified CD34+ PBSC by QPCR and by PCR product sequencing.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.