Stringent Regulation of DNA Repair During Human Hematopoietic Differentiation: A - 《stemcells》干细胞学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 27 下一页

返回列表

查看: 408925\|回复: 263	go Stringent Regulation of DNA Repair During Human Hematopoietic Differentiation: A [复制链接]

江边孤钓

版主

Rank: 7 Rank: 7 Rank: 7

积分: 0
威望: 0
包包: 483

楼主

发表于 2009-3-5 00:09 |只看该作者 |倒序浏览 |打印

作者：Tomke U. Brackera, Bernd Giebelc, Jan Spanholtzc, Ursula R. Sorgb, Ludger Klein-Hitpassa, Thomas Moritzb, Jrgen Thomalea . p8 V, w0 e, H8 u* A2 E5 h! [; e
   # a" P8 c& T6 Q
   ( ^8 |9 Z( \/ Y) B+ q) d





         : L6 y! E: o' B: p" ~% K' ]* @, U
                  1 v4 [0 A& d* L8 O& Y  L
      ' N  Z  X5 s/ ^4 |
      % N2 A! y8 j: k
      6 B! S* g$ I9 `% H: ~  y
      【摘要】/ b$ D2 F: }; }1 N, R( \
   For the lymphohematopoietic system, maturation-dependent alterations in DNA repair function have been demonstrated. Because little information is available on the regulatory mechanisms underlying these changes, we have correlated the expression of DNA damage response genes and the functional repair capacity of cells at distinct stages of human hematopoietic differentiation. Comparing fractions of mature (CD34¨C), progenitor (CD34 38 ), and stem cells (CD34 38¨C) isolated from umbilical cord blood, we observed: 1) stringently regulated differentiation-dependent shifts in both the cellular processing of DNA lesions and the expression profiles of related genes and 2) considerable interindividual variability of DNA repair at transcriptional and functional levels. The respective repair phenotype was found to be constitutively regulated and not dominated by adaptive response to acute DNA damage. During blood cell development, the removal of DNA adducts, the resealing of repair gaps, the resistance to DNA-reactive drugs clearly increased in stem or mature compared with progenitor cells of the same individual. On the other hand, the vast majority of differentially expressed repair genes was consistently upregulated in the progenitor fraction. A positive correlation of repair function and transcript levels was found for a small number of genes such as RAD23 or ATM, which may serve as key regulators for DNA damage processing via specific pathways. These data indicate that the organism might aim to protect the small number of valuable slow dividing stem cells by extensive DNA repair, whereas fast-proliferating progenitor cells, once damaged, are rather eliminated by apoptosis.
      【关键词】 Comet assay Ethylnitrosourea EtNU Melphalan Nucleotide excision repair Base excision repair Hematotoxicity
   INTRODUCTION

The cellular capacity to remove drug-induced lesions from the nuclear DNA strongly determines the sensitivity of normal and malignant hematopoietic cells to DNA-reactive agents, such as alkylator-type antineoplastic drugs . However, little information is available on the regulation of this network, the exact role of individual pathways, or crosstalk between pathways in the lymphohematopoietic system and other primary tissues.

In human leukocytes generated from peripheral blood or umbilical cord blood samples, a broad spectrum of individual DNA repair phenotypes has been observed regarding overall repair capacity and distinct repair functions . Additionally, we have correlated their regulatory pattern to the kinetics of DNA damage processing in these cells.
; S* i% v6 M2 e4 c  _7 M
Figure 1. Schematic outline of major DNA damage repair pathways and functional assays. Primary adducts are detected by the immunocytological assay (ICA) and secondary lesions by the comet assay. Abbreviations: BER, base excision repair; NER, nucleotide excision repair; GGR, global genomic repair; TCR, transcription coupled repair; MMR, mismatch repair; DSBR, double-strand break repair; HR, homologous recombination; NHEJ, nonhomologous end joining; SSB, single-strand break; DSB, double-strand break; MX, methoxyamine (a specific inhibitor of early BER).

MATERIALS AND METHODS, }8 o$ P& X1 P+ ~0 S

Preparation of Cells7 @/ R% a. u1 O4 V) A! O
; G+ R' C$ i) r0 v# F# N! r
Umbilical cord blood was obtained from individual donors after informed consent according to the Declaration of Helsinki. Mononuclear cells were isolated by Ficoll gradient centrifugation and resuspended in prewarmed Roswell Park Memorial Institute (RPMI) medium supplemented with 10% fetal calf serum (FCS; PAA Laboratories, Linz, Austria, http://www.paa.at). CD34  cells were purified by immunomagnetic isolation using the indirect MidiMACS technique (Miltenyi Biotech, Bergisch Gladbach, Germany, http://www.miltenyibiotec.com). The purity of CD34  fractions was controlled by flow cytometry and averaged 78% in samples for functional assays and 85% for gene expression analysis. CD34¨C cells were taken from the column flow-through.( b) y+ J* R1 r- x; c" V6 s- u! k

For functional analysis the primitive, stem cell-enriched fraction of CD34 38low cells and the progenitor-enriched CD34 38  cell fraction were isolated from preselected CD34  cells by fluorescence-activated cell sorting (FACS). After incubation with CD34-fluorescein isothiocyanate (FITC), CD38-PE, and CD45-PerCP-Cy5.5 antibodies (BD Pharmingen, San Diego, http://www.bdbiosciences.com/pharmingen; 15 minutes in phosphate-buffered saline (PBS)/1% FCS), cells were washed in PBS and sorted by using a FACSVantage (BD Biosciences). The purity of CD34 38low and CD34 38  fractions was >95%. During and after drug exposure, cells were kept in supplemented RPMI medium at 37¡ãC in a humidified atmosphere containing 5% CO2.5 y, f7 b9 W5 q! `+ L
* o0 `# M6 U9 M. }1 _) P0 j
For comparative microarray expression analysis of primitive and progenitor cells, CD34 38low and CD34 38  cells were isolated from the CD34 -enriched fraction by immunostaining with CD3-FITC, CD14-FITC, CD16-FITC, CD19-FITC, CD20-FITC, and CD56-FITC (lin1 antibody cocktail; BD Biosciences), as well as glycophorin A (GA)-FITC, CD38-PE, and CD34-PeCy5 antibodies (BD Pharmingen). lin1¨CGA¨CCD34 CD38low and lin1¨CGA¨CCD34 CD38  cells were highly purified using a Coulter EPICS Elite ESP fluorescence cell-sorting system equipped with the Expo32 software (Beckmann Coulter) with lin1¨CGA¨CCD34 CD38low comprising one-sixth of total lin1¨CGA¨CCD34  cells. Separated cell fractions were frozen in TRIzol (Invitrogen, Carlsbad, CA, http://www.invitrogen.com) and stored at ¨C80¡ãC until RNA was prepared.( @  e1 K3 w9 G( }# a. h6 {5 G2 V

Exposure to Drugs and Apoptosis Assay1 M, k! {  f7 X3 Y  W

For the measurement of repair kinetics, cells were exposed to N-ethylnitrosourea (EtNU; Sigma, 100 µg/ml) for 30 minutes or to melphalan (Alkeran; 10 µg/ml; GlaxoSmithKline, Research Triangle Park, NC, http://www.gsk.com) in RPMI medium (10% FCS, 5 mM HEPES) for 2 hours at 37¡ãC. Thereafter, cells were washed in PBS, resuspended in prewarmed RPMI, and incubated at 37¡ãC. Cell aliquots were taken prior and at different time points after drug treatment. For expression profile analysis, EtNU-exposed cells were postincubated for 2 hours prior to RNA isolation. For repair inhibition studies, cells were preincubated with 1 mM methoxyamine (MX; Sigma) for 1 hour prior to EtNU exposure and throughout the experiment. Drug-induced apoptosis in cells was measured 24 hours after the addition of cisplatin (Platinex; Bristol-Meyers Squibb, New York, http://www.bms.com), EtNU, or melphalan to the medium. The fraction of apoptotic cells was determined by annexin V-FITC staining (Annexin V Detection Kit I; BD Pharmingen) and FACS analysis.

Comet Assay- l/ i4 Z; k, M9 Z! q( K& \7 D

DNA strand breaks in individual nuclei of small cell fractions were measured by single-cell gel electrophoresis ("comet assay") modified according to McNamee et al. . In brief, from 8-well cell culture chamber slides (BD Falcon, Franklin Lakes, NJ, http://www.bdbiosciences.com) the glass bottom was removed and replaced by GelBond film (Cambrex, Walkersville, MD, http://www.cambrex.com; Biozym, Hess, Oldendorf, Germany, http://www.biozym.com). Aliquots of 104 cells were suspended in 45 µl of low-melting point agarose (0.75% in PBS, prewarmed at 42¡ãC; Metaphor; Biozym) and cast into the wells. After coagulation the frames were removed, and the gels on the film were soaked overnight at 4¡ãC in lysis buffer (2.5 M NaCl, 100 mM EDTA, 10 mM Tris, 10% dimethyl sulfoxide, 1% Triton X-100, 1% n-laurylsarcosinate, pH 10). Nuclear DNA was denatured by alkaline treatment (300 mM NaOH, 1 mM EDTA, 10 mM Tris-HCl, pH 12.7) for 15 minutes, and GelBond films were subjected to alkaline electrophoresis in the same buffer (20 minutes, 4¡ãC, 1.5 V/cm). After neutralization (30 minutes, 400 mM Tris-HCl, pH 7.5), gels were dehydrated in absolute ethanol (1 h) and air-dried. Before evaluation of comet formation, gel films were rehydrated, and the nuclear DNA was stained with SYBR-Green (dilution 1:10,000; Roche Diagnostics, Basel Switzerland, http://www.roche-applied-science.com).
  I; }. o+ i0 E; p9 t( O7 z: s
Immunocytological Assay

Immunoanalytical measurement of melphalan-induced adducts in the nuclear DNA of individual cells was performed as described  with minor modifications: 104 cells/sample were applied to precoated microscopic slides (ImmunoSelect; Squarix, Marl, Germany, http://www.squarix.de) and immuno-stained for melphalan-DNA adducts with rat monoclonal antibody Amp 4¨C42 (kindly provided by Dr. M. J. Tilby, University of Newcastle upon Tyne, Newcastle upon Tyne, U.K.). Binding of primary antibody was visualized by consecutive staining with rabbit anti-(rat Ig) and goat anti-(rabbit Ig), both labeled with Cy3 (Dianova).

Quantitative Image Analysis and Statistics
: L# M8 k5 Y2 g
Comet assay and immunocytological assay (ICA) were evaluated by quantification of fluorescence signals using a photomicroscope (Axioplan; Zeiss, Jena, Germany, http://www.zeiss.de) and a multiparameter image analysis system (ACAS; Ahrens Electronics, Bargteheide, Germany). Melphalan adduct levels of individual cell nuclei were calculated by normalizing the antibody-derived fluorescence signals for the DNA content of the same cell. The relative amount of DNA strand breaks (comet assay) was determined using the olive tail moment . Mean signal values (¡ÀSEM) were calculated from >100 individual cells per sample. Data of corresponding cell pairs from the same donor were analyzed by paired t test.

RNA Preparation

Cells were homogenized using a QIAShredder column (Qiagen, Hilden, Germany, http://www1.qiagen.com), and total RNA was isolated according to the manufacturer¡¯s instructions. RNA concentrations were measured by fluorescence staining (Ribo-Green Kit; Molecular Probes) using a microplate reader, and RNA quality was verified by electrophoresis in 1% agarose gels. Due to the small number of cells in the CD34 38low fraction, isolated RNA from four individual cord blood samples was pooled and compared with pooled RNA from the CD34 38  fractions of the same four donors.- z( r7 @- z! t2 ^" L8 n

RNA Amplification

To obtain sufficient amounts of labeled material for the Gene-Chip hybridization, RNA samples from individual CD34  or CD34¨C cell fractions were subjected to a two-round amplification procedure according to Baugh et al.  for 16 hours at 37¡ãC. Reactions were treated with 2 U of RNase-free DNase I (Ambion) for 30 minutes at 37¡ãC before purification on spin columns (Roche) and quantitated by optical density measurement. For second-round cDNA synthesis, 500 ng of first-round amplification products was used in all cases. Reverse transcription with SuperScript II was performed with 0.5 µg of random hexamer primer (Stratagene) in 15-µl reactions without T4gp32 as described above. Following heat inactivation, RNA templates were removed by digestion with 2 U of RNase H for 30 minutes at 37¡ãC. After annealing of T7-dT21V primer (100 ng) at 42¡ãC for 5 minutes, reactions were snap-cooled in ice water. Second-strand cDNA synthesis was performed in a final volume of 100-µl reactions (1x Second-strand buffer, 0.2 mM dNTPs, 1 U of RNase H, 20 U of E. coli DNA polymerase I) for 2 hours at 16¡ãC and trimmed with 10 U of T4 DNA polymerase for another 15 minutes at 16¡ãC. CDNA was purified, precipitated, and transcribed with T7 RNA polymerase as described above, except that ribonucleotide concentrations were 4 mM each for GTP and ATP, 1.4 mM each for CTP and UTP, and 0.6 mM each for biotin-11-CTP and biotin-11-UTP (PerkinElmer Life and Analytical Sciences, Boston, http://www.perkinelmer.com). Pooled samples of lin¨CCD34 38  or lin¨CCD34 38low-derived RNA (500 ng) were amplified by one round of cDNA synthesis using the MessageAmp II aRNA kit (Ambion) and in vitro transcription in the presence of biotinylated NTPs as described above.( E" M9 Y1 v8 Q, U1 v" I- x
8 E" w( g/ v" x; `5 j
Oligonucleotide Microarray Analysis
) Z& S) F. D5 V( }- a3 H
Biotin-labeled cRNA was purified on RNeasy columns (Qiagen), fragmented, and hybridized to HG-U133A GeneChips (Affymetrix, Santa Clara, CA, http://www.affymetrix.com) following the Affymetrix standard protocol. The arrays were washed and stained according to the manufacturer¡¯s recommendation and finally scanned in a GeneArray scanner 2500 (Agilent, Palo Alto, CA, http://www.home.agilent.com). Array images were processed to determine signals and detection calls (present, absent, and marginal) for each probe set using the Affymetrix Microarray Suite 5.0 software. Scaling across all probe sets of a given array to an average intensity of 1,000 was performed to compensate for variations in the amount and quality of the cRNA samples and other experimental variables of nonbiological origin.

Analysis of Microarray Data

For unsupervised hierarchical clustering, signals of individual probe sets were normalized to the mean probe set signal of all included arrays and log transformed. Log transformed ratios were subjected to UPGMA clustering using correlation as similarity measure (Spotfire DecisionSite for functional genomics). As additional criteria we used the present calls in  30% of the samples and a ratio of means of 1.5 or 0.67. To compare corresponding pairs of CD34 /CD34¨C or CD34 38 /CD34 38low cells from the same donor for the magnitude and direction of change, we employed the signal log ratio (SLR) algorithm giving the differences as binary logarithmic values.9 E9 r  X) E) b  \5 {6 l4 V
# {" k5 T- A' n) @
Quantitative Reverse Transcription-Polymerase Chain Reaction

For real-time polymerase chain reaction (PCR) analyses, total RNA was reverse-transcribed using random primers (High Capacity cDNA Archive Kit; Applied Biosystems). PCR was carried out in duplicate 20-µl reactions containing cDNA corresponding to 5 ng of total RNA, 1 µl of Taqman-based assay, and 1x master mix reagents (Applied Biosystems). PCR was performed on an ABI Prism 7900HT system as recommended by the manufacturer using the glyceraldehyde-3-phosphate dehydrogenase assay (Hs99999905_m1) as the endogenous reference and ATM (Hs00175892_m1), RAD23A (Hs00192541_m1), and RAD50 (Hs00194871_m1) as target assays. Differential expression was estimated by the comparative Ct method (ABI Prism 7700 Sequence Detection System User Bulletin #2: Relative Quantification of Gene Expression ).' M+ d& e. f7 y9 Y- H) Z8 A

RESULTS7 E7 @5 ?2 Q# F
+ {7 t+ Q' i+ x+ K
Expression of Damage Response Genes in CD34  and CD34¨C Cells$ f  I7 ~. ^5 f4 C7 e) g: j
# }4 o1 b! S3 A- [) W3 g
To assess the transcriptional activity of genes involved in DNA damage response at different stages of lymphohematopoietic differentiation, CD34  progenitor cells and their mature CD34¨C counterparts were prepared from seven individual cord blood samples. Total RNA was isolated, and gene expression profiles were determined using Affymetrix HG_U133A GeneChips. Hybridization signals from 1418 probe sets, representing a total of 803 DNA damage response genes, were evaluated. By our criteria (present calls in 30% of the samples), transcripts for 633 (79%) of these genes were detected. This set comprised 296 of 366 genes related to cell-cycle control, 254 of 330 genes related to apoptosis, and 153 of 189 genes related to DNA repair functions with some genes listed in more than one subgroup. Pronounced interindividual variances in specific transcript levels of DNA repair genes were observed at both stages of differentiation. For instance, the amount of mRNA transcribed from the mismatch repair gene MSH2 varied at a maximum range of 7.5-fold (fluorescence signals: 286¨C2,145) among CD34  and 4.6-fold (signals: 190¨C882) among CD34¨C cell samples. The corresponding values for RAD23A, a gene involved in nucleotide excision repair, were 6.7-fold (signals: 764¨C5,142) and 4.1-fold (signals: 2,909¨C12,024), respectively. For the majority of genes, interindividual variations were found in a 2.5- to 3.5-fold range, with a tendency to broader variations in progenitor cells. On the other hand, a small number of genes was expressed more constantly, showing less than 2-fold inter-individual variations within either of the cell fractions. One member of this group was the XPA gene coding for a key component of both sub-pathways of nucleotide excision repair (NER), that is global genomic (GGR) and transcription-coupled repair (TCR).
$ K$ a7 l. e% S
Despite the pronounced interindividual variation at the level of specific transcripts, stringently regulated, differentiation-dependent shifts in gene expression profiles were observed. Unsupervised hierarchical cluster analyses employing the whole set of damage response genes revealed a clear-cut separation between progenitor and mature blood cells (Fig. 2A). A similar pattern was found when focusing on genes associated with DNA repair functions or apoptosis (Fig. 2B, 2C), indicating that these genes are stringently regulated during hematopoietic cell development and that this regulation dominates interindividual variability in gene expression.4 b  P# J! W4 G$ U( d4 o3 f/ S
9 C. Y  L7 h) K$ p8 J4 e/ z  e
Figure 2. Expression profiles of DNA damage response genes separate mature (CD34¨C) and progenitor (CD34 ) cord blood cells. Fourteen RNA samples from seven corresponding pairs of hematopoietic cells were subjected to GeneChip analysis. Signals from individual probe sets were normalized to the mean and log-transformed, and ratios were analyzed by unsupervised hierarchical clustering (see Materials and Methods; relative expression: red ). The dendrograms and matrices represent the clustering of 962 probe sets (DNA damage response) (A), 102 probe sets (DNA repair) (B), and 219 probe sets (apoptosis) (C). Criteria for the depicted probe sets in (A): present call in 30% of the CD34  or the CD34¨C samples; additional in (B), and (C): ratio of means 1.5 or 0.667.

When transcript levels of individual pairs of CD34  and CD34¨C cells from the same donor were compared using the SLR algorithm, 37% (291/803) of the DNA damage response genes were found to be significantly up- or downregulated (cut-off: p values of SLRs  0.005 or  0.995 for at least four of seven cell pairs). More than half (175) of these genes displayed higher mRNA levels in the corresponding progenitor cell fraction. Among the genes related to DNA repair mechanisms, 58 of 153 detectable gene products were up-and 10 were downregulated in CD34  cells. When focusing on 97 genes coding for constituents of major DNA repair pathways  (see Fig. 1), 37 of 83 detectable genes were differentially expressed, namely 10/28 in base excision repair (BER), 13/38 in NER, 7/16 in mismatch repair (MMR), and 12/33 in double-strand-break repair (DSBR; Fig. 3). Most of these genes showed significantly higher transcript levels (mean, 2.6-fold) in the corresponding progenitor cell fraction. Only two genes (ATM and RAD23A) were identified as clearly downregulated in the progenitor fraction of all seven samples. These genes are employed in DSBR or NER functions, respectively.

Figure 3. Most differentially expressed genes contributing to major DNA repair pathways are consistently upregulated in progenitor compared with mature cells. SLRs of 37 differentially expressed genes with function in one or more of the major repair pathways: BER (A), NER (B), MMR (C), and DSBR (D). White bars show the SLRs of individual CD34 /CD34¨C cell pairs from the same donor; black bars represent the mean SLRs of all seven samples (criterion for significant up- or down-regulation: p  .005 or .995 for at least 4/7 CD34 /CD34¨C cell pairs). Abbreviations: SLR, single log ratio; BER, base excision repair; NER, nucleotide excision repair; MMR, mismatch repair; DSBR, double-strand break repair.

The array data were validated by real-time reverse transcription (RT)-PCR reanalysis of RNA from five CD34 / CD34¨C cell pairs and three different transcripts (ATM, RAD23A, and RAD50). In each single case the same magnitude and direction of regulation was measured by both assays (mean SLRs: ATM, ¨C1.78 and ¨C1.65; RAD23A, ¨C2.05 and ¨C1.64; and RAD50, 0.75 and 0.98 for RT-PCR and microarray analysis, respectively).
/ i: b; s1 N5 r0 C8 m* t( m
Functional Analysis of Cellular DNA Repair and Correlation with Gene Expression Profiles8 T3 Q# k) m4 V+ l0 f6 {4 R
3 ]+ m, M" V4 a& q5 o
The overall capacity of hematopoietic cells to process DNA damage was measured by single-cell gel electrophoresis ("comet assay"), which determines repair-induced DNA strand breaks in individual cell nuclei and by ICA allowing the direct measurement of drug-induced DNA adducts (Fig. 1). To induce a set of structurally defined DNA adducts, CD34  or CD34¨C cell fractions in liquid holding were exposed to a short pulse of EtNU, a fast-reacting monofunctional alkylator (t1/2 in cells: 7 minutes), which is not subject to active drug transportation. EtNU interacts with the cellular DNA to form about a dozen different ethylation products (among them 15% N7-guanine, 9% O6-guanine, 9% O2-/O4-thymine, 4% N3-adenine, and 3% O2-cytidine) , which simultaneously trigger repair responses via various pathways such as NER, BER, and MMR or direct removal by the alkyl-DNA alkyltransferase O6-methylguanine-DNA methyltransferase (MGMT).0 N  ]$ O# X0 E; o) R2 U- v

For quantitative evaluation of the repair kinetics by the comet assay, two parameters were utilized: 1) the amount of DNA strand breaks present directly after a 30-minutes period of drug exposure (OTM, t0: representing the efficiency of initial repair incision) and 2) the slope of the repair curve (OTM/t: depicting the efficiency of gap filling in religation steps) (Fig. 4A). Although analysis of both parameters confirmed the high interindividual variance in the DNA repair capacity of hematopoietic cells at the functional level, repair kinetics between mature and progenitor cells from the same donor differed markedly, too (Fig. 4B, 4C). Fewer early incisions into adducted DNA (8/8 samples, p = .004) and slower religation of repair gaps (7/8 samples, p = .018) both indicate less efficient repair processing of EtNU-induced DNA lesions in CD34  progenitor cells.3 H+ l% j8 h% ~2 j

Figure 4. Processing of DNA lesions in hematopoietic cells shows high interindividual variation and is impaired in progenitor compared with stem and mature cells. Following exposure to N-ethylnitrosourea (EtNU; 100 µg/ml; 30 minutes), the kinetics of repair-induced DNA strand breaks were determined by the comet assay (measurement of OTM) in CD34  or CD34¨C cells isolated from eight cord blood samples, as well as in CD34 38low or CD34 38  cells from four individual samples. (A): Two parameters for the cellular repair capacity, the efficiency of initial strand incision (OTM, t0), and the velocity of repair gap resealing (OTM/minutes) were determined. (B, C): Both parameters were significantly lower in the related CD34  cell fractions (paired t test: p = 0.004 and 0.018, respectively). Statistical evaluation is given in the box (25¨C75 percentile) and whisker (5¨C95 percentile) plots. (D): Kinetics for CD34 38low and CD34 38  cell fractions from an individual donor. (E, F): In addition to pronounced differences between four individual samples, the frequency of initial DNA strand breaks and repair velocity were consistently higher in the stem cell fraction. Abbreviation: OTM, olive tail moment.; f; [5 B; T4 y4 u7 x' D% D  o

To corroborate these findings for a different set of DNA lesions, corresponding pairs of isolated cell fractions were challenged with the alkylating anticancer drug melphalan (10 µg/ ml), which initially forms mono-adducts and, subsequently, interstrand crosslinks, preferentially at the N7-position of guanine. The level of primary DNA adducts, as measured 2 hours after drug exposure by quantitative immunocytological analysis, was significantly higher in progenitor compared with mature cells (mean, 1.7-fold, n = 3, p
9 k6 A% V- S' f
To link the removal of primary and secondary DNA lesions to chemosensitivity, the onset of apoptosis after exposure to EtNU, melphalan, or cisplatin was investigated (Table 1). A consistently higher apoptotic response in progenitor versus mature cells was observed, thus backing up our earlier observation of accelerated induced cell death after treatment with DNA-damaging agents in progenitor cells .( }; E9 {( T: G. D( h

Table 1. Higher frequencies of apoptosis in progenitor versus mature cord blood cells following ex vivo exposure to DNA-reactive drugs
5 L% I4 m0 {; N* {) A
As both repair functions and drug sensitivity are very likely affected by DNA replication, we have determined the cycle distribution in the progenitor cell fraction immediately after isolation. DNA histograms (FACS analysis) of 4,6-diamidino-2-phenylindole-stained cells confirmed observations of other groups that nearly all (>98%) CD34  cord blood cells are in G0/G1 .! t+ N! b' n$ l

To address the relative contribution of individual pathways to the overall repair activity, the pharmacological inhibitor MX was employed, which prevents the strand incision as an early step along the BER pathway  (Fig. 1). Thus, the proportion by which MX reduces the frequency of DNA strand breaks after drug treatment estimates the contribution of BER to the overall repair capacity of a cell. When comparing CD34  and CD34¨C cell pairs after exposure to EtNU alone or combined with MX, the relative contribution of BER was consistently (4/4 samples) higher in the CD34  subset than in mature cells (mean ¡À SEM: 51 ¡À 7% and 21 ¡À 12%, respectively). This observation is in agreement with the augmented mRNA levels of all (10/10) differentially expressed BER genes in the progenitor cell compartment (Fig. 3A).
. t% {" ~( J2 T0 l1 K* i' m) ^
No Immediate Transcriptional Response to EtNU-Induced DNA Damage
! s4 m7 h, S2 G5 X2 P* v% n5 r
It has been shown that exposure of human cell lines to DNA-damaging agents can induce a significant shift in the expression profile of DNA damage response genes . Therefore, the functional differences in repair capacity, which we observed between CD34  and CD34¨C cells, could be due to cell type-specific rapid up- or downregulation of DNA damage-related genes shortly after drug exposure. To investigate this possibility, we have compared gene expression profiles in both cell fractions prior to and 2 h after exposure to EtNU. However, no major shifts in mRNA levels upon initial DNA damage were observed either in mature or in progenitor cells (Fig. 5A, 5B). The changes induced by EtNU treatment were by far less pronounced than those observed between cells of different maturation status (Fig. 5C).

Figure 5. Lack of transcriptional response in mature and progenitor hematopoietic cells early after DNA alkylation stress. Scatter plots of the hybridization signals from 1,418 probe sets (representing 803 DNA damage response genes); comparison of cord blood¨Cderived cells prior to and 2 hours after ex vivo exposure to EtNU (100 µg/ml) CD34  (A); CD34¨C (B); and CD34  versus CD34¨C, both untreated (C). Abbreviation: EtNU, N-ethylnitrosourea.- ]. N" _0 Q. P% q# f& S/ K! l

Functional Repair Analysis and Expression Profiles in Primitive CD34 38low Cells0 i2 }' ~( u' j! d4 E1 T, _
1 |# m- n+ R' E1 n, W  G9 o
To analyze DNA repair capacity and DNA damage response in very early stages of hematopoietic differentiation CD34 38low cells were utilized. Repair kinetics after challenge with EtNU revealed consistently (4/4 samples) higher frequencies of initial strand breaks and faster resealing of repair gaps in primitive CD34 38low cells compared with the related progenitor cells (Fig. 4D¨C 4F). To reconcile functional analysis and gene expression profiles, mRNA from lin1¨CCD34 38low cells was subjected to microarray analysis. Due to the low cell number in this fraction, RNA from four individual cord blood samples was pooled. Analyzing the set of the 803 DNA damage response genes or the 189 genes related to DNA repair, 157 (20%) or 47 (25%) of them were differentially expressed in CD34 38low versus CD34 38  cells. Among the 97 genes contributing to the major DNA repair pathways, 19 were significantly up- or downregulated. Interestingly, and in contrast to the higher repair capacity of primitive CD34 38low cells, the vast majority of those genes (16/19) had reduced transcript levels in the stem cell fraction (Fig. 6). Again, ATM was among the few genes for which transcript levels were positively correlated to the overall repair capacity. This unique regulatory status was confirmed by RT-PCR analysis in three independent individual pairs of CD34 38 / CD34 38low cells (SLRs for ATM: ¨C1.84, ¨C0.79, and ¨C2.28 by RT-PCR vs. ¨C0.3 by array analysis of pooled RNA)., y& S% I  L9 q5 }+ u
/ U# u! G! z, w$ T1 j. n
Figure 6. Genes for major DNA repair functions are differentially expressed in primitive CD34 38low and CD34 38  progenitor cells. Up- or downregulation of transcripts (single log ratio ; see Fig. 3) as measured in pooled RNA samples from four cord blood specimens (criterion for significance: p values of SLRs .005 or  .995). Abbreviations: BER, base excision repair; NER, nucleotide excision repair; MMR, mismatch repair; DSBR, double-strand break repair.+ V, a9 ]2 R. ^! z; d

DISCUSSION% i- E. ^. j. Q0 S7 a/ v

Although pharmacotherapy with DNA-reactive agents has represented a cornerstone of systemic antineoplastic therapies for more than half a century, crucial steps of the molecular mechanisms determining the therapeutic potential, but also the side effects of these drugs, have remained ill-defined. We and others have recently introduced the cellular DNA repair capacity as an important modulator of drug susceptibility within the lymphohematopoietic system. In these studies, accentuated hematotoxicity and increased apoptotic response were associated with a consistently lower DNA repair capacity in human CD34  progenitor cells compared with their mature CD34¨C counterparts .

We now have employed gene array technology to allow a correlation of functional and gene expression analysis and have extended our studies to the hematopoietic stem cell compartment. We show that the differentiation-dependent alterations in DNA repair function and apoptotic response are accompanied by characteristic and consistent shifts in the expression profiles of related genes. Although genome-wide transcriptional signatures have been described for distinct stages of lymphohematopoietic differentiation . However, no comparable data are available on distinct developmental stages in the human system. The study presented here now reveals a strict differentiation-dependent regulation within the DNA damage response network of human hematopoietic cells. The uniform shift in transcript levels of regulated genes, up or down, in all seven cord blood specimens investigated implies a high biological relevance for this observation.9 \: F7 \& R& h2 q! g+ Z
5 i; @1 p, ~. k) X  F1 J
Despite this stringent regulation during development, intriguingly high interindividual differences were observed in repair gene expression, as well as functional repair capacity. Similar individual variations were detected for the induction of apoptosis following cytotoxic drug exposure (Table 1) or toxicity of chloroethylnitrosourea- or triazine-type alkylating agents for human clonogenic progenitor cells (personal observation). It can be speculated also that the substantial differences in antineoplastic response or hematotoxic side-effects observed with DNA-damaging agents in the clinical situation may be related to the individual DNA repair capacity of malignant or physiologic cells, respectively. Thus, dosage based on prior assessment of DNA repair capacity in tumor cells, as well as normal tissue, may represent a way to individualize antineoplastic therapy and improve results.

Rather surprising results were obtained when functional and expression profiling data were compared. Despite their significantly lower capacity to remove DNA adducts and secondary strand breaks, progenitor cells contained higher mRNA levels than their maturated progeny for 35/37 differentially expressed pathway-related repair genes. Comparing the progenitor and stem cell fractions, a similar discrepancy was noted. Here, reduced expression for 16/19 pathway-related genes in primitive cells was associated with higher repair efficiency. A possible explanation for this observation could be the impact of post-transcriptional control instances on the performance of the repair system such as ubiquitination-triggered protein turnover . In addition, low abundant gene products not detectable by oligonucleotide arrays may represent rate-limiting "bottleneck" positions along a given repair pathway and significantly determine its functional activity. Thus, important regulatory functions might be associated with the substantial number of repair genes not qualifying for present calls in our analyses (14/97 in mature vs. progenitor cell analysis, 21/97 in progenitor vs. stem cells).

On the other hand, the few pathway-related genes with shifts in transcript levels being conform to the alterations in functional repair may be essential for fine-tuning the cellular response to DNA damage. One of these genes encodes for ATM, a protein kinase that senses and signals the presence of DNA lesions, in particular double-strand breaks, to essential checkpoints and initiates rejoining .

The other gene with significantly diminished transcript levels in progenitor versus mature cells is RAD23A. The RAD23 proteins are involved in the regulation protein turnover via the ubiquitin/proteasome pathway . Interestingly, RAD23A is among the repair genes with the broadest interindividual variation of transcript levels in the cord blood samples analyzed and thus may be a key factor in controlling the individual repair capacity. Its relevance for the handling of DNA alkylation products is further strengthened by our observation that human XPC-RAD23 complexes can recognize EtNU-induced adducts in DNA and are essential for their NER-mediated excision in human lymphoid cells (unpublished data). Also in line with the downregulation of RAD23A and the reduced NER activity in progenitor cells is the higher relative contribution of BER (no downregulated constituents) to their overall repair capacity./ A% ~9 u  C0 t' e

Thus, in contrast to expression levels of the majority of genes involved, the least efficient DNA repair during hematopoietic differentiation resides within the progenitor cell compartment, whereas increased capacity is observed in more mature as well as more primitive cells. At first sight this may appear surprising, as an organism can be expected to equip its pool of particularly proliferation-competent cells with protective mechanisms to counteract DNA damage and the emergence of mutated daughter cells. From the perspective of the organism, however, eliminating damaged cells via apoptosis rather than attempting to restore their genomic integrity also represents an efficient defense mechanism against genotoxic stress in critical cell compartments. A well-known example of this principle is intrathymic T-cell development, where cells expressing nonfunctional T-cell receptors are eliminated via apoptosis . Interestingly, the relative repair deficiency of ES compared with fully differentiated cells was not accompanied by an increased mutation rate, and this was due to an effective induction of cell death programs even at low levels of DNA damage. These data suggest that apoptosis in stem cells is triggered preferentially by global genomic damage, whereas transcription-blocking lesions in active genes may represent the critical signals in maturated cells. The model is in agreement with our observations in the human lymphohematopoietic system, where the proliferation-competent progenitor cells exhibit reduced efficiency of global repair and increased apoptotic response upon exposure to DNA-damaging agents. Thus, this might be a general way to protect somatic "cell replenishment compartments" from the accumulation of genetic damage and thereby avoiding the expansion of mutated cells and their potential malignant transformation.

DISCLOSURES0 g% F0 A  K, N$ b' j; b5 |! ^

The authors indicate no potential conflicts of interest." w2 J; P1 u, g" f, n% I
1 i7 c$ s8 h* i/ U3 b
ACKNOWLEDGMENTS9 N' y/ [) L( F4 V5 D, @7 }* [
: x9 t# |9 k8 A& d+ ^3 y
We thank S. Niesert (Elisabeth-Krankenhaus, Essen, Germany) and G. Kögler and P. Wernet (Universitätsklinikum D¨¹sseldorf) for providing cord blood samples; M. J. Tilby (University of Newcastle upon Tyne) for the generous gift of anti-(melphalan-DNA adduct) antibodies; R. A. Hilger and M. Grubert (Universitätsklinikum Duisburg-Essen, Essen, Germany) for the measurement of intracellular melphalan levels; and A. Feldmann and M. Möllmann (Innere Klinik ) for excellent technical assistance. Research was supported by Wilhelm Sander-Stiftung Grant 1999.082.1 and in part by Bundesministerium f¨¹r Bildung und Forschung-Nationales Genomforshung-snetz Grant KR S06T06 and Mildred Scheel-Stiftung Grant 10-2039 to J.T. and T.M.9 H: b. X2 R& ~
      【参考文献】5 R7 r  O" L: \; M: ]
/ l) G) I4 i6 b& \

M¨¹ller MR, Buschfort C, Thomale J et al. DNA repair and cellular resistance to alkylating agents in chronic lymphocytic leukemia. Clin Cancer Res 1997;3:2055¨C2061.

Gerson SL. MGMT: Its role in cancer aetiology and cancer therapeutics. Nat Rev Cancer 2004;4:296¨C307.0 ], y8 V# ]" [) M. E

Spanswick VJ, Craddock C, Sekhar M et al. Repair of DNA interstrand crosslinks as a mechanism of clinical resistance to melphalan in multiple myeloma. Blood 2002;100:224¨C229.3 z, \2 \7 v- H, h, B

Christmann M, Tomicic MT, Roos WP et al. Mechanisms of human DNA repair: An update. Toxicology 2003;193:3¨C34.
+ i1 [: I  m2 v1 q$ x
Sancar A, Lindsey-Boltz LA, Unsal-Kacmaz K et al. Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem 2004;73:39¨C85.

Shiraishi A, Sakumi K, Sekiguchi M. Increased susceptibility to chemo-therapeutic alkylating agents of mice deficient in DNA repair methyl-transferase. Carcinogenesis 2000;21:1879¨C1883.( H- W- @2 l0 n  ~2 T; l
2 L% c4 {# C$ d5 d7 b) a
Reese JS, Liu L, Gerson SL. Repopulating defect of mismatch repair-deficient hematopoietic stem cells. Blood 2003;102:1626¨C1633.$ o& R% R* H6 m, u2 V* s- }# [

Essers J, van Steeg H, de Wit J et al. Homologous and non-homologous recombination differentially affect DNA damage repair in mice. EMBO J 2000;19:1703¨C1710.

Pollok KE, Hartwell JR, Braber A et al. In vivo selection of human hematopoietic cells in a xenograft model using combined pharmacologic and genetic manipulations. Hum Gene Ther 2003;14:1703¨C1714.
' a: n* D- h1 D6 b
Kobune M, Xu Y, Baum C et al. Retrovirus-mediated expression of the base excision repair proteins, formamidopyrimidine DNA glycosylase or human oxoguanine DNA glycosylase, protects hematopoietic cells from N,N',N''-triethylenethiophosphoramide (thioTEPA)-induced toxicity in vitro and in vivo. Cancer Res 2001;61:5116¨C5125.

Sawai N, Zhou S, Vanin EF et al. Protection and in vivo selection of hematopoietic stem cells using temozolomide, O6-benzylguanine, and an alkyltransferase-expressing retroviral vector. Mol Ther 2001;3:78¨C87.
( e8 M0 v' ^1 ~" G
Zielske SP, Reese JS, Lingas KT et al. In vivo selection of MGMT(P140K) lentivirus-transduced human NOD/SCID repopulating cells without pretransplant irradiation conditioning. J Clin Invest 2003; 112:1561¨C1570.$ k2 k( i( ~$ I1 Y6 A

Ragg S, Xu-Welliver M, Bailey J et al. Direct reversal of DNA damage by mutant methyltransferase protein protects mice against dose-intensified chemotherapy and leads to in vivo selection of hematopoietic stem cells. Cancer Res 2000;60:5187¨C5195.7 Y  ~! C2 \  M$ Z
, k8 c0 K1 j( r( d
Jansen M, Sorg UR, Ragg S et al. Hematoprotection and enrichment of transduced cells in vivo after gene transfer of MGMT(P140K) into hematopoietic stem cells. Cancer Gene Ther 2002;9:737¨C746.

Buschfort C, M¨¹ller MR, Seeber S et al. DNA excision repair profiles of normal and leukemic human lymphocytes: Functional analysis at the single-cell level. Cancer Res 1997;57:651¨C658.

Boffetta P, Nyberg F, Mukeria A et al. O6-Alkylguanine-DNA-alkyl-transferase activity in peripheral leukocytes, smoking and risk of lung cancer. Cancer Lett 2002;180:33¨C39.
) P4 U7 _: K* U. k, ~3 ], p( }
Vodicka P, Kumar R, Stetina R et al. Genetic polymorphisms in DNA repair genes and possible links with DNA repair rates, chromosomal aberrations and single-strand breaks in DNA. Carcinogenesis 2004;25:757¨C763.; R( l1 M4 g' s- f. |- _
& ^. w0 ?0 W( I0 j
Gerson SL, Phillips W, Kastan M et al. Human CD34  hematopoietic progenitors have low, cytokine-unresponsive O6-alkylguanine-DNA alkyltransferase and are sensitive to O6-benzylguanine plus BCNU. Blood 1996;88:1649¨C1655.
0 g6 X$ G+ Z% p7 n4 Q5 p3 k8 l
Terry MB, Gammon MD, Zhang FF et al. Polymorphism in the DNA repair gene XPD, polycyclic aromatic hydrocarbon-DNA adducts, cigarette smoking, and breast cancer risk. Cancer Epidemiol Biomarkers Prev 2004;13:2053¨C2058.9 R1 U- |, J' z( O* H  j  E
7 y  A, `3 [! O/ O0 _
Butkiewicz D, Popanda O, Risch A et al. Association between the risk for lung adenocarcinoma and a (¨C4) G-to-A polymorphism in the XPA gene. Cancer Epidemiol Biomarkers Prev 2004;13:2242¨C2246.. F8 |7 v3 ^5 Z. j% f( \

Myllyperkio MH, Vilpo JA. Increased DNA single-strand break joining activity in UV-irradiated CD34  versus CD34¨C bone marrow cells. Mutat Res 1999;425:169¨C176.
0 ~" O8 z. M3 r% Q0 u
Buschfort-Papewalis C, Moritz T, Liedert B et al. Down-regulation of DNA repair in human CD34( ) progenitor cells corresponds to increased drug sensitivity and apoptotic response. Blood 2002;100:845¨C 853.

Wood RD, Mitchell M, Sgouros J et al. Human DNA repair genes. Science 2001;291:1284¨C1289.

McNamee JP, McLean JR, Ferrarotto CL et al. Comet assay: Rapid processing of multiple samples. Mutat Res 2000;466:63¨C69.3 p" w8 |6 K# f2 u9 K% {' L

Olive PL, Banath JP, Durand RE. Heterogeneity in radiation-induced DNA damage and repair in tumor and normal cells measured using the "comet" assay. Radiat Res 1990;122:86¨C94.
; V2 @) z/ q& U( ~3 W  k. x+ P
Baugh LR, Hill AA, Brown EL et al. Quantitative analysis of mRNA amplification by in vitro transcription. Nucleic Acids Res 2001;29:E29.# v3 B# k) \( ^" m  G0 h: a

Beranek DT. Distribution of methyl and ethyl adducts following alkylation with monofunctional alkylating agents. Mutat Res 1990;231:11¨C30.  j' j1 B" D* A' j$ R7 f$ e

Summers YJ, Heyworth CM, De Wynter EA et al. Cord blood G0 CD34  cells have a thousand-fold higher capacity for generating progenitors in vitro than G1 CD34  cells. STEM CELLS 2001;19:505¨C513.# n+ K0 Q% Y: V

Rieger KE, Chu G. Portrait of transcriptional responses to ultraviolet and ionizing radiation in human cells. Nucleic Acids Res 2004;32:4786¨C4803.

Ramalho-Santos M, Yoon S, Matsuzaki Y et al. "Stemness": Transcriptional profiling of embryonic and adult stem cells. Science 2002;298:597¨C600.

Ivanova NB, Dimos JT, Schaniel C et al. A stem cell molecular signature. Science 2002;298:601¨C604.

Ustrell V, Hoffman L, Pratt G et al. PA200, a nuclear proteasome activator involved in DNA repair. EMBO J 2002;21:3516¨C3525.

Krogan NJ, Lam MHY, Fillingham J et al. Proteasome involvement in the repair of DNA double-strand breaks. Mol Cell 2004;16:1027¨C1034.
5 Y0 k, Y; h. Z# \. q/ H  ^
Rechsteiner M, Hill CP. Mobilizing the proteolytic machine: Cell biological roles of proteasome activators and inhibitors. Trends Cell Biol 2005;15:27¨C33.0 r* ]' @% ~. ?" t

Okuda Y, Nishi R, Ng JM et al. Relative levels of the two mammalian Rad23 homologs determine composition and stability of the xeroderma pigmentosum group C protein complex. DNA Repair (Amst) 2004;3:1285¨C1295.: ~5 C. C' y: V6 A
% Z, `: v8 J; `* j( e3 @
Strom L, Lindroos HB, Shirahige K et al. Postreplicative recruitment of cohesin to double-strand breaks is required for DNA repair. Mol Cell 2004;16:1003¨C1015.( G; Y& }- O- B. {& Y. ?! [. `

Jefford CE, Feki A, Harb J et al. Nuclear-cytoplasmic translocation of BARD1 is linked to its apoptotic activity. Oncogene 2004;23:3509¨C3520.
, N( x8 y: @8 w+ G4 s  r5 A, N+ H: w
Gill G. SUMO and ubiquitin in the nucleus: Different functions, similar mechanisms? Genes Dev 2004;18:2046¨C2059.
: s# @2 m4 H4 l
McGowan CH, Russell P. The DNA damage response: Sensing and signaling. Curr Opin Cell Biol 2004;16:629¨C633.

Debiak M, Nikolova T, Kaina B. Loss of ATM sensitizes against O6-methylguanine triggered apoptosis, SCEs and chromosomal aberrations. DNA Repair (Amst) 2004;3:359¨C368.
- n# x( M; S& k5 W* p/ V. v8 Z
Luo Y, Lin FT, Lin WC. ATM-mediated stabilization of hMutL DNA mismatch repair proteins augments p53 activation during DNA damage. Mol Cell Biol 2004;24:6430¨C6444.1 V* f% s9 C2 ?- Y
* C* {+ s9 r  I4 N% I5 d
Ito K, Hirao A, Arai F et al. Regulation of oxidative stress by ATM is required for self-renewal of haematopoietic stem cells. Nature 2004;431:997¨C1002." V4 B" D6 \& f( m. E( @- F

Raasi S, Pickart CM. Rad23 ubiquitin-associated domains (UBA) inhibit 26 S proteasome-catalyzed proteolysis by sequestering lysine 48-linked polyubiquitin chains. J Biol Chem 2003;278:8951¨C8959.
( n* S' Q; t" A  ~( a
Ortolan TG, Chen L, Tongaonkar P et al. Rad23 stabilizes Rad4 from degradation by the Ub/proteasome pathway. Nucleic Acids Res 2004; 32:6490¨C6500.% J' A6 C3 b0 c7 ?

Ng JMY, Vermeulen W, van der Horst GTJ et al. A novel regulation mechanism of DNA repair by damage-induced and RAD23-dependent stabilization of xeroderma pigmentosum group C protein. Genes Dev 2003;17:1630¨C1645.

Strasser A, Bouillet P. The control of apoptosis in lymphocyte selection. Immunol Rev 2003;193:82¨C92." n8 b2 r) r  W' C! _
4 N! q) V! n, R! ?
Van Sloun PP, Jansen JG, Weeda G et al. The role of nucleotide excision repair in protecting embryonic stem cells from genotoxic effects of UV-induced DNA damage. Nucleic Acids Res 1999;27:3276¨C3282." q' z$ z! e, C; a; S4 z2 s5 ~
" N5 L; B$ S; i7 E# y
de Waard H, de Wit J, Gorgels TG et al. Cell type-specific hypersensitivity to oxidative damage in CSB and XPA mice. DNA Repair (Amst) 2003;2:13¨C25.

收藏0 分享0 顶0 踩0