Muto, M., Fujimori, A., Nenoi, M., Daino, K., Matsuda, Y., Kuroiwa, A., Kubo, E., Kanari, Y., Utsuno, M., Tsuji, H., Ukai, H., Mita, K., Takahagi, M. and Tatsumi, K. Isolation and Characterization of a Novel Human Radiosusceptibility Gene, NP95. Radiat. Res. 166, 723–733 (2006).
The murine nuclear protein Np95 has been shown to underlie resistance to ionizing radiation and other DNA insults or replication arrests in embryonic stem (ES) cells. Using the databases for expressed sequenced tags and a two-step PCR procedure, we isolated human NP95, the full-length human homologue of the murine Np95 cDNA, which consists of 4,327 bp with a single open reading frame (ORF) encoding a polypeptide of 793 amino acids and 73.3% homology to Np95. The ORF of human NP95 cDNA is identical to the UHRF1 (ubiquitin-like protein containing PHD and RING domain 1). The NP95 gene, assigned to 19p13.3, consists of 18 exons, spanning 60 kb. Several stable transformants from HEK293 and WI-38 cells that had been transfected with the antisense NP95 cDNA were, like the murine Np95-knockout ES cells, more sensitive to X rays, UV light and hydroxyurea than the corresponding parental cells. In HEK293 cells, the lack of NP95 did not affect the activities of topoisomerase IIα, whose expression had been demonstrated to be regulated by the inverted CCAAT box binding protein of 90 kDa (ICBP90) that closely resembles NP95 in amino acid sequence and in cDNA but differs greatly in genomic organization. These findings collectively indicate that the human NP95 gene is the functional orthologue of the murine Np95 gene.
INTRODUCTION
Np95 was identified as a single 95-kDa mouse nuclear protein that appeared only in the S phase of normal mouse thymocytes but that was aberrantly accumulated in all stages of the cell cycle in mouse thymic lymphoma cells (1). This protein was localized as dot-like foci in S-phase nuclei, and almost all of these foci were colocalized with chromatin-bound PCNA in early S phase, although the degree of overlapping became nonexclusive in mid S phase and late S phase (2, 3). The expression of Np95 is apparently essential for entry from G1/G0 phase into S phase in adenovirus E1A-induced dedifferentiation of mouse myotube cells (4) but not in embryonic stem (ES) cells (5).
The Np95-encoding cDNA was cloned from a λgt11 expression library using Th-10a mAb and was verified as a novel nuclear protein with an open reading frame consisting of 782 amino acids (6). Np95 contains an N-terminal ubiquitin-like domain, a leucine zipper motif, a central zinc finger motif of the PHD finger type, and a C-terminal zinc finger of the RING finger type. It also possesses consensus sequences for an ATP/GTP binding site, a cyclin A/E-Cdk2 phosphorylation site, two retinoblastoma (Rb) protein-binding sites in the PHD finger, and the ring finger domains (6). The numbers of structural and functional domains existing in NP95 suggest multiple functions of this protein through protein-protein interactions and/or DNA bindings as a transcription factor for some crucial components in DNA metabolism, i.e. repair, replication and recombination. In fact, NP95 has recently been shown to interact with histones in vitro and to function as E3 ubiquitin ligase for histones, preferentially H3 (7).
Homozygously Np95-inactivated ES cells (Np95-null cells) are more sensitive to X rays, UV light, N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) and hydroxyurea (HU) than wild-type (Np95+/+) or heterozygously inactivated (Np95+/−) ES cells (5). The expression of the transfected Np95 cDNA in the Np95-null cells restored the resistance to these DNA-damaging agents and DNA replication blockers to a level similar to that of Np95+/− cells. Moreover, the frequency of spontaneous sister chromatid exchange was significantly higher in Np95-null cells than in Np95+/+ or Np95+/− cells (5). In view of these results, NP95 has been considered to function at least in ES cells as a common component in multiple DNA damage response pathways and to play a role in the maintenance of genomic stability (5), although the precise mechanism(s) remains largely to be defined. While it has been reported that DNA metabolism and cell cycle progression are controlled differently in ES cells than in somatic cells (8–13), no role of NP95 in maintaining genomic integrity has been suggested as yet in any kind of mouse somatic cells.
Whether NP95 and its human homologue serve similar functions in somatic cells from mice and those from humans, respectively, might be another question. The lack of the cross-reactivity of Th-10a with human cells (1) hampered our earlier characterization of human protein(s) corresponding to mouse NP95. Nonetheless, we had been successfully pursuing an attempt to isolate the full-length human homologue of the murine Np95 cDNA, NP95, by using the databases for expressed sequenced tags (ESTs) and the two-step PCR procedure. The exons of the human NP95 gene were partially sequenced and deposited as AB075601 in 2001. In this study we present the full-length cDNA sequence, genomic structure and chromosomal mapping of this gene. Furthermore, we generated several stable transfectants with human NP95 cDNA in both antisense and sense orientations and observed the effect of suppressed transcription of the NP95 gene on cellular resistance to DNA damages and DNA replication blocks including X rays in human somatic cells.
Human topoisomerase IIα functions in growth-dependent processes such as DNA replication and chromosome segregation (14, 15) mainly by alleviating torsional stress ahead of replication forks (16). In 2000, Hopfner et al. used the one-hybrid system with an inverted CCAAT box as the DNA target sequence in screening human cDNA libraries and discovered ICBP90, an inverted CCAAT box binding protein of 90 kDa, as being involved in a transcriptional regulation of the topoisomerase IIα gene expression in human cells (17–19). ICBP90 has similar domain structures with a 73.7% identity of amino acid sequences to those of mouse NP95 and has thus been considered as a candidate for a potential human counterpart of mouse NP95 (19). Inhibitors of topoisomerase IIα including m-AMSA and etoposide have been shown to sensitize cultured cells to ionizing radiation (20, 21), and so reduced expression of topoisomerase IIα could cause an increased sensitivity to ionizing radiation. We therefore examined whether knocking down of the NP95 gene in fact suppressed topoisomerase IIα gene expression together with the enzyme activity. Contrary to our expectation, the reduced expression of NP95 did not affect the topoisomerase IIα activity in human cells. The purified human NP95 protein indeed interacted with the critical ICB2 (inverted CCAAT box 2) element oligo-DNA, one of five CCAAT boxes in the inverted position of the topoisomerase IIα, but in fact lacked the binding specificity to the CCAAT box.
Thus the combination of genomic structure of the human NP95 being similar to that of the murine Np95, the resistance conferred by NP95-like murine NP95 against DNA insults and replication blocks, and a functional dissimilarity of human NP95 from ICBP90 in terms of the effect on topoisomerase IIα led us to propose that NP95 is the human functional orthologue of murine Np95.
MATERIALS AND METHODS
EST Database Screening and PCR-Based Full-Length cDNA Cloning
We searched for human homologues of the murine Np95 gene by scanning a human EST database using tblastn (22, 23) of the Basic Local Alignment Search Tool (BLAST) network service at the National Center for Biotechnology Information (NCBI). We used sequence information from the resulting matches, AA811055, AA903902, AA306523, AA354253 and AA488755, to synthesize several primers and used them in two-step PCR to amplify the core cDNA, cDNA 5 end and cDNA 3′ end from various human cDNA libraries (the detailed protocol of two-step PCR and sequences of primers are available on request).
Isolation of Human Bacterial Artificial Chromosome (BAC) Clones
To isolate BAC clones containing the human NP95 gene, PCR screening of the RPCI-11 library (24) was performed using the primers F2 (TGATCAGAAACT CAC CAA CACC) and R3 (CTTGCCACCCTTGAC ATTG) at a commercial facility (Nippon Technoservice Co, Chiba). The clones identified from the PCR screening were RPCI-11-165N13, RPCI-11-143D24 and RPCI-11-160C21.
Chromosome Preparation and Fluorescence In Situ Hybridization (FISH)
The FISH method was used for chromosome assignment of the human NP95 gene. Preparation of R-banded chromosomes and FISH were performed as described by Takahashi et al. (25, 26). The NP95 cDNA (2.4 kb) in the PCR2.1-TOPO vector (E43 clone) or BAC clones were labeled by nick translation with biotin-labeled 16-dUTP (Roche Diagnostics, Tokyo). The hybridized probes were allowed to react with goat anti-biotin antibodies (Vector Laboratories, Burlingame, CA) and then stained with fluorescent anti-goat donkey IgG (Nordic Immunology). The hybridization signals were visualized with Nikon filter sets B-2A and UV-2A. Kodak Ektachrome ASA 100 film was used for microphotography.
Generation of Expression Constructs Containing Human NP95 cDNA in both Antisense and Sense Orientations
pCR2.1.E43.hNP95 plasmid was made by subcloning the full-length human NP95 cDNA into the pCR2.1-TOPO TA cloning vector (Invitrogen). To insert NP95 cDNA into the mammalian expression vector in the antisense orientation, the DNA fragment was obtained by PCR using pCR2.1.E43.hNP95 as a template according to the manufacturer's protocol. This PCR product was inserted into the entry vector pDONR201 (Gateway system, Invitrogen) using BP clonase to create pDONR.AS.hNP95. The pDEST26.AS.hNP95 expression vector coding for NP95-antisense cDNA was created using the Gateway LR reaction with pDONR.AS.hNP95 and pDEST26 (Invitrogen). To construct the NP95 cDNA expression vector in the sense orientation, the reverse PCR product was converted to an entry clone by performing a BP reaction, and the pDEST26.hNP95 expression vector coding for NP95-sense cDNA was created using the Gateway LR reaction.
Cell Culture
HEK293 (JCRB9068) and WI-38VA13 (JCRB9057) cells were grown in DMEM/F12 medium (Sigma-Aldrich, St. Louis, MO) containing 10% FBS. The cells were obtained from the Japanese Cancer Research Resources Bank (JCRB).
In Vitro Treatment
X irradiation
Cells were trypsinized and plated on 100-mm dishes at a density of 1.0 × 103 cells per dish. After incubation at 37°C for 16 h in DMEM/F12 medium supplemented with 10% FBS, cells were exposed to various doses of X rays from a Shinai-III X-ray generator (Shimadzu Seisakusho Ltd, Kyoto) at 0.76 Gy/min (200 kVp, 20 mA, with filters of 0.5 mm copper and 0.5 mm aluminum).
UV irradiation
Cells were trypsinized, plated on 100-mm dishes and cultured at 37°C for 16 h. Cells were washed twice and then overlaid with 5 ml of Dulbecco's PBS(-) (Nissui Pharm. Co. Ltd., Tokyo). The cells were irradiated with UV light emitted from germicidal lamps (15 W, GL-15, Toshiba, Tokyo) at room temperature. The fluence rate was 0.285 J/m2/s, as measured with a UV intensity meter (Topcon UVR-254, Tokyo Kogaku, Tokyo).
HU treatment
Cells were trypsinized and plated on 100-mm dishes at a density of 1.0 × 103 cells per dish. After incubation at 37°C for 16 h, the medium was removed, and various concentrations of HU (Sigma-Aldrich) were added to the cultures. After treatment at 37°C for 24 h, cells were washed with 15 ml of Dulbecco's modified Eagle's medium and cultured further in DMEM/F12 medium supplemented with 10% FBS.
Colony Formation
After the treatments, cells were incubated with DMEM/F12 medium supplemented with 10% FBS for 10 days, and then colonies were scored. Plating efficiencies ranged from 15 to 20%. Relative survivals in terms of the loss of clonogenicity were based on at least two or three repeated experiments, which were carried out in triplicate.
Transfection
The day before transfection, 5 × 105 HEK 293 or WI-38-VA13 cells per well were plated on six-well plates and cultured in DMEM/F12 medium containing 10% FBS. These cells were then transfected with pDEST26.AS.hNP95 or pDEST26.hNP95 plasmid DNA using Lipofectamine 2000 according to instructions provided by the manufacturer (Invitrogen). After incubation at 37°C for 24 h, the cells were trypsinized, seeded in 100-mm dishes, and cultured further for 10 days in DMEM/ F12 medium with 10% FBS in the presence of Geneticin (300–500 μg/ ml, Invitrogen). Thirty colonies resistant to Geneticin were isolated in each group. Stable transfected cells were obtained 3 weeks after transfection.
Production of an Anti-human NP95 Polyclonal Antibody
An antibody against a unique protein sequence in human NP95 was created by the subcutaneous immunization of rabbits with a synthetic peptide (CFAPINDQE GAEAKDWRSGK) that was coupled to keyhole limpet hemocyanin. The antibody was affinity-purified on a column containing the peptide crosslinked to sepharose.
Western Blot Analysis
Three million cells were seeded in 100-mm dishes, and after 12 h they were washed twice with PBS (++) and immediately lysed in lysis buffer (2) plus SDS sample buffer (New England BioLabs, Beverly, MA). Equal amounts of whole-cell lysates were separated by 4–20% gradient SDS-PAGE (DAIICHI PURE Chem. Co. LTD, Tokyo) and transferred to Immobilon-P (PVDF) membranes (Millipore, Bedford, MA). Membranes blocked in 5% skim milk (DIFCO Laboratories, Detroit, MI) in PBS for 1 h were rinsed and incubated with the anti-NP95 antibody in 2% (v/v) skim milk in PBS for 1 h at room temperature and washed three times with PBS-T (0.1% Tween in PBS). The bound primary antibody was allowed to react with horseradish peroxidase-coupled anti-rabbit IgG antibody (P0448, DAKO, Kyoto) diluted at 1:5000 in 2% skim milk in PBS for 1 h at room temperature and washed three times with PBS-T. Signals were detected by enhanced chemiluminescence (Lumi-Light Plus, Boehringer Mannheim, Germany).
Preparation of Nuclear Extract and Topoisomerase II Assay
Five million cells were collected by trypsinization, washed once in phosphate-buffered saline, and resuspended in 200 μl NE buffer [10 mM Hepes, pH 7.9, 10 mM KCl, 0.75 mM spermidine, 0.15 mM spermine, 0.2 mM EDTA, 0.2mM EGTA, 0.5 mM dithiothreitol, and 1 mM phenylmethanesulfonyl fluoride (PMSF)] and incubated for 15 min on ice. Cells were then lysed by passage 10 times through a 26-gauge needle, and the lysate was centrifuged for 1 min in a microcentrifuge at 10,000 rpm. The resulting nuclear pellet was resuspended in 100 μl of NESG buffer [20 mM Hepes, pH 7.9, 0.4 M NaCl, 25% (v/v) glycerol, 0.75 mM spermidine, 0.15 mM spermine, 0.2 mM EGTA, 0.5 mM DTT, 1 mM PMSF] and incubated for 30 min on ice with frequent mixing. The extract was centrifuged for 20 min at 15,000 rpm at 4°C. The resulting supernatant (nuclear extract) was stored at −85°C, and the protein concentration was determined with a protein assay kit (Bio-Rad, Hercules, CA).
Topoisomerase II activity was assayed by the ATP-dependent decatenation of kinetoplast DNA (kDNA) (#1001, TopoGEN, Inc., Port Orange, FL). Human type II topoisomerase (p170 form, #2000H-1, TopoGEN) was used as a positive control. Decantenation assays were carried out using kDNA substrate in a final volume of 20 μl in topo II reaction buffer (50 mM Tris-HCl, pH 8.0, 120 mM KCl, 10 mM MgCl2, 0.5 mM ATP, 0.5 mM DTT, 30 μg/ml nuclease-free BSA) with 0.2 μg kDNA. Reaction products were analyzed on a 1% agarose gel containing 0.5 μg/ml ethidium bromide.
Purification of Recombinant Human NP95 Protein and Electrophoretic Mobility Shift Assay (EMSA)
To examine whether human NP95 binds with the critical ICB2, one of five CCAAT boxes in the inverted position of the human topoisomerase IIα promoter region (27), sense and antisense oligonucleotides; 5′ ATAAAAGGCAAGCTACGATTGGTTCTTCTGGACGGAGAC 3′ and 5′ GTCTCCGTCCAGAAGAACCAATCGTAGCTTGCCTTTTAT 3′, were synthesized and annealed. Mutated ICB2 (ICB2m) oligonucleotides, 5′ ATAAAAGGCAAGCTA CGATTCCTTCTTCTGGACGGAGAC 3 and 5′ GTCTCCGTCCAGAAG AAGGAATCGTAGCTTGCCTTTTAT 3′, were also synthesized, annealed and then subjected to the test for binding specificity. These double-stranded oligo DNAs were 32P-end labeled using T4 polynucleotide kinase (Takara, Tokyo) and (γ-32P)ATP (148 TBq/mmol, ICN Biomedicals).
For human NP95 purification, HEK293 cells were transfected with pCMV.Tag4A.hNP95 construct by using Lipofectamine Plus reagent (Invitrogen), and the cells were harvested 48 h after transfection. The nuclear extract was incubated with anti-Flag M2-agarose (H2220, Sigma) beads for 1 h at 4°C followed by washing with 50 mM Tris-HCl (pH 8.0) and 0.15 M NaCl. Proteins were eluted in 0.1 M glycine-HCl (pH 3.5). The purity of the NP95 preparation was monitored with SDS-PAGE and Coomassie blue staining (>95% in purity) (Fig. 5B). EMSA was performed as described previously (28). In brief, the purified NP95 (330 ng) was mixed with 1.4 ng of 32P end-labeled oligonucleotides in 12% glycerol, 20 mM Hepes (pH 7.9), 70 mM KCl, 5 mM MgCl2, 0.05% NP40, 1 mg/ml BSA, 0.5 mM DTT and 0.1 mg/ml poly(dI-dC) in 20 μl and incubated at 25°C for 20 min. For competition experiments, the indicated amounts of unlabeled oligonucleotides were first mixed with NP95 and incubated at 4°C for 25 min. The 32P end-labeled oligonucleotides were added to the reaction mixtures, which were subsequently incubated at 25°C for 20 min. To verify the binding of the labeled oligonucleotides with NP95, affinity-purified anti-NP95 antibody (14 μg) was added to the above-mentioned reaction mixture and incubated at 25°C for 20 min. These reaction mixtures were loaded on 6% polyacrylamide gels and subjected to electrophoresis.
RESULTS
Cloning of Human NP95 cDNA and Chromosomal Localization
Using two-step PCR, we identified human NP95 cDNA, which is a 4327-bp cDNA from the human testis cDNA library, and a 3229-bp cDNA from the human fetal thymus cDNA library (Accession Nos. AB177623, AB177624). These contain a single open reading frame (ORF) encoding a polypeptide of 793 amino acids and shows 73.3% homology to mouse NP95 at the amino acid level. The chromosomal location of the human NP95 gene was determined by direct R-banding FISH. NP95 was localized to chromosome 19p13.3 in a region with conserved synteny to the mouse chromosome 17 D-E1.1 on which the mouse Np95 gene was mapped (Fig. 1A). We searched the GenBank database using the human NP95 cDNA sequence as a probe and found two matches, accession numbers AC027319 and AC053467, which are localized to chromosome 19p13.3. These two DNA sequences are separated by a gap of approximately 5 kb genomic DNA. We cloned and sequenced the gap (exons 15–17 of human NP95) and submitted it to the database under accession number AB075601. Further analysis showed that NP95 includes 18 exons spanning 60 kb of genomic DNA (data not shown). After determining the sequence of the DNA fragment, we noted that the complete NP95 gene sequence had already been deposited by NCBI under the accession number NT_011255. The genomic structure of the human NP95 is similar to the mouse Np95 gene, which consists of 17 exons spanning 23 kb (5).
In the process of cloning the human NP95 gene from the RPCI-11 human male BAC library, we identified three clones, 165N13, 143D24 and 160C21, using a 138-bp cDNA probe (nucleotides 2068–2205 of the NP95 cDNA, Accession No. AB177623). Mapping by FISH, however, assigned those BAC clones to chromosome 12p12.2-p12.1 (Fig. 1B). We found that they all contained a DNA sequence 2.8 kb in length that was similar to a part of the human NP95 cDNA (nucleotides 1518–4327 of the 4327-bp human NP95 cDNA, Accession No. AB177623) and interestingly included polyadenylation signals followed by a poly-A tail at the 3′ end. This 2.8-kb sequence was termed the del hNP95 by us and was confirmed to exist in AC112777 of GenBank database, which had been assigned at 12p12.2-p12.1. Thus this sequence might have arisen from the integration into the chromosome of a naturally occurring cDNA sequence that had been generated as an RNA transcript by reverse transcriptase. In the mouse genome, a similar 2.7-kb sequence (an exonic part from exon 5 to exon 17 of the Np95 cDNA) has also been recognized on chromosome XD (Fig. 1C). The deduced amino acid sequence of human NP95 was identical to that of UHRF1 (Accession No. AF274048). The NP95 protein contains a ubiquitin-like domain, a leucine zipper motif, a zinc finger motif, a putative cyclin A/E-CDK2 phosphorylation site, an RB-binding motif and a ring finger domain (Fig. 1D).
The 5′ UTRs and 3′ UTRs of Human NP95 Transcripts
The 5′- and 3′-terminal regions of human NP95 cDNA were analyzed by RACE as described in the Materials and Methods. RT-PCR analysis was subsequently performed to further characterize the 5′ UTRs of these transcripts in various tissues. The sequences of the two cDNA clones, E43.hNP95 and Q15.hNP95, as well as that of RACE and RT-PCR products obtained from different tissues, were matched against their corresponding genomic sequences (accession no. AC027319, AC053467, AB075601, NT_011255). The results indicated that there are four noncoding exons (exons 1, 2, 3a and 3b) that undergo alternative splicing, forming divergent 5′ UTRs (Fig. 2A and B). 3′-RACE products included two polyadenylation signals (AATAAA) at 1152 bp and 1328 bp downstream from a TGA translation stop codon and contained a poly-A tail (A6-A15), indicating that the 3′ ends of the cDNA sequences are complete and that the polyadenylation signal is functional (AB177623, AB177624). To analyze the tissue specificity of different 5′ UTRs, we performed two-step PCR using panels of normalized, first-strand cDNA preparations from human fetal tissues (K1425-1, Clontech) and human adult tissues (K1420-1, K1421-1, Clontech). When sense primers derived from the 5′ UTR of exon 3b and antisense primers located in the 3′ UTR of exon 18 were used, a full-length NP95 cDNA was detected in the brain, lung, liver, kidney, thymus and skeletal muscle of the human fetus (Fig. 2C-1) and also in adult brain, testis and thymus (Fig. 2C-2). When sense primers from the 5′ UTR of exon 1 and antisense primers from the 3′ UTR of exon 18 were used, NP95 cDNA was detected in the testis (data not shown).
Increased Sensitivities to DNA Insults and Replication Arrests after Human NP95-Antisense Expression in Human Cells
We generated stable transformants from HEK293 and WI-38 cells that had been transfected with the human NP95 cDNAs in both antisense and sense orientations. Western blot analysis verified that the amount of NP95 protein increased dramatically in the NP95-sense transfected S22C cells and markedly decreased in the NP95-antisense transfected cells, 6AC (Fig. 3A), W28A, W1A and W30A (data not shown). Diminished expression of NP95 resulted in a growth retardation; the population doubling times for the parental HEK293 cells, HEK.V5 cells, 6AC cells, and S22C cells were approximately 24, 24, 36 and 24 h, respectively. We examined whether the altered level of NP95 expression affects the sensitivity to various types of DNA damage using NP95-antisense cDNA transfected cells. The NP95-antisense transfectants, 6AC, W28AS, W1AS and W30AS, were more sensitive to X rays than their parental cells (HEK293, WI-38) or the NP95-sense transfectant, S22C cells, and the empty vector transfectant, HEK.V5 cells (Fig. 3B and C). 6AC cells were also more sensitive to UV light than S22C cells and the parental HEK293 cells (Fig. 4A). Moreover, these NP95-antisense transfectants were more susceptible to DNA replication blocks after treatment with hydroxyurea than S22C, the NP95-sense cDNA transformants, or the parental WI-38 and HEK293 cells (Fig. 4B). Thus the expression of NP95 in human somatic cells appears to confer the ability to cope with DNA damage and DNA replication blocks, as is also the case with NP95 in mouse ES cells (5).
No Dependence of Topoisomerase IIα Activity on Human NP95 Expression
The topoisomerase IIα activities of the nuclear proteins extracted from HEK293, 6AC and S22C cells were compared using an assay of the ATP-dependent decatenation of kinetoplast DNA (kDNA). The topoisomerase II activities in these cells were equivalent to approximately 1 unit of human type II topoisomerase activity within the resolution of experimental condition (p170 form, #2000H-1, TopoGEN, Inc), and they did not correlate with the degree of NP95 expression (Fig. 5A-1 and 5A-2)
Nonspecific Binding of Human NP95 to the ICB2 Element of Topoisomerase IIα Gene Promoter
We next tested whether NP95 binds specifically to the ICB2 element of the promoter region in the topoisomerase IIα gene. Flag-tagged NP95 was expressed in HEK293 cells and purified to near homogeneity using anti-Flag M2-agarose beads (Fig. 5B). ICB2 oligo-DNA that contains CCAAT box or C→G mutated ICB2, i.e. ICB2m oligo-DNA that contains a GGAAT box, was 32P end-labeled as described by Hopfner et al. (19). The supershift of labeled ICB2 probe by anti-NFY antibody was used as a positive control for specific DNA binding, as shown in Fig. 5C. In Fig. 5D, the purified NP95 was mixed with 32P-end labeled oligo-DNAs and the reaction mixture was loaded on 6% polyacrylamide gels. The intensity of the bands representing oligonucleotides bound to NP95 (denoted as B) indicated that NP95 interacted with both ICB2 (lane 3) and ICB2m (lane 10) to a virtually similar extent. In EMSA, the addition of affinity purified anti-NP95 antibody to a reaction mixture resulted in the appearance of super shift bands (denoted as SB) and the concomitant disappearance of B bands, and there was no difference in the intensity of SB bands between ICB2 and ICB2m (lanes 2 and 9). When NP95 was assumed to bind specifically to ICB2, the intensity of B bands in lanes 3–7 should have been much denser than those representing interaction between NP95 and ICB2m in lanes 10–14. These results indicate that NP95 binds to ICB2 but not in a specific manner.
DISCUSSION
We have identified and characterized the NP95 gene, which is the human homologue of the murine Np95 gene. The NP95 gene at 19p13.3 has 18 exons spread over 60 kb of genomic DNA, and its exon-intron structure is similar to that of mouse Np95 (5). We characterized the 5′ and 3′ UTRs of the NP95 transcripts in a few human cDNA libraries from fetal thymus and testis. Several alternative splicings in the 5′ UTR and differential polyadenylation in the 3′ UTR were found in the NP95 transcripts, although the ORF remained unaffected. These divergent mRNA leaders may play a role in the translational regulation of NP95 (Fig. 2B and C), as is the case for the differential ATM expression among different tissues (29).
Since we had noted the hypersensitivity of Np95-knockout ES cells to X rays and UV light, MNNG and hydroxyurea compared with ES wild-type (Np95+/+) or heterozygously inactivated (Np95+/−) cells (5), we tested whether the same holds true for human NP95 expression in HEK293 and WI-38 cells. Under the experimental conditions used, it was found that the expression of the full-length antisense NP95 cDNA led to an essentially total deprivation of NP95 protein (Fig. 3A). In accordance with our expectation, the NP95-antisense cDNA transfectant clones, 6AC, W28A, W1A and W30A, were all more sensitive to X rays than their untransfected parental cells (HEK293, WI-38), empty vector-transfected cells (HEK-V5), or the NP95-sense cDNA transfectant S22C cells (Fig. 3B). 6AC cells were also more sensitive to UV light than S22C or HEK293 cells (Fig. 4A). Moreover, these NP95-antisense transfectants were more susceptible to DNA replication blocks induced by hydroxyurea than S22C cells or untransfected parental HEK293 and WI-38 cells (Fig. 4B). Thus, just like our findings with Np95 in mouse ES cells (5), the expression of NP95 in human somatic cells appears to confer the cellular capability to cope with both DNA insults and DNA replication arrests that could cause stalled replication forks in S-phase cells, a very vulnerable structure unless properly restored (30).
ICBP90, an inverted CCAAT box binding protein, has been demonstrated by Hopfner and his coworkers to be involved in the regulation of topoisomerase IIα transcription (18, 19). Since several inhibitors of human topoisomerase IIα have been reported to enhance cell killing by ionizing radiation (20, 21), the hypersensitivity to X rays of Np95 knockout mouse ES cells (5) and of NP95 knockdown human cells in this study may be attributable to a decreased expression of topoisomerase IIα. However, the specific binding of NP95 to the CCAAT box was not assured in this study, since purified NP95 bound not only to ICB2 but also to the mutated ICB2 to a similar extent (Fig. 5D). Moreover, our results indicated that the topoisomerase IIα activity in two different types of human cells employed were not correlated with the level of the expression of NP95 (Fig. 5A). Recently, the binding of this protein with methyl-CpG dinucleotides, especially methylated promoter region of some tumor suppressor genes, has been reported (31). No detectable amount of NP95 in the crude extracts from thymus tissue of C57BL/10 mice was demonstrated to bind to any of the seven CCAAT boxes in the upstream region of the mouse topoisomerase IIα gene (32), whereas NF-Y, the ubiquitous transcription factor also known as CBF, ACF and CP1 (27), was unequivocally bound to three or four CCAAT boxes as a major binding factor (M. Nenoi et al., data not shown). Thus NP95 and ICBP90 do not appear to share a common function in terms of the effect on topoisomerase IIα activity.
Hopfner et al. have reported that the ICBP90 gene spans approximately 35.8 kb on chromosome 19p13.3 and contains 6 coding exons named A to F (18). We have noted that NP95 and UHRF1 are identical in ORF and that ICBP90 and NP95 differ in ORF by only two amino acids (Lys383→Asn383, Ala457→Ser457) and seven nucleotides. However, we have been unable to recognize the sequence of exon F (2.6 kb) of the ICBP90 gene in NT_011255 of GenBank database at 19p13.3 (18). On the other hand, this exon F of the ICBP90 gene is quite similar to the del hNP95 sequence of 2.8 kb in our BAC clones (see AC112777). Exons through A to E of ICBP90 do exist at 19p13.3 (see NT_011255), not at 12p12.2-p12.1 (see AC129029 and AC112777). The gene encoding ICBP90 therefore appears to be distinct from the one encoding NP95.
NIRF (Np95/ICBP90-like ring finger protein)/human NP97/UHRF2 protein also resembles NP95 in possessing a ubiquitin-like domain, a PHD finger, an SRA (G9a) domain, an RB-binding motif, and a ring finger. Its gene is mapped at chromosome 9p24.1 (33). Since NP95/UHRF1 protein and NIRF/human NP97/UHRF2 protein have a similar organization, their corresponding genes found on mouse chromosomes (Fig. 1C) probably derive from a common ancestral gene. If we compare their coding region amino acid sequences, NP95 is closer to NP95 (73.3% similarity) than to NIRF/human NP97/UHRF2 (51.3%), implying that the NP95 mapped to 19p13.3 is the human orthologue of the mouse Np95.
In summary, judging from the similarity in the exon structure, the identity of deduced amino acid sequence, the hypersensitivity to X rays and other DNA insults and replication arrests in the NP95 knockdown cells, and the apparent lack of relevance to topoisomerase IIα gene expression, we propose that NP95 is the human functional orthologue of the murine Np95.
Acknowledgments
We thank Dr. T. Saito for discussion during this work and Dr. J. Rodrigue and Prof. B. Burke-Gaffney for editorial assistance. The technical assistance by Mr. and Ms. S. Sasanuma and secretarial assistance by Ms. Y. Kikuchi-Takahagi are deeply appreciated. This work was supported in part by a Special Grant for the Project Research “Genetic Control of Biodefence Mechanisms against Radiation” to MM and KT and by a Grant-in-Aid for Scientific Research to KT and AF (Grant 15310044 “Mechanism for Protection of Human Somatic Cells against Ionizing Radiation by Novel Nucleoprotein hNP95”) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
REFERENCES
FIG. 1.
Chromosomal localization of human NP95 (hNP95) and del hNP95 and structure of NP95. Panel A: The FISH was used for chromosomal assignment of the NP95 gene. A 2.4-kb NP95 cDNA fragment was used as a biotinylated probe. Arrows indicate the hybridization signals. The NP95 gene was localized to Chr 19p13.3. The metaphase spreads were photographed with Nikon B-2A (a, c) and UV-2A (b) filters. R- and G-banded patterns are shown in (a, c) and (b), respectively. Panel B: BAC clone 160C21 was used as a biotinylated probe. Arrows indicate the hybridization signals. 160C21 clone (del hNP95) was localized to Chr 12p12.1–12.2. The metaphase spreads were photographed with Nikon B-2A (a, c) and UV-2A (b, d) filters. R- and G-banded patterns are shown in (a, c) and (b, d), respectively. Panel C: Distribution of the Np95 gene family and the processed pseudogene on chromosomes. The figures in boxes are the percentages of amino acid sequence similarity between the ORFs of the corresponding pairs of genes. Panel D: Structure of NP95 and del hNP95
FIG. 2.
Nucleotide sequence and alternative splicing in the 5′ region of the human NP95 (hNP95) gene. Panel A: Nucleotide sequence of the genomic region spanning the 5′ end of the NP95 gene. Exons (in capitals) are boxed. The ATG initiation codon of the ORF is underlined in bold. Panel B: Alternative splicing of the 5′ UTR of the NP95 transcripts. Exons and introns are marked by boxes and solid lines, respectively. Panel C: Expression of NP95 mRNA in human fetal tissue. The full-length NP95 cDNA was detected by two-step PCR using primers designed on the basis of sequence information of the 5′ UTR in exon 3b and the 3′ UTR in exon 18. The first PCR was performed using primers U221 (AGGTGCTGGTAAAACTGATGG) and L3064 (AGGGAACGGAGGAAACACTAC), and human fetal tissue cDNA (K1425-1, Clontech, Palo Alto, CA) as a template. The second run used the first PCR product as a template and U36 (CTGATGGGGGTTTTTGCTGTC) and L3030 (GCCCACTAAGAACAAAATCAG) as primers
FIG. 3.
Human NP95 confers resistance against X radiation. Panel A: (1) Expression of NP95 protein in HEK293 cells and the NP95-antisense and -sense cDNA stably integrated HEK293 cells. Total cell extracts (6 μl out of 300 μl lysate prepared from 1 × 107 cells) were immunoblotted with anti-NP95 antibody. Lane 1, untransfected HEK293 cells; lane 2, 6AC (the NP95-antisense cDNA transfected HEK293 cells); lane 3, S22C (the NP95-sense cDNA transfected HEK293 cells). (2) The same amounts of lysates from above cell lines were immunoblotted with anti-α tubulin (DM1A, Sigma-Aldrich) and anti-β actin (AC-15, Sigma-Aldrich) as loading controls. Panel B: Right frame: Relative surviving fractions as a function of dose of X radiation in the NP95-antisense stably expressed HEK 293 cells, 6AC (○), the NP95-sense cDNA expressed HEK 293 cells, S22C (•), untransfected HEK 293 cells (▪), and the empty vector transfected HEK293 cells, HEKV5 (□). Left frame: Relative surviving fractions as a function of X-ray dose in the NP95-antisense cDNA expressed WI-38 cells, W30A (▪), W1A (•) and W28A (▴), and untransfected WI-38 cells (○). Each data point denotes a mean from the experiment carried out in triplicate
FIG. 4.
Human NP95 confers resistance against UV radiation and against treatment with HU. Panel A: Clonogenic survival curves after UV irradiation. Relative surviving fractions as a function of UV-radiation dose were determined for the NP95-antisense cDNA expressed HEK293 cells, 6AC (○), the NP95-sense expressed cDNA HEK 293 cells, S22C (▪), and untransfected HEK 293 cells (□). Data points denote means from the experiment performed in triplicate. Panel B: Increased sensitivity of the NP95-antisense cDNA stably integrated cells to continuous treatment with HU. Left: Relative surviving fractions in the NP95-antisense cDNA expressed HEK293 cells, 6AC (○), the NP95-sense cDNA expressed HEK293 cells, S22C (□), and HEK 293 cells (•) after treatment with HU for 24 h. Bars represent standard errors. Right: Relative surviving fractions in the NP95-antisense cDNA stably expressed WI-38 cells, W5A (♦), W28A (•) and W1A (▪) cells and untransfected WI-38 cells (○) after treatment with HU for 24 h. Data points denote means from a representative experiment carried out in triplicate
FIG. 5.
Irrelevance of human NP95 (hNP95) to expression of topoisomerase IIα. Panel A: Influence of the level of NP95 expression on activities of topoisomerase type II. (1) Representative data from the decatenation assay. The nuclear extracts (150 ng) from 6AC (▪), S22C (•), and HEK293 cells (♦) were prepared, and topoisomerase II activities were assayed by the ATP-dependent decantenation of kinetoplast DNA (kDNA). Human type II topoisomerase (p170 form) (TopoGEN, Inc.); 1 (○) and 2 units (□) were used as a positive control. The proportion of supercoiled kDNA remaining as a function of reaction time was determined. Panel B: (1) Coomassie staining of the purified Flag-hNP95 protein. (2) Immunoblot analysis with anti-Flag. Panel C: Electrophoretic mobility shift assays (NF-Y). Panel D: Electrophoretic mobility shift assays (hNP95). The purified human NP95 was reacted with 32P end-labeled ICB2 or ICB2m. The positions of bound, B, and free, F, probes are indicated. To verify the binding of the labeled oligonucleotides with NP95, affinity purified anti-NP95 antibody was added to the reaction mixture (lane 2 and lane 9), and the position of super shift probe bound, S.B., was exhibited
