ABSTRACT
Treatment of cells with DNA damaging agents leads to induction of a variety of genes involved in different cellular processes. We have applied a lacZ-based gene trap strategy to search for new mammalian genes induced by genotoxic stress. A population of 32 × 103 neor clones stably transfected with a gene trap vector was obtained, stained with fluorescein di-[beta]-d-galactopyranoside and analyzed by flow activated cell sorting and replica plating. This strategy allowed isolation of 30 neor `putative inducible' cell lines expressing lacZ only after a DNA damaging treatment. For three clones the site of integration and the degree of inducibility after UV treatment were determined by Southern blot and [beta]-galactosidase measurement respectively. One cell line (clone VI) showed a single integration site and a reproducible 3-fold induction of [beta]-galactosidase activity following UV irradiation. Fused transcripts were isolated from induced cells and a portion of the trapped gene was amplified by rapid amplification of cDNA ends. Sequence analysis and comparison with available gene and protein databanks revealed that the gene was novel.
Exposure of eukaryotic cells to DNA damaging agents induces numerous genes involved in many cellular processes, such as cell cycle control, signal transduction, DNA repair, apoptosis and other pathways associated with cell protection from injury (1 ). In Escherichia coli the existence of a regulatory pathway responding to DNA damage, the SOS system, has been well characterized (2 ). In mammalian cells a network of overlapping systems seems to be activated following exposure to DNA damaging agents. Many genes are induced specifically by UV (3 ) and [gamma]-rays (4 ), while others also respond to alkylating agents and to growth arrest (5 ). In many cases the cellular response to genotoxic treatment is triggered by signal transduction pathways which are not DNA damage specific, i.e. the UV response (3 ,6 ).
Several genes activated following treatment with DNA damaging agents are known proto-oncogenes or tumour suppressor genes (1 ). It is well documented that the DNA damage-inducible gene p53, which plays a key role in protection against genotoxic insult, is frequently mutated in human tumours (7 ). Thus identification of new genes participating in the DNA damage response can be particularly important in unravelling the carcinogenic process.
A gene trap strategy has been devised to isolate and clone trapped genes that display various expression patterns in mouse embryos at different developmental stages (8 -10 ). Moreover, this technique has recently been used to identify new genes involved in signal transduction pathways activated by retinoic acid in undifferentiated embryonic stem cells (11 ). This approach is based on transfection experiments with a gene trap vector containing a promoterless lacZ gene with a splice acceptor site. Expression of lacZ depends on its insertion within an active transcription unit; if insertion occurs in an inducible gene, one should be able to detect an increase in [beta]-galactosidase activity after an inducing treatment.
In this paper we present a new application of this approach which has been used to search for mammalian genes involved in the cellular response to genotoxic stress. We used a lacZ-based gene trap construct to isolate stably transfected Chinese hamster cell lines that showed inducibility of [beta]-galactosidase activity after a DNA damaging treatment.
A map of the gene trap vector pEN53 (Dr E.Neilan, PhD thesis, University of Stanford, CA) is outlined in Figure 3 , lower panel. The 3.8 kb BstEII-BamHI fragment contains the artificial adenovirus splice acceptor sequence fused to the lacZ reporter gene. The presence of the splice acceptor sequence upstream of lacZ allows expression of the reporter gene after insertion into an intron and, consequently, formation of spliced fusion transcripts with endogenous genes. The promoterless lacZ gene contains an artificial eukaryotic translation initiation sequence (12 ) and an SV40 bidirectional polyadenylation signal. The bacterial neomycin resistance gene (neo), encompassing the 1.2 kb BamHI-HindIII fragment, is driven by the TK promoter. The bacterial origin of replication (ori) and the ampicillin resistance gene (amp) are those of the vector Bluescript II KS- (Stratagene).
AA8 Chinese hamster ovary (CHO) cells were cultured under standard conditions as described (13 ). Stable transfectants were obtained by the Polybrene-DMSO method (14 ) using BstEII-linearized plasmids. A total of 2.5 × 108 cells were transfected in several independent transfection experiments. Cells were trypsinized 48 h after transfection and replated in selective medium containing 450 µg/ml G418. The medium was changed every 4 days and after 12-13 days neor colonies were collected. Sub-populations containing 5-10 × 103 independent clones were pooled and analysed by the flow activated cell sorting (FACS)-fluorescein di-[beta]-d-galactopyranoside (FDG) procedure. Cells were UV-C irradiated from a germicidal lamp emitting 254 nm light at a fluence rate of 0.13 J/m2/s, as monitored by a UVX digital radiometer (Ultraviolet Products, San Gabriel, CA). [gamma]-Irradiation was performed with 137Cs [gamma]-rays from a HWM-2000 machine (Siemens, Erlangen, Germany) at a dose rate of 1.2 Gy/min.
Cells were analysed by the FACS-FDG procedure according to Nolan et al. (15 ). Cells (1 × 106) were resuspended in 0.1 ml F10 medium containing 2% newborn calf serum and 10 mM HEPES. An equal volume of prewarmed (37°C) 2 mM FDG in distilled water was added and the cells incubated at 37°C for 1 min. FDG loading was stopped by adding 0.8 ml ice-cold F10 medium, 2% serum, 10 mM HEPES, containing 10 µg/ml propidium iodide (PI) to determine cell viability. PI is incorporated only by cells with damaged membranes; dead cells show a bright red fluorescence and can be easily gated out. After incubation on ice for 30 min cells were analysed and sorted with a FACStar Plus cell sorter (Becton Dickinson, Milano, Italy). The number of sorted LacZ+ cells varied among different experiments. Usually 200-400 cells were plated on one Petri dish and incubated to allow formation of visible colonies. Replica plating on polyester membranes (16 ) was performed according to Gal (17 ). Each replica contained the same number of clones of the master plate.
Cells were lysed by addition of lysis buffer (150 mM NaCl, 1 mM EDTA, 10 mM Tris-HCl, pH 7.5, 0.5% SDS, 100 µg/ml proteinase K) and subsequent incubation for 16 h at 37°C. DNA was phenol/chloroform extracted and precipitated with ethanol (18 ). DNA (15-20 µg) was digested with HindIII and BglII (2.6 U/µg DNA) according to the manufacturer's instructions (New England Biolabs). Digested DNA fragments were separated on a 0.8% agarose gel for 16 h at 25 V in TAE buffer. After electrophoresis the DNA was transferred to Hybond N+ nylon membrane (Amersham, Milano, Italy) in alkaline transfer solution (0.4 M NaOH, 0.6 M NaCl). The filter was hybridized with a 3.8 kb lacZ fragment isolated from agarose gel after digestion of pEN53 with XbaI, BamHI and HindIII. The probe was labelled with the Fluorescein Gene Images labelling kit (Amersham). The conditions for labelling, hybridization and detection were according to the manufacturer's protocols.
In situ hybridization was carried out as described (19 ). The biotinylated probe was detected with fluorescinated (FITC) avidin (Vector, Burlingame, CA). PI was used as counterstain. Slides were scored with a fluorescence microscope (Axioplan Zeiss, Oberkoken, Germany) equipped with a two wavelength filter combination suitable for FITC and PI detection.
[beta]-Galactosidase activity was measured through hydrolysis of 4-methylumbelliferyl-[beta]-d-galactoside (4-MUG). Cells were collected with a rubber policeman and cell extracts were prepared in 0.25 M Tris-HCl, pH 7.8, by three cycles of freezing and thawing as described (20 ). The enzymatic assay was performed with 2-5 µl cell extract in 1 mM MgCl2, 45 mM [beta]-mercaptoethanol, 0.85 mM 4-MUG, 100 mM HEPES, pH 8.0, in a final volume of 300 µl. The mixture was incubated for 1 h at 37°C and the reaction stopped by addition of 700 µl of 1 M Na2CO3. A sample without cell extract and one with 50 U E.coli [beta]-galactosidase (Sigma-Aldrich, Milano, Italy) were included as negative and positive controls respectively. Fluorescence was measured with a Luminescence Spectrometer LS 50B (Perkin Elmer, Milano, Italy) using an excitation filter of 390 nm and an emission filter of 460 nm. Absorbance at 595 nm was used to measure total protein content according to the Bradford technique (BioRad, Milano, Italy). [beta]-Galactosidase activity in each sample was expressed as fluorescence units (FU) normalized for absorbance at 595 nm.
To prepare total RNA a pellet of 5-10 × 106 cells was resuspended in 1 ml TriZol reagent (Gibco-BRL Life Technologies, Milano, Italy) and processed according to the manufacturer or stored at -20°C. Cloning of fusion transcripts was performed with a 5'-RACE kit (Gibco-BRL Life Technologies) according to the manufacturer's instructions. First strand cDNA was synthesized from 1 µg total RNA using a lacZ-specific primer (lac3, 5'-CCGTGCATCTGCCAGTTTGAGGGGA-3') and SuperScripttm II reverse transcriptase, digested with RNase H and tailed with oligo(dC) by TdT. A first round of PCR (40 cycles) was performed using the dG- and dI-containing anchor primer provided in the kit and lac3. The product of PCR was identified on agarose gels as a faint 1.4 kb band. The band was excised and purified using JETSORB resin (Genomed GmbH, Germany). The purified fragment was amplified in a second PCR with the anchor primer and an internal lacZ-specific primer (M4, 5'-GCCATTCAGGCTGCGCAA-3'), purified with JETSORB and directly sequenced using the ABI PRISMtm Dye Terminator Cycle Sequencing kit (Perkin Elmer).
The gene trap vector pEN53 was introduced into CHO cells by transfection and a large number of neor independent clones (32 × 103) was obtained. Discrimination between transfected cells expressing (LacZ+) or not expressing (LacZ-) lacZ was performed by the FACS-FDG technique. In this assay [beta]-galactosidase specifically cleaves a [beta]-galactoside analog, FDG, releasing free fluorescein, which can be detected by FACS. The vast majority of neor clones did not express lacZ, either because the tagged gene was expressed only under some conditions (i.e. different phases of the cell cycle, treatments, stress) or because of the type of insertion. Only 3-5% of neor clones were LacZ+ without any treatment, indicating that insertion occurred in constitutively expressed genes. This fraction of LacZ+ cells was discarded.
The obtained LacZ- cell population was expanded for 3 days before treatment with UV, to avoid LacZ+ cells arising, a phenomenon already observed by Nolan et al. (15 ) and due, according to those authors, to the influence of epigenetic factors and/or cell cycle-regulated expression of the trapped genes. The LacZ- cell population was treated with 254 nm UV light (2, 4, 8 or 16 J/m2) and [gamma]-rays (0.5 or 3 Gy). At different times after treatment (2, 6 and 20 h) cells were harvested, stained with FDG and LacZ+ cells were sorted by FACS. The average of treatment-induced LacZ+ cells was 0.05%. Figure 1 shows an example of flow sorting of cells 6 h after 2 (upper panel) or 8 J/m2 irradiation (lower panel).
Sorted LacZ+ cells were directly plated on Petri dishes and allowed to form colonies. Replica plating was used to obtain up to four replicates for each Petri dish. Each replica was treated with UV light, [gamma]-rays or left untreated and stained with X-gal to detect blue colonies expressing lacZ only after DNA damaging treatment (Fig. 2 ). Putative inducible colonies which showed either a newly arising or a more intense blue colour after UV or [gamma]-ray treatment were picked from the master plate or from one untreated membrane and expanded. We isolated 30 neor clones which showed intense blue staining after treatment on replicas. Three clones (VI, 4 and AC) that showed a more intense blue stain after UV irradiation were chosen for further analyses.
To evaluate the status of integration, genomic DNA of three clones was digested with HindIII, which cuts to the right of the neo gene of the vector, or with BglII, which does not cut the vector, and analysed by Southern blot using a lacZ probe. In both cases a single insertion event should result in a single hybridizing band. Among the three independent clones only clone VI had a single lacZ-hybridizing band with both restriction enzymes, indicating that insertion occurred at a single genomic site (Fig. 3 ). In clone 4 the two bands observed with BglII-digested DNA could be derived from two insertion events. However, HindIII digestion revealed only one band, suggesting a possible rearrangement within the construct. An analogous interpretation can be applied to clone AC.
In situ hybridization of metaphase chromosomes from clone VI with a biotinylated pEN53 probe showed the presence of a lacZ-hybridizing region on a single chromosome (Fig. 4 ).
The lacZ inducibility in these clones was evaluated by measuring the enzymatic activity of [beta]-galactosidase. We used a fluorimetric MUG assay (21 ) that in our hands was ~30-fold more sensitive (data not shown) than the colourimetric assay commonly used for measuring [beta]-galactosidase activity (20 ), based on hydrolysis of o-nitrophenol-[beta]-d-galactoside.
Among the clones analysed clone VI gave the most reproducible and highest level of induction of [beta]-galactosidase activity after UV treatment. In clones 4 and AC [beta]-galactosidase activity was only slightly increased (Table 1 ).
Table 1 .
Our study concentrated on clone VI, which showed a 3-fold induction of [beta]-galactosidase activity 8 h after UV irradiation. A slight increase in enzymatic activity was also observed with the same kinetics in mock-treated cells. Since the experimental procedure implies addition of fresh medium immediately after treatment, these results may indicate that activation of the trapped gene is not only UV dependent, but is also stimulated by serum factors (growth factors or cytokines). It is known that both UV and growth factors are able to activate the src family of tyrosine kinases and trigger the pathway leading to transcription of AP-1-regulated genes (3 ). Recently an involvement of growth factor receptors in the UV response has been elegantly demonstrated in HeLa cells (22 ). In clone VI, when conditioned medium was added after treatment, the response was completely abolished both in UV- and in mock-treated cells (Fig. 5 ), indicating that the trapped gene is induced by UV only in the presence of fresh serum. It is well known that growth factors induce different genes involved in proliferation control: early response genes are induced immediately after addition of serum, while delayed genes, which require protein synthesis, are induced more slowly (23 ). At the moment we cannot assign a function to this gene, but its requirement for serum might indicate a role in cell proliferation. We are now studying cell cycle-dependent regulation of this gene and preliminary data seem to suggest a role in S phase (manuscript in preparation). However, whatever the function of this gene might be, its involvement in the cellular response to UV irradiation needs to be further investigated.
An important advantage of using the gene trap approach is the possibility of easily isolating the endogenous trapped gene by 5'-RACE-PCR, taking advantage of the formation of a fused transcript between the endogenous gene and the lacZ gene. Moreover, clone VI is a good candidate for isolation of the trapped gene, since only one insertion site has been used for integration, as shown by Southern blot and FISH analyses (Figs 3 and 4 ). A portion of the endogenous trapped gene in clone VI was sequenced starting from lacZ fusion transcripts isolated from induced cells. 5'-RACE was used to amplify cDNA sequences upstream of the lacZ reporter gene. After the first amplification a second PCR with a nested lacZ-specific primer and an anchor primer was performed, giving rise to a 1.2 kb fragment containing 480 bp of the lacZ gene and ~0.7 kb of endogenous sequence. DNA sequencing was performed directly on purified PCR products. Figure 6 shows the sequence of the splice acceptor site used in the gene trap vector and the nucleotide sequence of the amplified cDNA. The splice acceptor site in the gene trap vector was used properly, since a novel sequence not present in the vector was found upstream of the splice junction. An open reading frame, in-frame with lacZ, extended upstream of the splice site. The identified gene was novel, since it did not correspond to any sequences in the GenBank database.
We thank Dr E.Neilan for providing the pEN53 construct, Dr A.Inga for his help in DNA sequencing and Dr H.Vrieling for critical reading of the manuscript. The authors wish to thank Dr M.Nüsse for his help in pilot FACS-FDG experiments. This work was partially supported by the Italian Association for Cancer Research (AIRC).
This article has been cited by other articles:
Clone
UV dosea
Time after UV (h)
(J/m2)
2
4
8
14
VI
4
1.5 ± 0.0
2.1 ± 0.1
3.4 ± 0.7
2.2 ± 0.1
4
4
1.1 ± 0.2
1.2 ± 0.4
1.8 ± 0.3
1.8 ± 0.5
AC
16
1.3 ± 0.0
1.4 ± 0.3
2.7 ± 0.5
nd
REFERENCES
CiteULike 
Connotea 
Del.icio.us What's this?
H. G.E. Sutherland, G. K. Mumford, K. Newton, L. V. Ford, R. Farrall, G. Dellaire, J. F. Caceres, and W. A. Bickmore
Large-scale identification of mammalian proteins localized to nuclear sub-compartments
Hum. Mol. Genet.,
September 1, 2001;
10(18):
1995 - 2011.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Print PDF (143K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (7)
Request Permissions
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Menichini, P.
Articles by Abbondandolo, A.
Search for Related Content
PubMed
PubMed Citation
Articles by Menichini, P.
Articles by Abbondandolo, A.
Social Bookmarking
What's this?