Abnormal response to DNA crosslinking agents of Fanconi anemia fibroblasts can be corrected by transfection with normal human DNA. (repair defect/gene ...
Proc. Nati. Acad. Sci. USA Vol. 83, pp. 7034-7038, September 1986 Medical Sciences
Abnormal response to DNA crosslinking agents of Fanconi anemia fibroblasts can be corrected by transfection with normal human DNA (repair defect/gene transfer/mitomycin C/selection by growth rate)
CATHERINE DIATLOFF-ZITO, DORA PAPADOPOULO, DIETRICH AVERBECK, AND ETHEL MOUSTACCHI Institut Curie-Biologie, 26 rue d'Ulm, 75231 Paris CUdex 05, France
Communicated by Richard B. Setlow, May 16, 1986
Primary skin fibroblast cell lines from paABSTRACT tients with Fanconi anemia were cotransfected with UVirradiated pSV2neo plasmids and high molecular weight DNA from normal human cells. Restoration of a normal cellular resistance to mitomycin C (MMC) was observed provided that a Fanconi anemia cell line is selected for DNA-mediated transformation (neo gene) and that at least two successive rounds of transfection are performed. Cells were selected by taking advantage of the higher proliferation rate and plating efficiency of the MMC resistant transformants. As estimated from reconstruction experiments, the frequency of transfer of MMC resistance lies between 1 and 30 X 10-. The MMC resistance phenotype was maintained for at least 10 generations following transfection. Evidence for DNA-mediated transformation also includes the recovery of a normal pattern of DNA semiconservative synthesis after treatment with 8-methoxypsoralen and 365-nm UV irradiation, and the presence of exogenous pSV2neo DNA sequences was shown by Southern blot analysis. The acquired MMC resistance is probably due to the presence of DNA from normal cells. Indeed, sensitivity to MMC was maintained when Fanconi anemia cells were cotransfected with the UV-irradiated pSV2neo plasmid mixed with their own DNA or with yeast or salmon sperm DNA. These negative results also render unlikely the selection of spontaneous MMC resistant revertants in transfection of Fanconi anemia cells with normal DNA. These experiments establish the prerequisites for the isolation of the gene(s) involved in the response to DNA crosslinking lesions in human cells.
of human cells to DNA crosslinking agents. To investigate these aspects, a project aimed at the isolation of the genes involved was initiated, and we established conditions for the DNA-mediated transfer and expression of wild-type genes in FA cell lines. The isolation and characterization of the transformed population obtained are described here.
MATERIALS AND METHODS Cell Lines and Cell Cultures. Three FA primary skin fibroblast cell lines: FA 150, FA 121, and FA 109 were obtained from R. Voss (Hadassah Hospital, Jerusalem). The normal cell line 1BR3 was obtained from C. Arlett (MRC, University of Sussex, UK), Jacquot skin fibroblasts were started in our laboratory from a 7-year-old male child, and HT-1080 human fibrosarcoma cell line was from ATCC (Rockville, MD). TK6-MA, a human lymphoblastoid cell line, was obtained from W. Thilly (MIT, Boston). The characteristics of the FA and Jacquot cell lines have been described (19). Cells were routinely maintained in minimum essential medium (MEM, GIBCO) supplemented with 13% (vol/vol) fetal calf serum and gentamycin at 16 pug/ml in an atmosphere of 6% C02/94% air. Determination of Cell Survival. For the measurement of cell survival to mitomycin C (MMC, Sankyo, Japan), 4 x 105 exponentially growing cells were incubated for 1 hr in aMEM medium (GIBCO) without nucleosides containing various concentrations of the drug. MMC was removed by washing with phosphate-buffered saline (NaCl, 137 mM/KCl, 2.7 mM/KH2PO4, 1.5 mM/Na2HPO4 12 H20, 8.1 mM, pH = 7.0), and after trypsinization the cells were plated on 12 dishes at different cellular concentrations for each dose of MMC to determine survival. Clones were counted 15 days later. The cloning efficiency was 20-30% for normal cells, whereas it was 10 times lower for FA cells. For reconstruction experiments, 1BR3 and FA 150 fibroblasts were cocultivated in various proportions from 10-2 to 10-, and the mixed populations were treated with MMC to measure cell survival. Preparation of Genomic DNA and Filter Blot Hybridization. For transfection experiments high molecular weight DNA was isolated from confluent cultures of FA or normal cells using the proteinase K/NaDodSO4 treatment (21) followed by two protein removal steps with phenol and chloroform/ isoamyl alcohol and RNase (22). In controls, genomic DNA was exposed to a dose of 254-nm UV irradiation (200 J/m2) before transfection. For Southern blot analysis DNA samples were either not digested or digested with the restriction endonucleases according to conditions recommended by the supplier (New England Biolabs), subjected to gel electrophoresis, and transferred to nitrocellulose filters by the method of Southern (23).
Fanconi anemia (FA) is an autosomal recessive disorder in which patients show pancytopenia and predisposition to leukemia and carcinoma (1, 2). These patients show a high level of chromatid-type damage in their lymphocytes and fibroblasts (2-4). Cultured FA cells are unusually sensitive to DNA crosslinking agents by chromosomal criteria (5-9) and by clonogenic cell survival (10-14) whereas their sensitivity to monofunctional agents or to radiation is close to normal (12, 15). A number of antitumor drugs and environmental pollutants produce interstrand crosslinks in DNA. Hence, it is important to understand the consequences of such lesions. Since FA is characterized by an almost specific sensitivity to DNA crosslinking agents it may serve as a model system for research in this area. The nature of the repair defect in FA cells is, however, still debated (11, 14, 16-18). Heterogeneous responses of various FA cell lines to a DNA crosslinking treatment (19) and the presence of at least two complementation groups in FA (20) has been shown. However, almost nothing is known about the chromosomal location, expression, and regulation of the genes controlling sensitivity The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Abbreviations: FA, Fanconi anemia; MMC, mitomycin C; 8-MOP, 8-methoxypsoralen; kb, kilobase(s); S, sensitive; R, resistant.
7034
Medical Sciences: Diatloff-Zito
et
al.
Transfection Protocol. The transfection procedure was essentially the calcium phosphate precipitation technique of Graham and Van der Eb (24). For each separate transfection experiment, FA fibroblasts were plated in 8-16 flasks at a concentration of 5 x 105 cells per 25-cm2 flask. Two days later the cells in each flask were exposed to a DNA precipitate of 13 pig of genomic DNA from normal cells with 2 ,ug of the plasmid pSV2neo (25). In all experiments the pSV2neo plasmid was UV irradiated (254 nm) at a dose of 200 J/m2, which causes an increase in the yield of transformants (26). After 4 hr of adsorption the medium was removed, and the cells subjected to a 2-min shock with 15% (vol/vol) glycerol in MEM. The glycerol was removed, and the cells washed twice with MEM plus 0.5% fetal calf serum. The cultures were then incubated in MEM for a week with a medium change every 3 days. At this stage the cells were again transfected and processed as above. For selection of G418resistant transferants, a week after the second transfection, the medium was replaced by MEM containing geneticin at 400 4g/ml (GIBCO). After 18-21 days, colonies of G418resistant cells were visible. For the selection of a MMC-resistant (R) population, 8 days after the 2nd transfection, cultures were plated at 4 x 105 cells per 25-cm2 flask and characterized with respect to MMC sensitivity 3 days later (day 11). After a dose of MMC (0.025 ,ug/ml) leaving 30% and 100% survivors in FA and normal cell lines, respectively, cultures were retested for MMC sensitivity after 4 passages at 1:2 culture ratio, on day 40. In parallel, cultures not selected with MMC were also grown for 4 passages and treated with MMC on day 40 to measure cell survival. As controls, cells transfected with UV-irradiated normal DNA, FA, salmon, or yeast DNA or mock transfected, always in the presence of UV-treated pSV2neo plasmid, were submitted to the same protocol. All treatments with MMC were performed by incubating the cells for 1 hr with the drug. Cultures, including the transfected ones, were routinely assayed for mycoplasma contamination (Myco Tect Assay, Bethesda Research Laboratories) and were found to be mycoplasma free.
RESULTS FA Cell Line Response to DNA Crosslinking Agents. In accord with results obtained by others (10-12, 14) the FA 109, FA 150, and FA 121 cell lines were found to be extremely sensitive to the cytotoxic action of MMC. FA 109 is 11 times and FA 121 and FA 150 are =40 times more sensitive than a normal cell line (1BR3) as estimated from the ratios of the doses at 37% survival level (LD37) (Fig. 1). The FA 150 cell line exhibits a pronounced sensitivity to MMC and belongs to the class of FA lines that did not recover a normal rate of DNA synthesis after another DNA-crosslinking treatment,
i.e., 8-methoxypsoralen (8-MOP) and 365-nm UV irradiation (19). These characteristics together with the competence for DNA-mediated gene transfer (see below) led us to focus our transfection experiments on FA 150. Correction of the FA Defect by Normal DNA. The competence for transfer of the bacterial neo gene was compared in FA 109, FA 121, FA 150, and HT 1080 cells, by transfection with the pSV2neo plasmid and carrier DNA. In the rapidly growing HT 1080 cell line, G418-resistant transformants were found at a frequency of 5 x 10-5 whereas such transformants were not detected in the three FA cell lines. UV irradiation of the plasmid had a stimulatory effect in that transformants were produced in the cell lines FA 150 and FA 121 at frequencies of 4 x 10-6 and 10-5, respectively. Transformants in these FA cell lines were observed only after two rounds of transfection. Therefore, the double transfection procedure was systematically applied. No G418-resistant
Proc. Natl. Acad. Sci. USA 83 (1986) 100
0.2
0.4
0.6
7035
0.8
I3 L-
10
MMC, Ag per ml per hr FIG. 1. Survival after MMC treatment (1 hr) of normal HT 1080
(e), normal 1BR3 (A), FA 109 (A), FA 121 (n), and FA 150 (o) cell
lines.
cells were derived from the FA 109 cell line under these conditions. Using the approach of selecting clones able to express, at the same time, a cloned bacterial gene cotransfected with a second marker (DNA repair gene) (27-3 1), we failed to isolate MMCR clones. Indeed, our primary untransformed FA skin fibroblasts were unable to grow under clonal conditions following the usual transfection and selection protocol. FA cells exhibit a slower growth rate than normal cells (12, 32, 33). Thus a protocol was devised that allows the selective pressure to favor the putative fast growing MMCR cells by keeping cells after transfection in conditions compatible with long-term cell growth. After transfection of FA 150 with high molecular weight DNA from normal human cells, clonogenic survival to MMC was characterized. The survival curves demonstrated a tailed shape indicating the presence of an important fraction of MMCR cells in the population (Fig. 2). When FA 150 cells were cotransfected with UV-irradiated pSV2neo plasmid and with UV-irradiated DNA from normal cells (eight experiments), from autologous cells (FA 150) (six experiments), from salmon sperm, or from yeast or were submitted to the transfection procedure without DNA, an 0
0.025
0.05
0.075
0.1
100-----------10
1
MMC, ,ug per ml per hr FIG. 2. Survival after MMC treatment of FA 150 fibroblasts transfected with Jacquot or TK6 MA DNA, 11 days after DNA transfer (A, o). For FA 150 cells transfected with TK6 MA DNA survival was established also 40 days after the transfection with (0) or without (i) MMC selection. The maximal standard deviation on each point is ±5%. Range of untransfected FA 150 cells and FA 150 cells cotransfected with UV-treated pSV2neo plasmid and their own DNA, UV-irradiated DNA from normal cells, yeast and salmon DNA, or mock transfected (hatched curves including 22 survival curves).
7036
Medical Sciences: Diatloff-Zito
et
al.
Proc. Natl. Acad. Sci. USA 83
MMCR fraction was never observed. In these cases the ratios of LD37 values of transfected over those of nontransfected populations ranged between 1 and 1.4 (Fig. 2). This implies that the plasmid by itself or the mock transfection do not confer MMC resistance. Among 10 transfection experiments performed with intact normal DNA, 3 experiments with DNA from the fibroblast cell line Jacquot (2) and from the lymphoblastoid cell line TK6 (1) gave positive results-i.e., showed the presence of MMCR cells as in Fig. 2-and were further analyzed. Six experiments lead to a MMC survival intermediary between a typical FA and a normal response [the sources of normal DNA were 1BR3, HT 1080, and Jacquot cell lines (two experiments in each case)] and were discarded. A failure in uptake of exogenous DNA does not explain these results since G418-resistant clones at the usual frequencies were observed in the same experiments. Finally, one experiment, performed with a DNA from a normal embryonic cell line (GM 10), was negative. Stability of the Acquired MMC Resistance. The transformed population was grown for 4 weeks (four passages) without MMC selection, and cells were then subjected to MMC. As seen in Fig. 2 the survival curves still had a tail, and the break in the curve was at a higher survival level indicating an enriched MMCR population. In other words, the MMC resistance was retained after several cell divisions. After treatment of the FA 150-transfected cells with MMC at 0.025 ,ug/ml for 1 hr, survivors were grown for four generations and subjected to MMC again. The survival curves are indicative for a homogeneous population of cells with a normal resistance to MMC (Fig. 2). Consequently, one round of MMC selection eliminated the MMC sensitive (S) cells. It should be noted that 10 generations after transfection, the MMC resistance was maintained, and the measured doubling times of FA 150 and MMCR fibroblasts were 4 and 2 days, respectively. Reconstruction Experiments. By mixing various proportions of normal (1BR3) and FA 150 cells and testing for MMC clonogenic cell survival in conditions mimicking a transfection experiment, tailed survival curves (Fig. 3) similar to those following transfection with normal DNA (Fig. 2) were obtained. Extrapolation to the ordinate axis of the second part of survival curves allows us to estimate the proportion of MMCR cells. Such calculated fraction (30-60%) is higher than the expected fraction from the experimentally mixed population (10-4 to 10-2). A diffusible factor produced by 0
0.025
0.05
0.075
0.1
(1986)
normal cells cannot explain this result. By cocultivation of FA and normal cells at a 1:2 ratio, which is much higher than in our case, an increase in chromosomal aberrations in the normal cells and only a slight correction in FA cells were observed (34). The higher growth rate and plating efficiency of MMCR cells seems to account for the data (Fig. 3). Comparison of survival curves of FA cells transfected with normal DNA (Fig. 2) and reconstruction survival curves (Fig. 3) indicates that the proportion of corrected FA 150 cells in transfection experiments may be 10- to 10-4. Some Characteristics of Transfected FA 150 Cells. Determination of HLA antigens. To exclude inadvertent contamination of FA 150 cells by MMCR cells in transfection experiments, FA 150 cells and MMCR cells derived from a transfection were typed for histocompatibility antigens (HLA). A standard serologic assay (35), based on complement-dependent microcytotoxicity with alloantisera, allows identification of HLA A2, A24, B5, B13, BW4, and 6 antigens at the cell surface of both cell lines analyzed. When the two cell lines were compared by serological reactions using 102 monospecific reagents, 98 reactions were found to be concordant leading to an r value of 0.88 (using a two-by-two table of comparison with one degree of freedom, the x2 value was 79, P
