Office of Biological and Environmental Research

DOE Low Dose Radiation Program Workshop V

2005 Abstract

Title: Cooperation Between Homologous Recombination and the Fanconi Anemia Cancer Suppressor Proteins in Minimizing Spontaneous and Radiation-Induced Chromosomal Instability

Authors: Larry H. Thompson, John M. Hinz, Robert S. Tebbs, and N. Alice Yamada

Institutions: Biosciences Directorate, Lawrence Livermore National Laboratory, Livermore, California

Purpose and experimental approach. This study addresses the genetic basis of spontaneous mutagenesis as a means of understanding the DNA damage-response pathways that maintain chromosome stability. It is our view that knowledge of these processes is fundamental to understanding how low dose ionizing radiation (IR) produces chromosomal rearrangements that lead to carcinogenesis. Endogenous oxidative DNA damage is presumed to be a major factor determining the spontaneous mutation rate at individual gene loci in cultured cells and cancer rates in people. The hprt locus is widely used as a surrogate marker for loss of gene function because it is readily measured in dividing cells and can be analyzed in detail for the molecular nature of the mutations.

Using Chinese hamster CHO cells, we performed gene targeting at the fancg and rad51d loci to make isogenic knockout mutants that are defective in the Fanconi anemia (FA) cancer suppressor "pathway" and homologous recombination (HR), respectively. In each case the defect in the mutant cells was corrected by transfection and complementation by the normal gene derived from a BAC-clone library of CHO genomic DNA. Importantly, essentially full complementation was observed for both genes, demonstrating that the phenotypic changes seen with each mutant are caused by the specific gene knockout. A detailed comparison of fancg and rad51d mutant cells has provided a new perspective on how cells minimize chromosomal rearrangements when they replicate damaged DNA and has indicated the potential relevance of the FA pathway to low dose IR damage.

Insights from fancg knockout cells. Fancg mutant cells grow almost normally and have increased sensitivity to killing by DNA crosslinking chemicals, as expected, but also simple alkylating agents such as methyl methanesulfonate and methylnitrosourea ¹. Notably, the sensitivity to monofunctional ethylnitrosourea and crosslinking chloroethylnitrosourea is the same (3-fold). Sensitivity to killing by IR is slight, but, interestingly, it is not associated with any change in of cell-cycle-dependence of IR sensitivity. We think this finding is consistent with the idea that fancg acts during S phase in a manner than is independent of cell-cycle phase at irradiation. Fancg cells have normal spontaneous sister-chromatid exchange, IR-induced Rad51 focus formation, and spontaneous chromosomal aberrations. Importantly, gene amplification rates for the CAD and DHFR loci are elevated 3- to 4-fold in fancg cells as shown by fluctuation analysis. We find a reduced spontaneous mutation rate at the hprt locus in fancg cells as well as reduced mutagenesis at this locus produced by UV-C, ethylnitrosourea, and IR γ-rays delivered to G1-synchronized cells. These results suggest that fancg is important for replicating DNA containing poorly repaired oxidative base damage produced by IR; potentially mutagenic base damage in or near the hprt gene may be converted to lethal events in fancg cells. This is the first instance of any mammalian cell showing reduced spontaneous or IR-induced mutation rates.

In synchronized cell populations obtained by centrifugal elutriation, we find that fancg cells have relatively normal levels of spontaneous γH2AX foci, which increase as cells progress from G1 to G2 phase. These results suggest that γH2AX foci normally arise at replication fork-blocking, spontaneous DNA lesions, perhaps without breakage of blocked forks (in contrast to γH2AX-marked double-strand breaks [DSBs] that arise from IR independently of DNA replication). However, after a 30-minute mitomycin C (MMC) treatment of early-S phase cells (which is nontoxic to wild type and produces 85% survival of fancg cells), we see more γH2AX foci in fancg cells than in control cells. Their persistence is associated with a lengthening of S phase, suggesting that HR can slowly, but effectively, restart most broken forks arising from the processing of crosslinks and produce viable fancg cells at this minimally toxic dose.

Based on all these findings with fancg cells and the FA literature, we propose that the primary role of most FA proteins, mediated through monoubiquitination and nuclear focus formation of FANCD2, is to: (a) promote mutagenic translesion synthesis (TLS) past oxidative DNA damage and thereby prevent the collapse of DNA replication forks at the cost of producing point mutations; (b) promote the restoration of stand continuity, by HR, of broken forks that arise spontaneously or after mutagen treatment. Thus, in the case of crosslinking damage, the FA proteins are uniquely needed for both processes, first HR-mediated restart of the broken fork associated with crosslink unhooking by XPF/ERCC1 endonuclease and, second, TLS past unhooked crosslinks. From our findings, we predict that the FA proteins are needed for efficient replication of DNA containing non-DSB bistranded, clustered damage from low dose IR. This hypothesis will be tested in our new DOE project using primary human fibroblasts at low IR doses.

Insights from rad51d knockout cells. The important role of the five Rad51 paralogs in chromosome stability after exposure to IR, and many other DSB-producing agents, is well documented in studies of rodent and chicken mutants, but the relationship of the paralogs to mutagenesis per se has not been examined. We used Cre-Lox conditional gene targeting to produce CHO cells targeted at both alleles of RAD51D. Importantly, rad51d cells grow quite well with a doubling time of ~16 h compared with ~13 h for the parent lines (51D1Lox and wild-type AA8). The high level of spontaneous chromatid breaks (6-fold increase) can account for the diminished growth capacity and plating efficiency. The rad51d cells are ~75-fold more sensitive than the parental cell lines to killing by MMC and ~1.8-fold sensitive to IR; they also show UV-C and MMS sensitivity. Rad51d cells have a 4- to 7-fold increase in spontaneous gene amplification at both the DHFR and CAD loci, suggesting increased chromosomal breakage is driving amplification. Very importantly, fluctuation analysis reveals a large increase (~12-fold) in spontaneous mutation rate at the hprt locus in the rad51d cells, and the majority of these mutants have deletions of part or all of the hprt gene. This study is the first to determine mutation and gene amplification rates for a Rad51 paralog mutant, revealing the critical role of HR in minimizing gene rearrangements.

Fig. 1. Model to account for the altered hprt mutation rates and chromosomal instability associated with mutations in the FA pathway or the homologous recombination pathway. Details of our model of the physiological function of FA proteins are presented ina recent review ².

Molecular model of FA and HR proteins in chromosome stability. Fancg and rad51d mutations cause reduced or increased hprt mutations rates, respectively, but in both cases the mutation spectrum shifts toward a higher proportion of deletions/rearrangements (Fig. 1). Thus, we propose that HR efficiently restarts broken replication forks without mutation in normal cells, but not in fancg or rad51d cells. Our model distinguishes between broken forks that are stabilized by the FA proteins versus those that become displaced (defined as fork collapse) in the absence of fancg or other FA proteins (2). Thus, the FA and HR proteins cooperate to ensure replication fork stability for a wide variety of DNA damages. Specific protein interactions that have been identified include a direct interaction between fancg and the Rad51 paralog XRCC3 in human cells.

(Work done under the auspices of the US DOE by the University of California, LLNL, under Contract No. W-7405-Eng-48 and funded by the DOE Low-Dose Program and NCI/NIH grant CA89405.)

1. Tebbs, R.S., Hinz, J.M., Yamada, N.A., Wilson, J.B., Salazar, E.P., Thomas, C.B., Jones, I.M., Jones, N.J. and Thompson, L.H. (2005). New insights into the Fanconi anemia pathway from an isogenic FancG hamster CHO mutant. DNA Repair 4:11-22.
2. Thompson, LH, Hinz JM, Yamada NA, Jones NJ (2005). How Fanconi anemia proteins promote the four Rs: Replication, recombination, repair, and recovery. Environmental and Molecular Mutagenesis 45:128-142.

» Return to top