Repair and Persistence of DNA Damage in Chromosomes With Diverse Gene Densities: Insights From Normal and Xeroderma Pigmentosum Cells Using Reverse FISH* *

1998 EEMS-Young Scientist Award Lecture by Jordi Surrallés

Fax: +343 581 23 86; Phone: +343 581 25 97; E-mail: jordi.surralles@blues.uab.es

Keywords: Chromosomes; CpG islands; Gene density; FISH; Repair; Xeroderma pigmentosum

Introduction: from micronuclei to DNA repair

My initiation in the field of Mutagenesis was made possible thanks to my former professor of Genetics Ricardo Marcos who put his trust in me almost 10 years ago. He was leading a small research group at the Department of Genetics of the Autonomous University of Barcelona. This group is nowadays the Group of Mutagenesis, one of the biggest and most dynamic research teams in Spain. He asked me to read a paper by Michael Fenech describing the micronucleus (MN) assay in human lymphocytes (Fenech and Morley, 1985). I soon learned that MN had something to do with Paramecium aurelia also being chromosome fragments or whole chromosomes that are left behind during anaphase and appear in the cytoplasm of daughter cells as small additional nuclei. As Ricardo knows very well, my attempts to break and lose chromosomes, and therefore to induce MN, turned out to be my life sentence: thousands of hours of microscope, hundreds of blood samples, and too many hours in front of a computer screen. I even used to dream I was floating inside a giant MN trying to find out whether it was a chromosome fragment or a misegregating whole chromosome.

Soon after the beginning of my PhD I discovered that travelling was fun and I started

to practice scientific tourism by visiting Caterina Tanzarella's laboratory in Rome, in 1991. We set up a fruitful collaboration with the aim standardizing the MN assay. During my stay in Rome, Caterina showed me a nice picture of the first MN after kinetochore immunolabelling (Degrassi and Tanzarella, 1988). This photograph was the end of my dream and the beginning of a nightmare since I had to learn extremely complicated protocols to study the dual origin of MN: fluorescence in situ hybridization (FISH) and immunocytogenetics. I then moved to Michelline Kirsch-Volders's laboratory where I used to rush behind Paul van Hummelen while he was doing his first pancentromeric FISH. I gained more experience with FISH in Marja Sorsa's and Hannu Norppa's laboratory in Helsinki and in immunocytogenetics with Michael Fenech and Peter Jeppesen in Australia and Scotland, respectively. My most sincere acknowledgement to all these people that have contributed to my scientific training.

My trip to Australia was not only tourism. Michael Fenech proposed a simple way to study DNA repair by converting excision repairable DNA lesions to MN with the use of a DNA repair inhibitor Ara-C, that blocks the refilling step of excision repair. The unfilled gaps were then converted to chromosome breaks and, subsequently, to MN after S phase (Fenech and Neville, 1992). We proved this model by converting UV-light induced excision repairable DNA lesions to MN and, as hypothesized, all the induced MN were kinetochore negative and, therefore, derived from chromosome fragments (Fenech et al., 1994). We then applied this methodology to study chemically induced excision repair (Surrallés et al. 1995). This is the silly way a MN lover was introduced into the field of DNA repair at the cytogenetic level.

Heterogeneous repair in chromosomes with high and low gene density

As any young scientist interested in breaking and sticking chromosomes, I was following the steps of a wise man called A.T. Natarajan, Nat. I was really lucky he accepted me to join his laboratory in the Department of Radiation Genetics and Chemical Mutagenesis in Leiden, Holland. After a few coffee breaks and late-night beers, Nat completely changed my concept of science. He had and still has a deep personal and scientific influence on me, so I focused all my efforts on his main research line: the mechanisms of chromosomal aberrations and how chromatin structure and DNA repair modulate chromosome fragility.

I initially recovered the MN approach of using DNA repair inhibitors to convert excision repair unfilled sites to chromosome breaks with the aim of comparing induced breakage in chromosomes with high and low gene density. The rational of this approach is straight: gene density reflects the heterogeneity of diverse chromosomes in terms of gene activity and chromatin structure. Assuming that DNA repair is highly influenced by chromatin structure and genetic activity, we hypothesized that high gene density chromosomes would be preferentially repaired and, therefore, more sensitive to the inhibitory effects of the DNA repair inhibitor.

In order to compare DNA repair in chromosomes with high and low gene density, the next step was to find out an easy way to "see" the gene density of a given chromosome. The use of the word "see" was deliberate. Some years ago a Nobel laureate said that nothing is proved until it is demonstrated at the molecular level. My problem is that I only believe what I see with my own eyes. This is probably one of the many reasons why he got the Nobel prize and I will never do. Anyhow we managed to visualize gene density cytogenetically by using reverse FISH with a probe enriched in CpG islands that we had isolated from human genomic DNA (Surrallés et al., 1997a). This approach belongs to Craig and Bickmore (Craig and Bickmore, 1994) and it is based on the widely accepted assumption that most genes have CpG islands in their regulatory regions. We digested genomic DNA with the restriction enzyme HpaII cutting CCGG sites. As high gene density segments are rich in CCGG sites, many tiny fragments are generated by the HpaII digestion. After ³²P end-labelling and electrophoresis, these tiny fragments can be recovered from the gel, labelled with digoxigenin and used as a probe to paint back the chromosomes by FISH.

After this procedure, Nat was plethoric when he "saw with his eyes" a metaphase where some chromosomes were completely banded and some others were not. Chromosome 18 and 19 appeared to be the ones with the lowest and highest gene density signal, respectively. Amongst the big chromosomes, we selected chromosome 1 and chromosome 4 as examples of high and low gene density chromosomes, respectively. We then treated human lymphocytes with the ethylating agent EMS and Ara-C, and analysed induced chromosome breakage in chromosomes 1,4,18, 19 and 20 by chromosome painting. After correcting for the DNA content, chromosomes with increasing gene density were increasingly more sensitive to the inhibitory effect of Ara-C upon the EMS induced repair sites. Considering that functional base and nucleotide excision repair is required for the clearance of ethylation damage from DNA (Sitatam et al., 1997), we interpreted our results as showing that chromosomes with high gene density are preferentially repaired in human cells (Surrallés et al., 1997a).

Towards a repair karyotype: insights from xeroderma pigmentosum cells

It is well known that nucleotide excision repair (NER) is preferentially directed to actively transcribed genes and their transcribed strands through transcription coupled repair mechanisms (Friedberg et al., 1996). Since gene density and activity is clustered in some chromosomes and chromosomal regions, the next challenge was to visualize those chromosomal regions subject to preferential repair. These regions should theoretically co-localize with those regions rich in CpG islands, since chromosomal bands harboring actively transcribed genes would be expected to be preferentially repaired. The final aim was to generate what we called a "repair karyotype". This repair karyotype would allow us to corroborate our observations that chromosomes with high gene density, such as chromosome 1 or 19, are preferentially repaired when compared to gene poor chromosomes, such as chromosome 4.

When somebody is fully ignorant about something, the best she/he can do is to seek for advice, so I suggested the idea of generating a repair karyotype to a world authority in the field of DNA repair, Leon Mullenders. Considering that Leon was on his feet all day long talking about DNA repair and that his office was only two meters away from my laboratory bench, our interaction, as well as my first contact with repair deficient cell lines, was almost unavoidable.

Xeroderma pigmentosum group C (XPC) cells are known to be partially deficient in NER since they can only repair the transcribed strands of active genes (Venema et al., 1991). UV-induced repair patches in confluent XPC cells were labelled with BrdU. After EcoRI digestion and CsCl gradient, unreplicated DNA fragments containing BrdU-repair patches were isolated with antibodies against BrdU and an immunomagnetic system which is too long to explain now (Kalle et al., 1991). The extracted repaired fragments were random primed with biotin and used as a probe to perform reverse FISH to metaphases from normal cells. In other words, we mapped repaired DNA by FISH. The chromosomal distribution of FISH signals was compared to that found with the unrepaired fractions containing non-repaired bulk DNA and the non-transcribed strands of active genes. To further control the distribution of repair sites in human chromosomes we also generated similar probes with wild type cells and xeroderma pigmentosum group A (XPA) cells, which are known to be completely deficient in NER. Preliminary observations indicated that there are clusters of transcription coupled repair in early replicating R bands (light G bands) and, therefore, that overall early replicating chromosomes such as gene-rich chromosomes 1 and 19 are preferentially repaired when compared to late replicating chromosomes. All avenues seem to lead to the same conclusion: high gene density chromosomes are preferentially repaired in human cells.

The inactive X chromosome and heterochromatin: presenting a model to study heterogeneous repair at the chromosome level

The specific time interval during DNA-synthesis (S-phase) at which a given DNA sequence is being replicated is highly associated with the transcriptional activity of that sequence. Accordingly, expressed DNA loci undergo early replication whereas unexpressed loci replicate late. A clear example is the inactive X chromosomeand the non-coding satellite DNA forming the constitutive heterochromatin, which replicate late in S-phase (see Yeshaya et al., 1998, and references therein). In order to further visualize heterogeneous repair at the chromosome level we applied the above repair inhibition approach and interphase FISH with tandem probes to provide visual evidences for a low level of DNA repair in human constitutive heterochromatin, band 1q12. We speculated that the observed repair deficiency in 1q12 probably accounts for its fragility and involvement in cancer chromosomal aberrations (Surrallés et al., 1997b).

In our attempts to find out models to study heterogeneous repair at the chromosome level, we concluded that the inactive X chromosome had all the features a good model required. X chromosome inactivation results in silencing of transcriptional activity in all but one X-chromosome in somatic female cells, hence allowing dosage compensation of X-linked genes between males and females (Lyon, 1961). In previous studies comparing euchromatin and heterochromatin with respect to chromosome fragility and repair it was not possible to know whether it is the heterochromatic state or the heterogeneous genetic background which actually accounts for the observed differences between heterochromatin and euchromatin. However, the inactive X chromosome provides us with a unique opportunity to overcome this limitation as both X chromosomes (active and inactive) have the same genetic background and, therefore, the active X chromosome serves as internal control to study the role of transcriptional activity and chromatin conformation in DNA repair.

One of the prerequisites of using the inactive X chromosome as a model to study intragenomic heterogeneity in DNA repair is that X inactivation is not spread to autosomal material in the case of translocations X chromosome-autosome. Otherwise, selection against cells carrying translocations involving the inactive X chromosome would lead to biassed results. We therefore analyzed de novo induced translocations between the inactive X chromosome and autosomes to study the spreading of X-inactivation with respect to the position of the XIST gene This gene is known to control the spreading and initiation of inactivation during early embryonic development (Brown et al., 1991; Penny et al., 1996; Herzing et al., 1997; Lee and Jaenisch; 1997). We had to develop new methods to check for cis-limited spreading of X inactivation in X chromosome-autosome translocations. Thus, translocations involving any of the two X-chromosomes were detected by means of FISH with X-chromosome specific red painting probes. The activation status of the chromosomes involved in the translocation was determined by simultaneous immunocytogenetics with antibodies either against acetylated histone H4 as a cytogenetic marker of gene expression (Jeppesen and Turner, 1993; Surrallés et al., 1996) or against BrdU incorporated at late S-phase as a cytogenetic marker of the late-replicating inactive X chromosome (Willard and Latt, 1976). Xq13 band carrying the XIST gene was localized by computer-assisted generation of the DAPI banding pattern. Alternatively, the position in cis or trans of the XIST gene in the reciprocal products of the translocation was determined by simultaneous XIST gene specific FISH and computer enhancing. Our study in differentiated somatic cells provided visual demonstration that the X-inactivation was not spread to the translocated autosomes irrespective of the position of the XIST gene (Surrallés et al., 1998a), so the inactive X chromosome could be faithfully used as model to study heterogeneousDNA repair at the chromosome level.

Accordingly, we induced chromosome breakage in human lymphocytes with X-rays in the presence or absence of an inhibitor of double strand break repair, adenine 9--_d-arabinofuranoside (Ara-A). Our data surprisingly indicated that both chromosomes are equally radiosensitive. However, the inactive and highly condensed state of the inactive X chromosome enhanced the inhibitory effect of Ara-A upon the repair of X-ray-induced breaks. Thus, the observed fragility is the result of a balance between the actual number of breaks induced in each chromosome and their differential processing (Surrallés et al., 1998b). The next step will be to compare the extent of UV-radiation induced NER in the active and inactive X chromosome. As NER is preferentially directed to actively transcribed regions, we would expect a lower level of repair in the inactive X chromosome.

Does gene density modulate the persistence of chromosome damage?

Once I had managed to settle back into the Spanish society, I had in mind that, besides different radiosensitivity and repair, another factor that might modulate interchromosomal differences in the number of breaks detected after past exposure is a differential persistence of translocations involving different chromosomes. Low persistence of translocations might be related to cell lethality as a result of gene truncation or position effects. If this is true, then one would expect a lower persistence of translocations involving chromosomes with high gene density. To verify this hypothesis we are analysing the persistence of translocations involving chromosomes 1 and 19 (with high gene density) and 4 and 18 (with very low gene density). Translocations were induced by ionizing radiations in a wild-type lymphoblastoid cell line and samples were collected and harvested 1, 3, 7, 14, 28, 42 and 56 days after irradiation. After scoring 4000 metaphases per chromosome and time point (including untreated controls), chromosome aberrations involving chromosome 1 declined slightly faster than those involving chromosome 4, suggesting that aberrations involving the gene-rich chromosome 1 are less stable. Further analysis will allow us to state whether this tendency is biologically relevant and actually related to gene density.

Back in my home laboratory, I am really lucky my PhD students are continuing the job. I am trying to pass them all I have learnt from my mentors and teachers so the knowledge will not fade but, like wine, improve through the years. This is probably the tribute I can pay to all the people that have made it possible for me to get this Award and that have put their trust in me over the past 10 years, including my family and my wife. Last but not least. Thank you all.

Acknowledgements

This overview is the subject of the EEMS Young Scientist Award 1998 Lecture I gave during the 28th-EEMS Annual Meeting in Salzburg. I would like to thank the Award Committee for the prize and for giving me the opportunity to attend the EEMS-Meeting.

References

Brown, C.J., A. Ballabio, J.L. Rupert, R.G. Lafreniere, M. Grompe, R. Toniorenzi and H.F. Willard (1991) A gene from the region of the human X-inactivation centre is expressed exclusively from the inactive X-chromosome. Nature 349: 38-44.

Craig,J.M. and W.A. Bickmore (1994) The distribution of CpG islands in mammalian chromosomes. Nature Genet. 7: 376-382.

Degrassi, F. and C. Tanzarella (1988) Immunofluorescent staining of kinetochores in micronuclei: a new assay for the detection of aneuploidy. Mutation Res. 203: 339-345.

Fenech and Morley (1985) Measurement of micronuclei in human lymphocytes. Mutation Res. 149: 27-36.

Fenech, M. and S. Neville (1992) Conversion of excision repairable DNA lesions to micronuclei within one cell cycle in human lymphocytes. Environ. Mol. Mutagen. 19: 27-36.

Fenech, M., J. Rinaldi, and J. Surrallés (1994) The origin of micronuclei induced by cytosine arabinoside and its synergistic interaction with hydroxyurea in human lymphocytes. Mutagenesis 9: 273-277.

Friedberg,E.C. (1996) Relationships between DNA repair and transcription, Annu. Rev. Biochem. 65, 15-42.

Herzing, L.B.K., J.T. Romer, J.M. Horn, A. Ashworth (1997) Xist has properties of the X-chromosome inactivation centre. Nature 386: 272-275 (1997).

Jeppesen, P., and B.M. Turner (1993) The inactive X chromosome in female mammals is distinguished by a lack of histone H4 acetylation, a cytogenetic marker for gene expression. Cell 74: 281-289.

Kalle, W.H.J., A.-M. Hazekamp-van Dokkum, P.H.M. Lohman, A.T. Natarajan, A.A.van Zeeland, and L.H.F. Mullenders (1992) The use of streptavidin-coated magnetic beads and biotinylated antibodies to investigate induction and repair of DNA damage: analysis of repair patches in specific sequences of uv-irradiated human fibroblast. Anal. Biochem. 208: 228-236.

Lee, J.T. and R. Jaenisch (1997) Long-range cis effects of ectopic X-inactivation centres on a mouse autosome. Nature 386, 275-279.

Lyon, M.F. (1961) Gene action in the X-chromosome of the mouse (Mus musculus L.), Nature, 190, 372-373.

Penny, G.D., G.F. Kay, S.A. Sheardown, S. Rastan, N. Brockdorff (1996) Requirement for Xist in X-chromosome inactivation. Nature 379: 131-137.

Sitatam, A., G. Plitas, W. Wang and D. A. Scicchitano (1997) Funtional nucleotide excision repair is required for the preferential removal of N-ethylpurines from the transcribed strand of the dihydrofolate reductase gene in Chinese hamster ovary cells. Mol. Cell. Biol. 17: 564-570.

Surrallés J., N. Xamena, A. Creus, and R. Marcos R (1995) The suitability of the micronucleus assay in human lymphocytes as a new biomarker of excision repair. Mutat Res. 342: 43-59.

Surrallés, J., P. Jeppesen, H. Morrison, A.T. Natarajan (1996) Analysis of loss of inactive X chromosome in interphase cells. Am. J. Hum. Genet. 59: 1091-1096.

Surrallés, J. S. Sebastian, and A.T. Natarajan (1997a) Chromosomes with high gene density are preferentially repaired in human cells, Mutagenesis 12: 437-442.

Surrallés, J., F. Darroudi, and A.T. Natarajan (1997b) Evidence for low level of DNA repair in human chromosome 1 heterochromatin, Genes Chromosomes and Cancer 20: 173-184.

Surrallés, J. and A.T. Natarajan (1998a) Position effect of translocations involving the inactive X chromosome: physical linkage to XIC/XIST does not lead to long-range de novo inactivation in human somatic cells. Cytogenet. and Cell Genet. (in press)

Surrallés, J. and A.T. Natarajan (1998b) Radiosensitivity and Repair of the Inactive X chromosome. Insights from FISH and Immunocytogenetics. Mutation Res. (in press)

Venema, J., A. Van Hoffen, V. Karcagi, A.T. Natarajan, A.A. van Zeeland, amd L.H.F. Mullenders (1991) Xeroderma pigmentosum complementation group C cells remove pyrimidine dimers selectively from the transcribed strand of active genes. Mol. Cell. Biol. 11: 4128-4134.

Willard, F.H. and S.A. Latt (1976) Analysis of deoxyribonucleic acid replication in human X chromosomes by fluorescence microscopy. Am. J. Hum. Genet 28: 213-227.

Yeshaya, J., R. Shalgi, M. Shohat and L. Avivi (1998) Replication timing of the various FMR1 alleles detected by FISH: inferences regarding their transcriptional status. Human Genet. 102: 6-14.