Fen1 expression: a novel marker for cell proliferation

J. Pathol. 186: 319–324 (1998)

FEN1 EXPRESSION: A NOVEL MARKER FOR CELLPROLIFERATION

1*, . 2 . 2

1Department of Biochemistry, and 2Department of Molecular and Cellular Pathology, University of Dundee, Dundee, DD1 4HN,Scotland, U.K.

SUMMARY

The identification of antigens whose expression is associated with the cell cycle is a particularly attractive method with which to defineproliferative populations in histological and cytological preparations. A polyclonal antibody 3220 has been raised which recognizes thestructure-specific endonuclease Fen1 and can be used for a wide range of applications including western blotting, immunoprecipitationand immunohistochemical analysis. This antibody has been used to examine Fen1 levels by immunoblotting and its subcellularlocalization in cultured cells and tissue samples by immunostaining. Although the role Fen1 plays in DNA replication has been wellcharacterized, its function in DNA repair is not so clear. The possible roles of Fen1 in repair have been investigated by examining anychanges in level or localization of Fen1 in response to DNA damaging agents. We find that Fen1 is a nuclear antigen, that it is expressedby cycling cells, and that it co-localizes with PCNA and polymerase á during S phase. Fen1 expression is topologically regulated in vivoand is associated with proliferative populations. No change has been found in either patterns or levels of Fen1 expression induced byDNA damaging agents, either in vivo or in vitro.

This anti-Fen1 antiserum is well suited to the analysis of proliferation in histological material, since (1) the proportion of labelled cellsequals the experimentally determined growth fraction in an experimental xenograft system and (2) unlike markers such as PCNA, Fen1is not induced by DNA damage. ? 1998 John Wiley & Sons, Ltd.

KEY WORDS—Fen1; antibody; cell cycle; DNA damage; DNA replication

*Correspondence to: E. Warbrick, Cancer Research CampaignLaboratories, Department of Biochemistry, Medical Sciences Institute,University of Dundee, Dundee, DD1 4HN, U.K. E-mail:[email protected]

INTRODUCTION

Metazoan organisms must co-ordinately regulatethree critical processes: the specification of cell type bythe regulated expression of differentiation-related genes;the number of differentiated cells in any given micro-anatomical structure; and the spatial arrangement ofthose cells. Central to this process is the regulationof cell number by the balance of cell loss (by apoptosisand exfoliation) and the production of new cells byproliferative lineages.1 In the analysis of normal andpathological tissues, the careful quantitation of prolifer-ation depends upon either the use of exogenous markerssuch as Budr or tritiated thymidine, or the measurementof endogenous molecules whose expression is associatedwith aspects of cell proliferation.2 The latter categoryhas particular advantages in the context of the examin-ation of human material, since no exogenous reagents arerequired to be administered. In the past, molecules suchas PCNA and Ki67 have been employed; the former hasdisadvantages, since its expression is regulated by DNAdamage and there are well-documented methodologicaldrawbacks.3 The latter molecule has many advantages,though at present the function of Ki67 is unknown.4Novel reagents that recognize biochemically well-characterized molecules whose expression is associatedwith proliferation and which are not expressed in othersituations are thus desirable for use in experimental andclinical work.

CCC 0022–3417/98/110319–06 $17.50? 1998 John Wiley & Sons, Ltd.

We describe here the analysis of expression patterns ofthe structure-specific endonuclease Fen1, (also known asrad2hs, DNAse IV and MF1).5–9 Fen1 was first identifiedbiochemically (as DNAse IV) as a structure-specificendnuclease showing high specificity of binding andactivity towards 5* flap structures, which comprisedouble-stranded DNA with a displaced 5* strand.10 Itsactivity has subsequently been extensively characterized:it is independent of flap length and is highly specific, asother branched structures such as 3* flap structures,Holliday junctions etc. are not cleaved.11 Fen1 also has5* to 3* exonucleolytic activity which is maximal whenacting at a nick, and has lower efficiency at a gap orrecessed 5* end on double stranded DNA.5,12

Subsequently, it became apparent that this protein isidentical to the factor MF1, identified as essential forin vitro DNA replication.7 Fen1 is essential for SV40replication in vitro and has been shown to be required,in combination with RNAse H, for Okazaki fragmentprocessing.7,13 Fen1 homologues have now been ident-ified in human mouse and the yeasts S. cerevisiae andS. pombe.5,6,14–16 Fen1 binds to PCNA, which results inthe exonucleolytic and endonucleolytic activities of Fen1being stimulated 10–50-fold in vitro.17–19 It seems likelythat the interaction with PCNA may stabilize Fen1 atthe branch point of the replication fork. p21 appears tobe involved in the regulation of the Fen1–PCNA inter-action: Fen1 and p21 PCNA-interacting domains showconservation of essential residues, both interact withthe same regions of PCNA, and p21 can disrupt theinteraction between Fen1 and PCNA in vitro.19,20

The role of Fen1 in DNA repair is not so clear:reconstitution of human base excision repair (BER) in

Received 3 February 1998Revised 24 April 1998

Accepted 29 June 1998

320 E. WARBRICK ET AL.

vitro shows that Fen1 is essential for ‘long patch’ BER,which also involves PCNA.21 Disruption of the Fen1homologue in S. pombe (Rad2) results in UV sensitivity,though this is due to the involvement of Rad2 in the UVdamage endonuclease (UVDE) repair pathway, ratherthan in nucleotide excision repair (NER).22 Disruptionof the S. cerevisiae Fen1 homologue (Rad27/Rth1)results in a temperature-sensitive phenotype with aterminal phenotype characteristic of a defect in DNAreplication. However, at the permissive temperature,the strain is viable and shows increased sensitivityto alkylating agents, also pointing to a role for Fen1in base excision repair.15,16 Fen1 shares significantsequence homology with the XPG family of endo-nucleases, which are essential for NER, although Fen1 isnot involved in this form of repair process.23

In order to examine the role of Fen1 in replication andrepair in mammalian cells and in vivo in tissues, we haveraised a Fen1 polyclonal antibody 3220 which can beused for a wide range of applications including westernblotting, immunoprecipitation and immunohistochemi-cal analysis. We have used this antibody to examine thesubcellular localization of Fen1 in cultured cells andtissue samples by immunostaining. We have also inves-tigated the possible roles of Fen1 in DNA repair byexamining any changes in level or localization of Fen1 inresponse to DNA damaging agents.

MATERIALS AND METHODS

described.
Expression of human Fen1 in E. coli and generation ofthe polyclonal serum 3220
The Fen1 full-length open reading frame was clonedinto pET23d (Novagen) as an NcoI–BglII fragment froma Fen1 clone isolated in the vector pACT in a two-hybrid screen which was carried out with human PCNAas the bait protein.19 Competent E. coli BL21 (DE3)cells were transformed with this plasmid and induced toexpress Fen1 by IPTG treatment. The majority of Fen1expressed in this way was shown to be insoluble andinclusion bodies were isolated from transformed celllysates. Fen1 protein was partially purified by prepara-tive SDS–PAGE and used to immunize a rabbit bysubcutaneous injection in Freund’s complete adjuvant,followed by monthly boosts with the protein in Freund’sincomplete adjuvant.

Western blotting

Proteins were separated on 12·5 per cent SDS–PAGEthen electrophoretically transferred to nitrocellulose.Blots were blocked, incubated either with a 1 in 1000dilution of 3220 in two per cent non-fat milk/PBS orwith PC10 hybridoma supernatant for one hour. Afterwashing, blots were incubated with secondary HRP-conjugated IgG serum at 1:500 for one hour. Boundantibody was visualized using the ECL system accordingto the manufacturer’s instructions (Amersham plc.)

Immunohistochemistry

Immunostaining was performed as previouslydescribed.24 For histological material, 4 ìm sections of

? 1998 John Wiley & Sons, Ltd.

formalin-fixed and wax-embedded human and rodenttissues were cut, mounted on silane coated glass slidesand air dried at 37)C. Sections were dewaxed andendogenous peroxidase activity blocked by immersion inmethanol with 0·5 per cent hydrogen peroxide. Micro-wave antigen retrieval was performed as previouslydescribed.24 Sections were rinsed in PBS and immuno-staining was performed with 3220 serum at a dilution of1 in 100 overnight at 4)C, after a blocking step withnormal swine serum. After washing in PBS, the primaryantibody was detected by one-hour incubation withperoxidase-labelled swine anti-rabbit antibodies. Afterfurther washes with PBS, this was visualized with thechromogen diaminobenzidine/hydrogen peroxide fol-lowed by a haematoxylin counterstain. For culturedcells, acetone methanol fixed and air-dried cells grownon glass slides or cover slips were incubated with 3220antiserum at a dilution of 1 in 100 for one hour at roomtemperature and detected with fluorescein-labelled swineanti-rabbit antibodies.

For double labelling experiments, PCNA or polym-erase á were detected using monoclonal antibodies,followed by rhodamine-conjugated anti-mouse anti-bodies. Preparations were viewed in a BioRad 600 laserscanning confocal microscope with Zeiss optics.

Histological sections of xenografts of the LoVo cellline grown in nude mice in which the tumour growthfraction had been experimentally determined25 werestained as described above for Fen1 expression and thenumber of labelled nuclei assessed by quantitatingstained tumour cell nuclei in 1000 cells as previously

25

RESULTS

Antiserum 3220 recognizes Fen1 protein

A rabbit antiserum was raised (see Materials andMethods) against full-length human Fen1 protein. Theresultant serum (3220) recognizes recombinant Fen1protein expressed in E. coli and a band corresponding toapproximately 48 kD in lysates of mammalian cells (seebelow). 3220 was tested by western blotting againstprotein samples form a variety of organisms and wasshown to recognize a band of similar molecular weightto that seen in human protein samples in hamster, rat,mouse and Xenopus samples (data not shown).

Fen1 protein is a nuclear antigen that is expressed bycycling cells

In order to analyse the subcellular localization ofFen1, asynchronously-growing cultured cells (HeLa andNIH 3T3) were immunostained with 3220. Very littlecytoplasmic staining is observed, the vast majority ofFen1 immunoreactivity being localized in the nucleus(Fig. 1), which is in agreement with the role of Fen1 inDNA replication. Within the cell population there isvariation in the pattern and intensity of Fen1 staining:occasional cells contain nuclei which are unstained orweakly stained and in a minority population the staining

J. Pathol. 186: 319–324 (1998)

321FEN1 EXPRESSION

has a dot-like or granular appearance. This pattern ofstaining is suggestive of that seen when cells are stainedfor markers such as PCNA, which has been shown tolocalize to sites of DNA replication termed replicationfoci.

In order to examine Fen1 protein levels as cellsprogress from quiescence into the cell cycle, Fen1expression in lysates of serum-starved 3T3 cells and cellsat different time points after refeeding were examined bywestern blotting (Fig. 2a). The level of Fen1 expressionis very low in non-cycling cells and is induced as the cellsenter the cell cycle. Peripheral human blood lym-phocytes were also used to examine Fen1 levels follow-ing stimulation of the cell cycle. In freshly isolatedlymphocytes, Fen1 expression is barely detectable. How-ever, following stimulation with the mitogen phyto-haemagglutinin there is a dramatic induction of Fen1expression concomitant with the observed mitogenicresponse (Fig. 2b). This increase in Fen1 protein levelparallels the induction of PCNA (Fig. 2c).

Fen1 co-localizes with PCNA and polymerase á duringS phase

During S phase, sites of DNA replication within thenucleus are associated with dense structures, which havebeen termed DNA replication factories or foci. Thesechange in number and distribution as the cells progressthrough S phase and have various proteins associatedwith them, including DNA polymerase á, PCNA, DNAligase 1 and RPA. These proteins show a characteristicpunctate pattern of localization in S phase cells. Theobservation of a minority population of cells with agranular or dot-like distribution of Fen1 expressionprompted us to examine the possible association of Fen1with replication foci. We used double immunofluor-escence to examine whether Fen1 co-localizes with eitherpolymerase á or PCNA in S phase HeLa cells. We found


that cells showing a punctate pattern of staining witheither polymerase á or PCNA also showed a similarpattern of staining with the anti-Fen1 antiserum 3220. Acloser examination of the patterns of staining show thatthey are coincident, which strongly suggests that Fen1is predominantly localized to replication foci duringS phase (Fig. 3).

Fig. 1—Fen1 expression in asynchronously growing NIH 3T3 cells.Cells were immunostained with 3220 anti-Fen1 antiserum and Fen1immunoreactivity was seen on only occasional nuclei, typically with apunctate pattern

Fig. 2—Fen1 is expressed in proliferating cells in vitro. (a) Fen1expression in lysates of NIH 3T3 cells was examined by westernblotting following serum starvation and refeeding. A single band of48 kD corresponding to the predicted molecular weight of Fen1 wasdetected. The level of Fen1 expression is low in non-cycling cells (lane1, t=0 hours) and is induced as the cells enter the cell cycle (lane 2,t=15 hours; lane 3, t=24 hours; lane 4, t=41 hours). (b) Fen1expression is induced by PHA in peripheral human blood lym-phocytes. Fen1 expression by western blotting (panel A; with pre-immune serum in panel B) is barely detectable in peripheral humanblood lymphocytes (lane 1). After PHA stimulation there is a dramaticinduction of expression (lane 2, t=24 hours; lane 3, t=30 hours; lane 4,t=42 hours; lane 5, t=48 hours; lane 6, t=72 hours). Panel C showsthe concomitant induction of PCNA in this experiment by westernblotting with the anti-PCNA antibody PC10

Fen1 protein expression is topologically regulation invivo and is associated with proliferative populations

We have shown that levels of Fen1 expression are lowin non-cycling cells in culture, and are induced as cellsenter the cell cycle. In order to investigate if this in vitropattern is reflected in vivo, the distribution of Fen1expression was examined in histological preparationsfrom human and rodent tissues. Using immunoperoxi-dase methods, we found that patterns of Fen1 expres-sion correlate with the proliferative state of a wide rangeof tissues (Fig. 4). The normal adult brain, smooth andskeletal muscle, myocardium and peripheral nerve con-tain no proliferating cells. Fen1 expression was notdetectable in these tissues, but was identifiable in thegerminal centres of lymphoid tissue and in the epi-thelium of the skin, stomach, small intestine, colon,bladder and proliferative phase endometrium. In thesetissues, where there is spatially-regulated distribution ofproliferation, Fen1 expression is seen in the nuclei of

J. Pathol. 186: 319–324 (1998)


cells within the morphologically-defined proliferativezones. This is similar to that observed for PCNA26 andthe inverse of the distribution of p21Cip1 expression.27,28

During murine development there is very widespreadexpression of Fen1 protein, correlating closely with the


extensive cell proliferation seen in organogenesis andgrowth. In the xenografts, the fraction of Fen1 stainedcells was the same as the experimentally determinedgrowth fraction, indicating the ultility of Fen1 stainingin assessing the proliferation of cell populations inhistological material. These results show that Fen1expression as determined by 3220 immunohistochemis-try is a useful marker for proliferation not only in vitro,but in a wide range of mouse and human tissues.

Fig. 3—Fen1 co-localizes with PCNA and polymerase á to replicationfoci. Asynchronously growing HeLa cells were immunostained withthe anti-Fen1 serum 3220, and with either anti-PCNA antibodies oranti-polymerase á antibodies. Cells showing a granular pattern ofanti-Fen1 staining (red channel, (B) and (D)) also show a granularpattern of PCNA (A) and polymerase á (C) staining (green channel),and these two patterns show co-localization

Fig. 4—Fen1 protein expression is associated with proliferative populations. Fen1 immunoreactivity isseen in the basal cells of adult mouse epidermis (panel A). In the murine small intestinal epithelium(panel B) Fen1 immunoreactivity is seen in the crypts but not in the lamina propria stromal cells or inthe villus cells. In contrast, in the non-proliferative mouse liver (panel C) or myocardium (panel D),Fen1 immunoreactivity is not detected

Fen1 protein expression is not induced by DNAdamaging agents in vitro or in vivo

The role of Fen1 in DNA replication has previouslybeen well defined by in vitro studies and, consistent withthis, we have shown that Fen1 is associated with repli-cation foci in cycling cells. The role of Fen1 in repair,however, has not been so well defined. In order toinvestigate how Fen1 expression is regulated in responseto DNA damage, we have analysed changes in Fen1level and/or localization in response to a range of DNAdamaging agents.

Human skin irradiated with 1·5 MED of solar simu-lated light shows a dramatic induction of p53 andPCNA protein.29 The expression of Fen1 and p21Cip1

was examined in formalin-fixed wax-embedded materialfrom this experiment. While p21Cip1 expression wasinduced, there was no change in Fen1 immunoreactivityat the times examined (1, 2, 4, 8, 24, 48, 96, 156, and 312hours). Specifically, Fen1 staining was restricted to thebasal and immediate parabasal keratinocytes and therewas no discernible increase of expression or of thenumber of cells staining positively (data not shown).NIH 3T3 cells exposed to 20 Jm"2 of UV showed a

J. Pathol. 186: 319–324 (1998)

323FEN1 EXPRESSION

marked induction of p53, p21Cip1 and PCNA. Over aperiod of 24 hours there was no induction of Fen1expression in parallel experiments (Fig. 5).

In other experiments it has been shown that the p53pathway is induced by gamma irradiation in some, butnot all, tissues in mouse.24,30,31 We found no detectablealteration in either level or localization of Fen1 in arange of different tissue types (spleen, liver, myo-cardium, brain and gut) examined six hours after 5 Gyof gamma irradiation (data not shown).

DISCUSSION

We have generated and characterized an antiserumraised against full-length human Fen1, and shown that itrecognizes Fen1 protein in human and rodent cells. Fen1is a structure-specific endonuclease which binds toPCNA and which shows high specificity of binding andactivity towards 5* flap structures. It has been shown tobe essential for in vitro SV40 replication and, unlikePCNA, is not required for nucleotide excision repairreconstituted in vitro.23 However, the phenotype ofmutants in homologues of Fen1 in both budding andfission yeasts suggests that Fen1 is involved in BER andUVDE repair pathways.6,15,16,22

We have examined patterns of Fen1 subcellular locali-zation by immunostaining and have also examined anychanges in the level or localization of Fen1 in responseto DNA damaging agents. We show that Fen1 is anuclear protein and that it is associated with PCNA andpolymerase á in replication foci. This is entirely con-sistent with its proposed role as a protein involved inDNA replication. Furthermore, we show that Fen1 isexpressed at very low levels in quiescent cells and thatlevels of expression rise as cells enter the cell cycle. Fen1protein expression is topologically regulated in vivoand is associated with proliferative populations. Thesepatterns of expression are similar to those of PCNA,which acts as a processivity factor for polymerase ä andwhich binds Fen1. Changes of levels or patterns of Fen1expression in response to DNA damaging agents wereexamined: there was no change either in protein levels orin tissue sections in response to UV treatment. This is incontrast to the induction seen when p53, p21Cip1 orPCNA levels are examined. In addition, no detectable


induction of Fen1 occurred in tissues following 5 Gy ofgamma irradiation.

The ability to define proliferative populations inhistological and cytological preparations with confi-dence is of great value33 and can be achieved by a varietyof different approaches. The identification of antigenswhose expression is cell-cycle associated is a particularlyattractive method and a number of antigens have beenproposed as good markers of proliferation. Many havedisadvantages: for example, PCNA expression isupregulated by DNA damage,32 levels may be high innon-proliferating cells2,34 and a variety of technicalcaveats exist.3 Ki67 expression is tightly linked to pro-liferation and can, with the help of antigen retrievalsteps, now be detected in fixed material. However, Ki67is not evolutionarily well conserved and thus the avail-able reagents cannot be used easily in animal studies,though an antibody to rat Ki67 has recently beencharacterized.35 In vivo labelling with thymidineanalogues is not always possible. Consequently, theavailability of antibodies to new cell-cycle associatedantigens that are evolutionarily conserved and whoseexpression is not induced by DNA damage is highlydesirable. As shown in this report, our antiserum toFen1 fulfils these criteria. Specifically, (1) it identifies anuclear antigen in the proliferative compartment oftissues known from earlier studies to be proliferating;36

(2) expression is not seen in quiescent populations;(3) expression is not induced by DNA damaging agentsin vivo or in vitro; (4) the proportion of labelled cellsequals the experimentally-determined growth fraction inan experimental xenograft system;25 and the antiserumrecognizes Fen1 from a diverse range of species. In thefuture, monoclonal antibodies may also prove to be ofgreat utility.

ACKNOWLEDGEMENTS

This work was supported by the AICR, the EuropeanCommunity, and the Cancer Research Campaign. Wethank Professor Rob Hume for assistance with photo-microscopy, Dr S. J. Campbell for immunohistology,and Dr Alan Prescott for help with confocal microscopy.

Fig. 5—Fen1 expression is not induced by UV treatment of NIH 3T3cells in vitro. Fen1 expression was examined by western blotting insynchronously-growing NIH 3T3 cells exposed to 20 Jm"2 of UV. Noinduction of Fen1 was detected (lane 1, t=0 hours; lane 2, t=1 hour;lane 3, t=3 hours; lane 4, t=6 hours; lane 5, t=24 hours). This is incontrast to the marked induction of p53, p21Cip1 and PCNA (data notshown). The single band of 48 kD corresponds to the predictedmolecular weight of Fen1

REFERENCES

1. Hall PA, Coates PJ, Ansari B, Hopwood D. Regulation of cell number inthe mammalian intestine: the importance of apoptosis. Journal of CellScience 1994; 107: 3569–3577.

2. Hall PA, Woods A. Immunohistochemical markers of cell proliferation.Achievements, problems and prospects. Cell and Tissue Kinetics 1990; 23:549

3. McCormick D, Hall PA. The complexities of proliferating cell nuclearantigen. Histopathology 1992; 21: 591–594.

4. MacCallum DE, Hall PA. Biochemical characterisation of pKi67 withthe identification of a mitotic specific form associated with hyper-phosphorylation and altered DNA binding. Submitted.

5. Harrington JJ, Lieber MR. The characterisation of a mammalian DNAstructure-specific endonuclease. EMBO J 1994; 13: 1235–1246.

6. Murray JM, Tarassoli M, Al-Harithy R, et al. Structural and functionalconservation of the human homologue of the Schizosaccharomyces pomberad2 gene, which is required for chromosome segregation and recovery fromDNA damage. Mol Cell Biol 1994; 14: 4878–4888.

7. Waga S, Bauer G, Stillman B. Reconstitution of complete SV40 DNAreplication with purified replication factors. J Biol Chem 1994; 269:10 923–10 934.

J. Pathol. 186: 319–324 (1998)


8. Robins P, Pappin D, Wood RD, Lindahl T. Structural and functionalhomology between mammalian DNAase IV and the 5* nuclease domain ofE. coli polymerase I. J Biol Chem 1994; 269: 28 535–28 538.

9. Lieber MR. The FEN-1 family of structure-specific nucleases in eukaryoticDNA replication, recombination and repair. BioEssays 1997; 19: 233–240.

10. Lindahl T, Gally JA, Edelman GM. Deoxyribonuclease IV: a newexonuclease from mammalian tissues. Proc Natl Acad Sci USA 1969; 62:597–603.

11. Harrington JJ, Lieber MR. DNA structural elements required for FEN-1binding. J Biol Chem 1995; 270: 4503–4508.

12. Murante R, Huang L, Turchi J, Bambara R. The calf 5* to 3* exonuclease isalso an endonuclease with both activities dependent on primers annealedupstream of the point of cleavage. J Biol Chem 1994; 269: 1191–1196.

13. Siegal G, Turchi JJ, Myers TW, Bambara RA. A 5* to 3* exonucleasefunctionally interacts with calf polymerase å. Proc Natl Acad Sci USA 1992;89: 9377–9381.

14. Hiraoka LR. Sequence of human FEN-1, a structure specific endonuclease,and chromosomal localization of the gene FEN1 in mouse and human.Genomics 1995; 25: 220–225.

15. Reagan MS, Pittenger C, Siede W, Friedberg EC. Characterisation of amutant strain of Saccharomyces cerevisiae with a deletion of the RAD27gene, a structural homolog of the RAD2 nucleotide excision repair gene.J Bact 1995; 177: 364–371.

16. Sommers CH, Miller EJ, Dujon B, Prakash S, Prakash L. Conditionallethality of null mutations in RTH1 that encodes the yeast counterpart of amammalian 5*- to 3*-exonuclease required for lagging strand DNA synthesisin reconstituted systems. J Biol Chem 1995; 9: 4193–4196.

17. Li X, Li J, Harrington J, Lieber MR, Burgers PMJ. Lagging strand DNAsynthesis at the eukaryotic replication fork involves binding and stimulationof FEN-1 by proliferating cell nuclear antigen. J Biol Chem 1995; 270:22 109–22 112.

18. Wu X, Li J, Li X, Hsieh C-L, Burgers PMJ, Lieber MR. Processing ofbranched intermediates by a complex of human FEN-1 and PCNA. NuclAcids Res 1996; 24: 2036–2043.

19. Warbrick E, Lane DP, Glover DM, Cox LS. Homologous regions of Fen1and p21Cip1 compete for binding to the same site on PCNA: a potentialmechanism to co-ordinate DNA replication and repair. Oncogene 1997; 14:2313–2321.

20. Chen J, Chen S, Saha P, Dutta A. p21Cip/Waf1 disrupts the recruitment ofhuman Fen1 by proliferating-cell nuclear antigen into the DNA replicationcomplex. Proc Natl Acad Sci USA 1996; 93: 11 597–11 602.

21. Klungland A, Lindahl T. Second pathway for completion of human DNAbase excision-repair: reconstitution with purified proteins and requirementfor DNase IV (FEN1). EMBO J 1997; 16: 3341–3348.

22. Yonemasu R, McReady SJ, Murray JM, et al. Characterisation of thealternative excision repair pathway of UV damaged DNA in Schizosac-charomyces pombe. Nucl Acids Res 1997; 25: 1553–1558.


23. Aboussekhra A, Biggerstaff M, Shivji MKK, et al. Mammalian DNAnucleotide excision repair reconstituted with purified protein components.Cell 1995; 80: 859–868.

24. MacCallum DE, Hupp TR, Midgely CA, et al. The p53 response to ionisingradiation in adult and developing murine tissues. Oncogene 1996; 13:2575–2587.

25. Scott RJ, Hall PA, Haldane J, et al. A comparison of immunohistochemicalmarkers of cell proliferation with experimentally determined growth frac-tion. Journal of Pathology 1991; 165: 173–178.

26. Hall PA, Levison DA, Woods AL, et al. Proliferating cell nuclear antigen(PCNA) immunolocalisation in paraffin sections. An index of cell prolifer-ation with evidence of deregulated expression in some neoplasms. Journal ofPathology 1990; 162: 285–294.

27. El-Deiry WS, Tokino T, Waldman T, et al. Topological control ofp21WAF1/CIP1 expression in normal and neoplastic tissues. Cancer Res 1995;55: 2910–2919.

28. Fredersdorf S, Milne A, Hall PA, Lu X. Immunochemical analysis ofp21waf1/Cip1 expression in normal human tissues using a panel of novelmonoclonal antibodies. American Journal of Pathology 1996; 148: 825–835.

29. Hall PA, Mckee PH, Du P Menage H, Dover R, Lane DP. High levels ofp53 protein in UV-irradiated normal human skin. Oncogene 1993; 8:203–207.

30. Merritt A, Potten CS, Kemp CJ, et al. The role of p53 in spontaneous andradiation induced apoptosis in the gastrointestinal tract of normal and p53deficient mice. Cancer Res 1994; 54: 614–617.

31. Midgley CA, Owens B, Briscoe CV, Thomas DB, Lane DP, Hall PA.Coupling between gamma irradiation, p53 induction and the apoptoticresponse depends upon cell type in vivo. Journal of Cell Science 1995; 108:1843–1841.

32. Hall PA, Mckee PH, Du P Menage H, Dover R, Lane DP. High levels ofp53 protein in UV-irradiated normal human skin. Oncogene 1993; 8:203–208.

33. Hall PA, Levison DA. Assessment of cellular proliferation in histologicalmaterial. Journal of Clinical Pathology 1990; 43: 184–192.

34. Hall PA, Coates PF, Goodlad R, Hart IR, Lane DP. Proliferating cellnuclear antigen expression in non-cycling cells may be induced by growthfactors in vivo. British Journal of Cancer 1994; 70: 244–247.

35. Gerlach C, Golding M, Larue L, Alison MR, Gerdes J. Ki-67 immuno-expression is a robust marker of proliferative cells in the rat. Lab Invest1997; 77: 697–698.

36. Ansari B, Hall PA. The kinetic organisation of tissues. In: Hall PA, LevisonDA, Wright NA, eds. Assessment of Cell Proliferation in Clinical Practice.Berlin, Springer Verlag, 1992; 45–62.

J. Pathol. 186: 319–324 (1998)

Documents

Fen1 expression: a novel marker for cell proliferation