HUMANMUTATION 27(11),1065^1071,2006

RESEARCH ARTICLE

Deletion Hotspot in the Argininosuccinate LyaseGene: Association With Topoisomerase II and DNAPolymerase a Sites

John Christodoulou,1,2� Hugh J. Craig,1,2 David C. Walker,1 Linda S. Weaving,1

Christopher E. Pearson,1,2 and Roderick R. McInnes1,2�

1Program in Genetics and Genomic Biology, Research Institute, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada;2Department of Pediatrics and Molecular and Medical Genetics, University of Toronto, Toronto, Ontario, Canada

Communicated by David Cooper

Molecular analysis of argininosuccinate lyase (ASAL) deficiency has led to the identification of a deletionhotspot in the ASL gene. Six individuals with ASAL deficiency had alleles that led to a complete absence of exon13 from the ASL mRNA; each had a partial deletion of exon 13 in the genomic DNA. In all six patients, thedeletions begin 18 bp upstream of the 30 end of exon 13. In four cases, the deletions were 13 bp in length, andended within exon 13, whereas in two other patients the deletions were 25 bp and extended into intron 13. Thesequence at which these deletions begin overlaps both a putative topoisomerase II recognition site and a DNApolymerase a mutation/frameshift site. Moreover, the topoisomerase II cut site is situated precisely at thebeginning of the deletions, which are flanked by small (2- and 3-bp) direct repeats. We note that a similarconcurrence of these two putative enzyme sites can be found in a number of other deletion sites in the humangenome, most notably the DF508 deletion in the CFTR gene. These findings suggest that the joint presence ofthese two enzyme sites represents a DNA sequence context that may favor the occurrence of small deletions.Hum Mutat 27(11), 1065–1071, 2006. rr 2006 Wiley-Liss, Inc.

KEY WORDS: CFTR; ESE; ASL; argininosuccinate lyase; urea cycle; deletion hotspot; slipped mispairing

INTRODUCTION

Spontaneous mutations, including single-base substitutions,deletions, and gene rearrangements, are not uniformly distributedover the genome [Cooper and Krawczak, 1990; Krawczak andCooper, 1991], which suggests that the local DNA environmentmay sometimes predispose to the occurrence of mutation [Benzer,1961; Krawczak et al., 1998]. Over 7,000 small deletions (less than20 nucleotides) have been identified in up to 1700 human genes(Human Gene Mutation Database; www.hgmd.org) [Stensonet al., 2003] and are often flanked by short regions of sequencehomology [Antonarakis et al., 2000]. A number of geneshave been identified in which there were independentsmall deletions in apparently unrelated individuals involving thesame region of the gene (deletion hotspots) [Cooper et al., 1998;Huff et al., 1995; Jouanguy et al., 1999; Krawczak and Cooper,1991; Lebo et al., 2001]. Short direct repeats were almostinvariably found in the vicinity of the deletion hotspots, but therepeats did not necessarily flank the deletions. In addition, asequence similar to the (TGGNGT) sequence for a polymerase-a(pol-a) mutation/frameshift site [Kunkel, 1985] was identified atmost of these hotspots. Pol-a has also been implicated in thegeneration of spontaneous mutations [Fry and Loeb, 1992]. Otherpossible deletion hotspots have also been recognized at theargininosuccinate synthetase locus [Jackson et al., 1989]. In thatstudy, 5 out of 48 mutant alleles were found to have one of twodeletions in genomic DNA, but the breakpoints were notestablished.

DNA replication requires the concerted action of a number ofdifferent enzymes [Frouin et al., 2003]. DNA pol-a is involved inreplication of the lagging strand at the replication fork, whiletopoisomerase II acts as a swivel to release the torsional stresscreated during replication. Topoisomerase II alters the topologicalstate of nucleic acids by passing an intact helix of DNA through adouble-stranded break that it generates in a separate DNA helix.Topoisomerase II has been found to mediate illegitimaterecombination in vitro [Bae et al., 1991, 1988; Han et al., 1993].

Argininosuccinate lyase deficiency (ASAL; E.C. 4.3.2.1; MIM]207900) is an autosomal recessive defect of ureagenesis, andaffected individuals are prone to hyperammonemia associated with

Published online 29 August 2006 inWiley InterScience (www.inters-cience.wiley.com).

DOI10.1002/humu.20352

The Supplementary Material referred to in this article can beaccessed at http://www.interscience.wiley.com/jpages/1059-7794/suppmat.

Received 1 March 2005; accepted revised manuscript 5March 2006.

Current a⁄liation for John Christodoulou: Discipline of Paediatricsand Child Health, University of Sydney, NSW, Australia.

�Correspondence to: John Christodoulou, Western SydneyGenetics Program, Children’s Hospital at Westmead, Locked Bag4001,Westmead, NSW, 2145, Australia. E-mail: johnc@chw.edu.au;Roderick R. McInnes, Program in Genetics, Research Institute,Hospital for Sick Children, 555 University Ave., Toronto, Ontario,CanadaM5G1X8. E-mail: mcinnes@sickkids.ca

rr 2006 WILEY-LISS, INC.

an accumulation of argininosuccinic acid and relative argininedepletion. It is notable for its phenotypic heterogeneity, rangingfrom severe hyperammonemic encephalopathy in the newbornperiod, through to intermittent milder though still damagingelevations of ammonium during periods of catabolic stressor excessive protein intake [Brusilow and Horwich, 2001].Previous studies have demonstrated the presence of at least 12complementation subgroups in patients with defects at the ASLlocus [McInnes et al., 1984; Walker et al., 1997]. To furtherinvestigate the clinical and genetic heterogeneity in ASALdeficiency, we performed mutation analysis in a cohort of 28ASAL-deficient fibroblast cell strains. We describe here a newdeletion hotspot in the ASL gene (MIM] 608310). We postulatethat topoisomerase II and pol-a may be directly implicated in thegeneration of mutations at this location.

MATERIALSANDMETHODSCell Strains

Cultured skin fibroblasts or lymphoblasts were obtained from 28unrelated individuals with ASAL deficiency, in accordance withguidelines of institutional review boards. These individualsincluded representatives of each of the three clinical phenotypes

[McInnes et al., 1984]. Fibroblasts were cultured in a-minimalessential medium without antibiotics, and supplemented with 10%fetal bovine serum [Walker et al., 1990]. Lymphoblasts werecultured as above, but supplemented with from 15 to 20% fetalbovine serum.

Reverse Transcription and PCR Ampli¢cation

Messenger RNA and genomic DNA were extracted as previouslydescribed [Walker et al., 1990]. The mRNA was reverse-transcribed [Walker et al., 1990], and cDNA or genomic DNAwas amplified using a modification of the ‘‘hot start’’ PCR method[Chou et al., 1992]. The reaction mixture (total volume 100ml)contained 62.5 mM of each dNTP, 0.8mM of each primer, 2.4 mMbovine serum albumin, PCR buffer consisting of 10 mM Tris-HClpH 8.3, 50 mM KCl, and 1.5 mM MgCl2, and DNA template. TheDNA was denatured by heating the reaction tubes in a water bathat 951C for 3 minutes. The tubes were then transferred to a heatingblock set at 801C, 5 units of Perkin Elmer (Wellesley, MA;www.perkinelmer.com) Amplitaq added, and then overlaid with70ml of mineral oil. PCR was performed using the Perkin Elmer-Cetus Thermal Cycler 480 under the following conditions:denaturation step 981C for 30 seconds, annealing/elongation step681C for 2 minutes, run for 35 cycles. For PCR of the 30 third of the

FIGURE 1. A deletion of exon13 in ¢veASL-de¢ciency alleles. A: PCR-ampli¢edDNA from the 30 third of theASL cDNA.The smallerfragments of strains 647,926,1006,1040, and1190 lackexon13 and are 84 bp shorter than the control 30 cDNA (theupper band). B:Sequencing of the PCR-ampli¢edASL cDNA from a control ¢broblast strain (left panel) and one allele of strain 1006 (right panel),demonstrating the deletion of exon13 in strain1006.

1066 HUMANMUTATION 27(11),1065^1071,2006

Human Mutation DOI 10.1002/humu

ASL cDNA, the primers (50 - 30) used were ASL cDNA (gi68303544) positions 983–1009 and positions 1495–1469. For PCRof genomic DNA encompassing exon 13, the same PCR methodwas used, except that a concentration of 1.6mM of previouslydescribed intronic primers [Barbosa et al., 1991] was used.

DNA Puri¢cation and Sequencing

PCR products amplified from cDNA or genomic DNA[Sambrook et al., 1989] were purified from a 5% nondenaturingpolyacrylamide gel except that the elution buffer was 10 mM Tris-EDTA, pH 8.0. Double-stranded direct sequencing was performedby the Sanger dideoxy chain termination technique using eitherthe Sequenase kit (United States Biochemical, Piscataway, NJ;www.amersham.com) or the Circumvent cycle-sequencing kit(New England Biolabs, Ipswich, MA; www.neb.com/nebecomm/default.asp).

HaplotypeAnalysis

To exclude identity by descent, genetic markers from theUniversity of California, Santa Cruz (UCSC) Genome Browser(http://genome.ucsc.edu/cgi-bin/hgTracks?position 5 chr7:64984963-65002090&hgsid 5 67375949&knownGene 5 pack&hgFind.mat-ches 5 ,) and the International HapMap Project (www.hapmap.org/cgi-perl/gbrowse/gbrowse/hapmap/?ref 5 chr7;start 5 64952391;stop 5 64969936;version 5 ) were selected from within and aroundthe ASL gene for PCR amplification followed by sequencing. Theseincluded the microsatellite D7S2549 (AFMA055XG9), locatedapproximately 146 kb upstream of the ASL gene start site, and withits map position confirmed by DeCODE (http://genome.ucsc.edu/cgi-bin/hgTracks?position 5 chr7:64964624-65022430&hgsid 5

67083678&stsMap 5 pack), two intronic simple tandem repeatsfrom the UCSC Genome Browser (http://genome.ucsc.edu/cgi-bin/hgTracks?position 5 chr7:64984963-65002090&hgsid 5 67375949&knownGene 5 pack&hgFind.matches 5 ,)and four intronic SNPs from HapMap, rs1183245(–) (SNP1),rs313830(–) (SNP2), rs313829(–) (SNP3), and rs1167406(1)(SNP4). The short tandem repeats were TG dinucleotide repeatat position chr7:64996368–64996420 and a TAAAT repeat atposition chr7:64999383–64999407. D7S2549 was amplified usingthe published primers and the following PCR conditions: 1.5 mMMgCl2 and 0.2ml Platinum Taq polymerase (Invitrogen, Burlington,Ontario, Canada; www.invitrogen.com) at 961C for 5 minutesfollowed by 951C for 30 seconds, 601C for 30 seconds, and 721C for30 seconds for 30 cycles, followed by 721C for 5 minutes. The TGrepeat was amplified using the primers 50-TGGGGACAAGTATTCTGTAC-30 (forward) and 50-TTTCAGGTGTGTCTGCAGG-30 (reverse) with the same conditions as described above,but with an annealing temperature of 551C. Amplification of theTAAAT repeat consistently produced nonspecific products, and wasdiscarded. Primers from the HapMap website (http://www.hapmap.org/cgi-perl/snp_details?name 5 rs118245) were used to amplifySNPs 1 and 3. Primers for SNPs 2 and 4 were 2f 50-ATGGCTTGGTCGTCACCTTAG-30, 2r 50-TTAGGGATTGA-CAGAGCTGGG-30, 4f 50-TAAGAGAGTATCTGCCCAAGG-30,and 4r 50-CCATGTTCTCTTGGTGAATCTG-30. Amplification ofSNPs was performed as described above but with an annealingtemperature of 501C.

RESULTSIdenti¢cation of Exon13 Skipping

When the 30 third of cDNA from the 28 cell strains wasamplified, six strains were found to have two cDNA populations

of different sizes, in roughly equal amounts; the PCR products offive strains are shown in Fig. 1A (the products of strain 1041 arenot shown). In these six strains, direct sequencing of the largerfragment yielded normal ASL sequence, while the smallerfragment was found to be deleted for exon 13 (Fig. 1B). Thisdeletion results in an in-frame removal of 84 bp from thetranscript, the translation of which would generate a polypeptideshortened by 28 amino acid residues.

Sequencing of Genomic DNA

Direct sequencing of the amplified products of genomic DNAidentified one of two deletions in each of the six cell strains withskipping of exon 13. In four strains (647, 926, 1006, and 1190), inaddition to the normal sequence, there was a 13-bp deletionwithin exon 13 (c.1045_1057del), which ended 5 bp upstream ofthe 30 splice junction of exon 13 (Fig. 2A). The other two strains(1040 and 1041) had a 25-bp deletion (c.1045_106217del) thatbegan at precisely the same position as the 13-bp deletiondescribed above, but extended into intron 13, 7 bp beyond the 30

splice site of exon 13 (Fig. 2A). It should be noted that there are2-bp direct GC repeats near the ends of the 13-bp deletion and3-bp TCA direct repeats near the ends of the 25-bp deletion.

SequenceAnalyses

An analysis of the DNA sequence in the region of the 50 end ofthe deletions was undertaken to identify sequences known to beassociated with deletions. Two such sequences were recognized(Fig. 2B). The first, TGGCGT, corresponds to nucleotides1041–1046 of the ASL cDNA, which matches perfectly therecognition sequence (TGGNGT) of a human pol-a mutation site[Kunkel, 1985]. This hexanucleotide sequence overlaps the 50 endof the two deletions. The second sequence of interest,GGCCACTGGCGTCATCTC (corresponding to nucleotides1035–1052 of the ASL cDNA, begins 50 to the beginning ofthe deletion and contains the TGGNGT pol-a mutation site.This sequence is similar to the consensus recognition sequence forhuman topoisomerase II [Muller et al., 1988], matching 7 out of10 nucleotides. The start of the two deletions in the ASL sequenceis shown in Figure 2B, along with the locations of the DNA pol-amutation site and topoisomerase II recognition sequence.

HaplotypeAnalyses

All of the patients shared common alleles for both D7S2549and the four SNPs. The TG tandem repeat was not polymorphic.Therefore no conclusion could be drawn as to the descent of eachof the mutations. Amplification of a second microsatellite marker,D7S663, located 1.2 Mb downstream of the ASL gene, was alsoattempted but was unsuccessful.

DISCUSSION

We have identified two deletions in the ASL gene in ASAL-deficient cell strains. Each of these deletions begins at precisely thesame nucleotide, but one deletion extends for only 13 bp, whereasthe other is 25-bp-long. Since these deletions are of two types,they have occurred independently, suggesting that this regionof ASL exon 13 is a deletion hotspot [Krawczak and Cooper,1991]. Moreover, these deletions occurred in individuals notknown to have any genetic relationship, and have beenindependently reported by one other group [Barbosa et al.,1991]. The 13-bp deletion within exon 13 leads to skipping ofexon 13 during splicing. It is possible that this deletion may lead toexon skipping because alterations in the secondary structure

HUMANMUTATION 27(11),1065^1071,2006 1067

Human Mutation DOI 10.1002/humu

caused by the deletion result in inaccurate splice-site selection[Schwarze et al., 1999]. It has been shown that normal splicing isdependent on the recognition of the 50 splice-donor site,the branch point, and the 30 splice-acceptor site by elements ofthe spliceosome machinery [Reed and Palandjian, 1997]. It istherefore possible that the 13-bp deletion interferes with normalsplicing of exon 13 in the ASL gene by removing the targetsequence for one of the spliceosome elements. Alternatively,the 13-bp deletion could lead to the loss of existing or generationof new exonic splicing enhancer (ESE) sites. Analysis of the ASL

sequence using ESEfinder Release 2.0 [Cartegni et al., 2003]indeed confirms that the deletion leads to the loss of SF2/ASF,SRp40, and SRp55 motifs, and the gain of an SC35 motif (score4.300027; threshold 2.383).

As stated above, this deletion has previously been reported inone apparently unrelated patient with ASAL deficiency [Barbosaet al., 1991], making this the most common ASL mutationidentified to date, and accounting for 5 out of 60 (8%) mutantalleles that we [Walker et al., 1997] and others [Barbosa et al.,1991] have examined. The 13-bp deletion is flanked by a GC

FIGURE 2. Two di¡erent deletions in theASL gene lead to a deletion of exon13 from themRNA. A: Sequencing of the PCR-ampli¢edASL genomic DNA froma control strain (left), ASAL-de¢cient strain1006 with a13-bp deletion (middle), andASAL-de¢cient strain1040 with a 25-bp deletion (right). Exon sequence is inuppercase text and intron sequence is in lowercase text. Both deletions begin18 bp 50 to the 30 end of exon13, at exactly the position of the predicted topoisomerase II cut site (arrow; also see Fig.2B).The13-bpdeletion ends 5 bp 50 to the end of exon13,whereas the 25-bp deletion extends 7 bp beyond the 30 splice site, into intron13. B:The¢rst panel shows theASLDNA sequence at the 30 end of exon13 and the location of the topoisomerase II and DNA pol-a consensussites.The asterisk denotes the start of the 13-bp and 25-bp deletions. In the second panel, a schematic of the twoASL deletions,and the putative topoisomerase II and DNA pol-a sites in exon 13 of theASL gene.The small solid bars near the ends of deletions(open boxes) indicate the location of theGC andTCA direct repeats.

1068 HUMANMUTATION 27(11),1065^1071,2006

Human Mutation DOI 10.1002/humu

direct repeat, and the 25-bp deletion by TCA repeats; directrepeats are commonly found near small deletions [Krawczak andCooper, 1991; Lehrman et al., 1986].

The identification of this deletion hotspot and analysis of thenucleotide sequences within and flanking it have suggested a

possible mechanism that may underlie the deletions. The sequenceat which these deletions begin overlaps both a putativetopoisomerase II consensus recognition site [Muller et al., 1988]and a DNA pol-a mutation/frameshift site [Krawczak and Cooper,1991; Kunkel, 1985]. The DNA pol-a mutation/frameshift site

FIGURE 3. A:The topoisomerase II and DNA pol-a sites within exon 10 of the human CFTR gene, in relation to the dF508 deletion.B:Two deletions in exon 9 of the humanHPRTgene, in relation to overlapping topoisomerase II and DNA pol-a sites. Note the 2-bpand 3-bp repeats (black bars) near the ends of the deletions. C: A 34-bp deletion in exon 1 of the humanWT1 gene, in relation tooverlapping topoisomerase II andDNApol-a sites. Note the 3-bpCTG repeats (black bars) £anking the deletion.

HUMANMUTATION 27(11),1065^1071,2006 1069

Human Mutation DOI 10.1002/humu

was defined as a region of DNA that was highly prone, in vitro, tosubstitutions and frameshifts by human DNA pol-a. The putativetopoisomerase II cut site is situated precisely at the beginning ofthe deletions, and in vitro studies in other systems havedemonstrated that type II topoisomerases can mediate illegitimaterecombination [Asami et al., 2002; Bae et al., 1988; Chiba et al.,1989; Ikeda, 1986], and are often associated with short regions ofhomology at the sites of recombination [Chiba et al., 1989]. Whenpol-a pauses at the pol-a mutation/frameshift site, it would resultin the DNA molecule forming a bent structure, predisposing to theinteraction of topoisomerase II with the DNA molecule at theblocked end (bend) of this DNA structure [Howard et al., 1991].At least in the human mitochondrial genome, it has been shownthat bent DNA structures may colocate with deletion hotspots[Hou and Wei, 1996]. Recognition of these bend DNA structuresand subsequent DNA strand breakage by topoisomerase II wouldpredispose to illegitimate recombination, particularly in thepresence of flanking direct repeats [Bae et al., 1988; Ikeda,1986; Mita et al., 1990]. In summary, we therefore propose thatboth the 13-bp and 25-bp deletions result from pol-a pausing atthe pol-a mutation/frameshift site, leading to a bent structure thatpredisposes to the binding of topoisomerase II. Slipped mispairingat the direct repeats would then predispose to illegitimaterecombination and deletion of the nucleotides between therepeats. Our finding of the topoisomerase II recognition andDNA pol-a sites in joint proximity to the small ASL deletions, andflanked by direct repeats, may be of general significance.Furthermore, the ability of the intervening sequence betweenthe direct repeats to fold upon itself may further facilitate slippage.Analysis of the deletion regions using Mfold [Zucker, 2003]revealed that several imperfect intrastrand DNA structures arepossible (Supplementary Fig. S1 available online at http://www.interscience.wiley.com/jpages/1059-7794/suppmat). The looped-out sequences may be deleted, possibly via replication bypass.

We have not observed, in other published small deletions andtheir surrounding sequences, the concurrence of the three putativepredisposing factors we have identified here: the two enzyme sites,the topoisomerase II cut site at the beginning of the deletion, andthe direct repeats. However, a large deletion in the NPH1 gene,responsible for juvenile nephronophthisis type 1, was found to beassociated with multiple topoisomerase II recognition sites, flankedby a 7-bp direct repeat, and with a pol-a site in the immediatevicinity of the break points [Otto et al., 2000].

Interestingly, the small deletion with one of the highest profilein human genetics, the DF508 deletion in the CFTR gene, has aperfect 5 out of 5 bp match for the pol-a site, and a 7 out of 10 bpmatch for topoisomerase II, whose cut site is precisely at the 30 endof the nucleotides deleted in DF508, although it is not flanked bydirect repeats (Fig. 3A).

A 5 out of 5 bp match to the DNA pol-a site and a 7 out of10 bp match to the topoisomerase II site are also present near thedeletion (c.474_476delCTG) that gives rise to Hb Coventry[Patrinos et al., 2004], a deletion in exon 3 of b-globin (HBB).We have identified similar joint occurrences of sequencesclosely matching the pol-a site (4/5 bp) and the topoiso-merase II site (7/10 bp) enzyme sites in exon 1 of b-globin (HbLyon; c.102_107delAAGGTG; cDNA ENST00000335295)[Patrinos et al., 2004], two deletions of exon 2 of b-globin (HbMontreal; c.270_278delGATGGCCTG and Hb Dresden;c.151_156delTGGTCT) [Patrinos et al., 2004], and two deletionsin exon 9 of HPRT (c.620_622delGTT and c.621_623delTTG;cDNA ENST00000298556) [Krawczak and Cooper, 1991].Notably, all of these deletions also have flanking direct repeats.

Although the topoisomerase II cut site is not directly within all ofthese deletions, the presence of a nearby double-stranded DNAbreak may be more important in contributing to the generation ofdeletions than the precise location of the cut. We have alsorecognized that a 34-bp deletion in exon 1 of the WT1 gene,which is flanked by 3-bp direct repeats, has close matches forthe pol-a (4/5) and topoisomerase II (7/10) sites overlapping withthe start of the deletion, with the topoisomerase II cut being only1 bp distal to the start of the deletion [Huff et al., 1995].The relationship of these mutations to the HPRT sequence isshown in Figure 3B, along with the locations of the DNA pol-amutation site and topoisomerase II recognition sequence, and forthe WT1 sequence in Figure 3C. These findings, taken togetherwith our observations on the exon 13 deletions of ASL, suggestthat the joint presence of these enzyme sites may create a region atincreased risk for deletions in the human genome.

To our knowledge, the association of a topoisomerase II site anda DNA pol-a mutation site in the region of a deletional hotspothas not been previously recognized, and furnishes a possible reasonfor the location of this deletional hotspot in the ASL gene. Inconclusion, we have identified a new deletional hotspot in theASL gene, occurring in association with topoisomerase II andDNA pol-a mutation sites. We suggest that this association maynot be coincidental, and that it provides indirect evidence for amechanism for the generation of deletions in this region of theASL gene. Our observation that similar sequences are also locatedin the region of the DF508 deletion in the CFTR gene, in anNPH1 deletion, and in HPRT, HBB, and WT1 variants, suggeststhat the mechanism that we propose as predisposing to thesedeletions may have general relevance.

ACKNOWLEDGMENTS

We thank Andrew Grimm, Genetic Metabolic DiseasesResearch Unit, Children’s Hospital at Westmead, with helpfuladvice and comments relating to this manuscript.

REFERENCES

Antonarakis SE, Krawczak M, Cooper DN. 2000. Disease-causingmutations in the human genome. Eur J Pediatr 159:S173–S178.

Asami Y, Jia D-W, Tatebayashi K, Yamagata K, Tanokura M, IkedaH. 2002. Effect of the DNA topoisomerase II inhibitor VP-16on illegitimate recombination in yeast chromosomes. Gene291:251–257.

Bae Y-S, Kawasaki I, Ikeda H, Liu LF. 1988. Illegitimaterecombination mediated by calf thymus DNA topoisomerase IIin vitro. Proc Natl Acad Sci USA 85:2076–2080.

Bae Y-S, Chiba M, Ohira M, Ikeda H. 1991. A shuttle vector foranalysis of illegitimate recombination in mammalian cells: effectsof DNA topoisomerase inhibitors on deletion frequency. Gene101:285–289.

Barbosa P, Cialkowski M, O’Brien WE. 1991. Analysis of naturallyoccurring and site-directed mutations in the argininosuccinatelyase gene. J Biol Chem 266:5286–5290.

Benzer S. 1961. On the frequency of the genetic fine structure.Proc Natl Acad Sci USA 47:403–417.

Brusilow SW, Horwich AL. 2001. Urea cycle enzymes. In: ScriverCS, Beaudet AL, Sly WS, Valle D, editors. The metabolic andmolecular bases of inherited disease. New York: McGraw-Hill.p 1909–1963.

1070 HUMAN MUTATION 27(11),1065^1071,2006

Human Mutation DOI 10.1002/humu

Cartegni L, Wang J, Zhu Z, Zhang MQ, Krainer AR. 2003.ESEfinder: a web resource to identify exonic splicing enhancers.Nucleic Acids Res 31:3568-3571.

Chiba M, Shimizu H, Fujimoto A, Nashimoto H, Ikeda H. 1989.Common sites for recombination and cleavage mediated bybacteriophage T4 DNA topoisomerase in vitro. J Biol Chem264:12785–12790.

Chou Q, Russell M, Birch DE, Raymond J, Bloch W. 1992.Prevention of pre-PCR mis-priming and primer dimerizationimproves low-copy-number amplifications. Nucleic Acids Res20:1717–1723.

Cooper DN, Krawczak M. 1990. The mutational spectrum ofsingle base-pair substitutions causing human genetic disease:patterns and predictions. Hum Genet 85:55–74.

Cooper DN, Krawczak M, Antonarakis SE. 1998. The nature andmechanisms of human gene mutation. In: Vogelstein B, KinzlerKW, editors. The genetic basis of cancer. New York: McGraw-Hill. p 65–95.

Frouin I, Montecucco A, Spadari S, Maga G. 2003. DNAreplication: a complex matter. EMBO Rep 4:666–670.

Fry M, Loeb LA. 1992. A DNA polymerase a pause site is a hotspot for nucleotide misinsertion. Proc Natl Acad Sci USA 89:763–767.

Han Y-H, Austin MJF, Pommier Y, Povirk LF. 1993. Small deletionand insertion mutations induced by the topoisomerase IIinhibitor Teniposide in CHO cells and comparison with sitesof drug-stimulated DNA cleavage in vitro. J Mol Biol 229:52–66.

Hou J-H, Wei Y-H. 1996. The unusual structures of hot-regionsflanking large-scale deletions in human mitochondrial DNA.Biochem J 318:1065–1070.

Howard MT, Lee MP, Hsieh TS, Griffith JD. 1991. Drosophilatopoisomerase II-DNA interactions are affected by DNAstructure. J Mol Biol 217:53–62.

Huff V, Jaffe N, Saunders GF, Strong LC, Villalba F, RuteshouserEC. 1995. WT1 exon 1 deletion/insertion mutations in Wilmstumor patients, associated with di- and trinucleotide repeats anddeletion hotspot consensus sequences. Am J Hum Genet 56:84–90.

Ikeda H. 1986. Bacteriophage T4 DNA topoisomerase mediatesillegitimate recombination in vitro. Proc Natl Acad Sci USA 83:922–926.

Jackson MJ, Kobayashi K, Beaudet AL, O’Brien WE. 1989.Analysis of deletions at the human argininosuccinate synthetaselocus. Mol Biol Med 6:179–186.

Jouanguy E, Lamhamedi-Cherradi S, Lammas D, Dorman SE,Fondaneche M-C, Dupuis S, Doffinger R, Altare F, Girdlestone J,Emile J-F, Ducoulombier H, Edgar D, Clarke J, Oxelius V-A, BraiM, Novelli V, Heyne K, Fischer A, Holland SM, Kumararatne DS,Schreiber RD, Casanova J-L, 1999. A human IFNGR1 smalldeletion hotspot associated with dominant susceptibility tomycobacterial infection. Nat Genet 21:370–378.

Krawczak M, Cooper DN. 1991. Gene deletions causing humangenetic disease: mechanisms of mutagenesis and the role of thelocal DNA sequence environment. Hum Genet 86:425–441.

Krawczak M, Ball EV, Cooper DN. 1998. Neighbouring-nucleotideeffects on the rates of germ-line single-base-pair substitutionin human genes. Am J Hum Genet 63:474–488.

Kunkel TA. 1985. The mutational specificity of DNA polymerasesalpha and gamma during in vitro DNA synthesis. J Biol Chem260:12866–12874.

Lebo RV, Ikuta T, Milunsky JM, Milunksy A. 2001. Rett syndromefrom quintuple and triple deletions within the MECP2 deletionhotspot region. Clin Genet 59:406–417.

Lehrman MA, Russell DW, Goldstein JL, Brown MS. 1986.Exon-Alu recombination deletes 5 kilobases from the lowdensity lipoprotein receptor gene, producing a null phenotypein familial hypercholesterolemia. Proc Natl Acad Sci USA83:3679–3683.

McInnes RR, Shih V, Chilton S. 1984. Interallelic complementa-tion in an inborn error of metabolism: genetic heterogeneityin argininosuccinate lyase deficiency. Proc Natl Acad Sci USA81:4480–4484.

Mita S, Rizzuto R, Moraes CT, Shanske S, Arnaudo E, Fabrizi GM,Koga Y, DiMauro S, Schon EA. 1990. Recombination viaflanking direct repeats is a major cause of large scale deletions ofhuman mitochondrial DNA. Nucleic Acid Res 18:561–567.

Muller MT, Spitzner JR, DiDonato JA, Mehta VB, Tsutsui K,Tsutsui K. 1988. Single-strand DNA cleavages by eukaryotictopoisomerase II. Biochemistry 27:8369–8379.

Otto E, Betz R, Rensing C, Schatzle S, Kuntzen T, Vetsi T, Imm A,Hildebrandt F. 2000. A deletion distinct from the classicalhomologous recombination of juvenile nephronophthisis type 1(NPH1) allows exact molecular definition of deletion break-points. Hum Mutat 16:211–223.

Patrinos GP, Giardine B, Riemer C, Miller W, Chui DH, AnagnouNP, Wajcman H, Hardison RC. 2004. Improvements in theHbVar database of human hemoglobin variants and thalassemiamutations for population and sequence variation studies.Nucleic Acids Res 32(Database issue):D537–D541.

Reed R, Palandjian L. 1997. Spliceosome assembly. In: KrainerAR, editor. Eukaryotic mRNA processing. Oxford: IRL Press.p 103–129.

Sambrook J, Fritsch EF, Maniatis T. editors. 1989. Gel electro-phoresis of DNA. In: Molecular cloning: a laboratory manual.Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.p 6.46–6.48.

Schwarze U, Starman BJ, Byers PH. 1999. Redefinition of exon 7in the COL1A1 gene of type I collagen by an intron 8 splice-donor-site mutation in a form of osteogenesis imperfecta:influence of intron splice site order on outcome of splice-sitemutation. Am J Hum Genet 65:336–344.

Stenson PD, Ball EV, Mort M, Phillips AD, Shiel JA, ThomasNST, Abeysinghe S, Krawczak M, Cooper DN. 2003. HumanGene Mutation Database (HGMD): 2003 update. Hum Mutat21:577–581.

Walker DC, McCloskey DA, Simard LR, McInnes RR. 1990.Molecular analysis of human argininosuccinate lyase: mutantcharacterization and alternative splicing of the coding region.Proc Natl Acad Sci USA 87:9625–9629.

Walker DC, Christodoulou J, Craig HJ, Simard LR, Ploder L,Howell PL, McInnes RR. 1997. Intragenic complementation atthe human argininosuccinate lyase locus: identification of themajor complementing alleles. J Biol Chem 272:6777–6783.

Zucker M. 2003. Mfold web server for nucleic acid folding andhybridization prediction. Nucleic Acids Res 31:3406–3415.

HUMAN MUTATION 27(11),1065^1071,2006 1071

Human Mutation DOI 10.1002/humu