DNA bending in the replication zone of the C3 DNA ... - Springer Link

Molecular Biology Reports (2006) 33: 71–82 DOI 10.1007/s11033-006-0009-4


Springer 2006

DNA bending in the replication zone of the C3 DNA puff amplicon of Rhynchosciara americana (Diptera: Sciaridae) Adriana Fiorini1, Fabiana de Souza Gouveia1, Maria Albertina de Miranda Soares2, Ann Jacob Stocker3, Ricardo Rodrigues Ciferri4 & Maria Aparecida Fernandez1,* 1

Departamento de Biologia Celular e Gene´tica, Universidade Estadual de Maringa´, Av. Colombo, 5790, 87020-900, Maringa´, Parana´, Brasil; 2Departamento de Biologia Estrutural, Molecular e Gene´tica, Universidade Estadual de Ponta Grossa, Parana´, Brasil; 3Centre for Environmental Stress and Adaptation Research, University of Melbourne, Parkville, Victoria, Australia; 4Departamento de Informa´tica, Universidade Estadual de Maringa´, Maringa´, Parana´, Brasil; *Author for correspondence (Phone: +55-44-3261-46-83; Fax: +55-44-3261-48-93; E-mail: [email protected]/[email protected])

Accepted 18 January 2006

Key words: gene amplification, intrinsic bent DNA, replication origin, Rhynchosciara americana

Abstract Intrinsic bent DNA sites were identified in the 4289 bp segment encompassing the replication zone which directs DNA amplification and transcription of the C3-22 gene of Rhynchosciara americana. Restriction fragments showed reduced electrophoretic mobility in polyacrylamide gels. The 2D modeling of the 3D DNA path and the ENDS ratio values obtained from the dinucleotide wedge model of Trifonov revealed the presence of four major bent sites, positioned at nucleotides )6753, )5433, )5133 and )4757. Sequence analysis showed that these bends are composed of 2–6 bp dAÆdT tracts in phase with the DNA helical repeat. The circular permutation analysis permitted the verification that the fragments containing the bending sites promote curvature in other sequence contexts. Computer analyses of the 4289 bp sequence revealed low helical stability (DG values), negative roll angles indicating a narrow minor groove and a putative matrix attachment region. The data presented in this paper add to information about the structural features involved in this amplified segment.

Introduction In all organisms the sequence features present in replication and transcription regions have been investigated. One of those features included intrinsic bent DNA sites, which are usually associated with the repetition of short adenine and thymine (dAÆdT) tracts phased close to the helix repeat, 10.5 bp/turn [1]. Curved DNA sequences have been found at regulatory regions or nuclear processes including prokaryotic promoters [2–4], eukaryotic promoters [5–10], recombination sites [11], regions of nucleosome formation [12, 13], Scaffold/Matrix Attachment Regions – S/MARs [14–17] and DNA

replication origins [18–20]. In relation to replication origins, the presence of DNA curvature has been reported near or within replication origins, although the role of these structures on replication control is not well known. Sequence-derived bends have been found associated with replication origins in a number of systems, such as prokaryote genome origins [21–23], plasmid origins [24–26], the SV40 replication origin [27, 28], autonomously replicating sequences – ARS [29, 30] and eukaryote genome origins [20, 31–34]. Some of the studies have suggested that DNA bending was necessary for origin activity [20, 21, 24, 28, 30, 34]. Although specific replication origins in simple genomes have been well characterized, a consensus sequence that

72 may contribute to origin function in metazoan chromosomes has not been well established and the origin can be localized from random to specific sites [35–39]. We are interested in searching for structural characteristics present in amplified regions of sciarid flies. The replication process in the salivary glands of sciarid flies is reiterated during late larval development, i.e., additional DNA synthesis occurs at specific chromosomal loci. This differential replication results in DNA amplification, which occurs in the chromosomal regions named DNA puffs and also involves intense transcription of these regions. The DNA amplification process provides an interesting and useful model for studying the control of the initiation of replication and transcription throughout development in eukaryotes. It has been proposed that the role of amplified genes within the DNA puffs is to encode secretory proteins synthesized at large quantities to build the pupal cocoon [40, 41]. The gene C3-22, localized within the C3 amplicon of Rhynchosciara americana, has been studied in detail because of its high level of amplification (about 32 times) during the time of DNA puff formation [42]. The zone of initiation of replication that directs this DNA amplification begins about 2 kb upstream of the C3-22 transcription unit and extends for about 6 kb [42]. In the present paper, we report the characterization of bent DNA sites in the Rhynchosciara americana C3 amplicon origin zone. These findings will provide further data for the studies of structural features in this amplified region.

Materials and methods DNA samples The 6809 bp fragment ()7893 to )1084 position, referring to the transcription start point as +1) containing the putative replication initiation zone is cloned in pBluescript II – Stratagene, named pSc6.5 [42], and was kindly provided by Dr. Ann J. Stocker and Dr. Francisco J. S. Lara (Universidade de Sa˜o Paulo – USP). The accession number for the 6809 bp sequence is AY225408, but the pSc6.5 clone analyzed in this work contains a deletion of 386 bp within the 3525–3911 bp segment, relative to the published sequence. To certify

this, both strands of the pSc0.7 kb internal clone were sequenced using the DYEnamic ET Dye Terminator Kit (Amersham Biosciences) in an automated DNA sequencer MegaBACE 1000. This difference probably occurs because R. americana flies are collected in nature and may present sequence polymorphisms. The recombinant plasmid was isolated from E. coli competent strains by the CTAB method [43] and/or by alkaline lysis [44]. Restriction of recombinant plasmids and electrophoretic analysis Fragments from clone pSc6.5 were obtained by digestion with restriction enzymes according to the manufacturers instructions. The mobilities of the fragments were compared after electrophoresis in 1% agarose gels (control gels) at 3.5 V/cm at room temperature for 5 h versus electrophoresis in 6% polyacrylamide gels (PA) (acrylamide/bisacrylamide ratio 30:0.8) at 7 V/cm at 4 C, for 12–14 h. Since DNA fragments containing unusual structures show an anomalous mobility in PA [45], this technique is powerful for investigation of the curvature and flexibility of DNA segments. To confirm the presence of bent DNA, the samples were maintained overnight with 1 lg/ml ethidium bromide (EthBr) and electrophoresed in 6% PA previously run with 1 lg/ml EthBr in the running buffer. The EthBr acts locally between base pairs, abolishing the intrinsic DNA curvature and straightening the fragments [46]. All gels were run in 1  TBE buffer (45 mM Tris–borate, 1 mM EDTA, pH 8.0), stained after running with 0.1 lg/ ml of EthBr and documented by a UVP BioImaging System under UV light. In all three gel systems, an R-value, corresponding to the ratio of the observed length to the expected length, was calculated for each DNA fragment to determine the mobility alteration in the gels. The DNA molecular weight standard 1 kb ladder (GIBCO BRL) was electrophoresed on both sides of the gels. Circular permutation assay This assay was performed as previously described to determine the bending-center site on permuted fragments [47, 48] and verify if the bending sites promote curvature in other sequence contexts as

73 reported by Du et al. [33]. When the bend center is located in the middle of the DNA fragments, the effect of the curvature is most pronounced, resulting in retarded migration. To generate fragments that contain the four main bent DNA sites, the pSc6.5 clone was digested separately with the enzymes EcoRI, ScaI+EcoRI and ClaI+DraI. The 528 bp EcoRI/EcoRl, 485 bp ScaI/EcoRI and 534 bp ClaI/DraI resultant fragments (oriented by sequencing), which contain the bent DNA sites (b1), (b2 and b3) and (b4), respectively, were recovered from agarose gels by elution and cloned into pBluescript vector compatible ends (Figures 4a and b). All the three fragments were excised from the vector using XbaI and SalI restriction polylinker sites (Figure 4a), gel-purified and cloned into XbaI/SalI-cleaved pBendBlue [49] (Figure 4c). The pBendBlue vector contains two identical DNA segments with 17 restriction sites in a repeat spanning a central region containing XbaI and SalI cloning sites. Figure 4c shows only the restriction sites used in this work. To generate a set of circularly permuted DNA fragments that differ in the position of the inserts relative to their ends, the resultant constructions were digested separately with the restriction enzymes BglII, XhoI, DraI, PvuII, NruI and KpnI. The permutated fragments were electrophoresed in 6% PA at 4 C and 11 V/cm during 5 h. The relative mobility (relative to normal migration, 100%) of each fragment was determined and plotted against the distance from the 5¢ end cleavage site of the first enzyme used in the polylinker to the midpoint of the fragment generated by each of the restriction enzymes used. Computer analyses of replication zone sequence The computer analyses of curvature of a DNA sequence give results that are compatible with electrophoretic data [50]. In this report, we have used the dinucleotide wedge model of Trifonov [51, 52] to predict the 3D path of the 4289 bp DNA segment (from the pSc6.5 clone), which contains the major bent elements and the preferential replication origin region. The projection of the three dimensional path and maps of the structural parameters (DNA curvature, helical stability variations and roll angles) were calculated by Map15a and 3D15m1 software using the algorithm of Eckdahl and Anderson [50] and the

helical parameters of Bolshoy et al. [52] as described by Pasero et al. [53] and Marilley and Pasero [54]. Each plotted value takes into account the contribution of the surrounding nucleotides. Putative bent segments were identified using the ENDS ratio, which shows the ratio of the contour length of the segment’s helical axis to the shortest distance between the fragments ends. The AT percentage, the DG values (which predict the stability of the DNA duplex) and variations of roll angle (which estimate the rolling-open of the base pairs along their long axes) maps were computed by Map15a software using a 120 bp window width and a 10 bp step. This window was chosen according to the fragment length, 4289 bp. By convention, if base pairs open toward the minorgroove side, the roll angle is positive. On the other hand, negative roll angles that cause a narrow minor groove suggest a bent region [22]. In the 2D projected figures, the roll angles variation and the DG values were calculated in a 20 bp window. This window was applied in accordance to the short fragment lengths, allowing a better 2D visualization of the fragments as a whole. Putative S/MARs were predicted for the 4289 bp fragment using the MAR-Finder tool [55], now known as MAR-Wiz available online at http:// www.futuresoft.org/MAR-Wiz/. This tool uses sequence characteristics found in S/MARs to identify potential S/MAR sites. A MAR-potential is determined based on the probability of finding combinations of these sequence motifs in a given stretch of DNA. The resulting MAR-potential values are normalized between 0 and 1. A ‘‘true MAR’’ as defined by the program’s authors has a high potential value (above 0.6) for a least three consecutive readings.

Results Restriction fragments within the replication zone show anomalous electrophoretic mobility Our initial assay for the identification of intrinsic bent DNA sites in the C3 DNA puff amplicon was the analysis of electrophoretic behavior of DNA sequences within this region. The restriction fragments of the 6.8 kb segment containing the replication zone (pSc6.5 clone) were analyzed in agarose and PA and the results are shown in

74 Figure 1. In PA, the migration of DNA fragments is dependent on both size and structure. Fragments containing bent DNA segments migrate through the mesh of this gel with lower or higher apparent molecular sizes than expected from the fragment length. In contrast, the migration of fragments through agarose gels, which form an irregular mesh, is dependent only on their size. The mobility reduction of restriction fragments (Rvalue, determined as the apparent size divided by its actual size) was measured in three gel systems: agarose gels (AG), PA and polyacrylamide gels with ethidium bromide (PA+EthBr). R-values between 0.9 and 1.10 signify no alteration in the fragment mobility and R-values ‡1.11 signify a reduced mobility indicative of bent DNA [7]. R-values 100 signify that the fragment has a higher mobility. The 97% relative mobility also indicates that the permuted fragment has not showed mobility alteration. The bent site b4 in the 534 bp ClaI/DraI insert is already located near the 3¢ end and the presence of the polylinker sequence at this end, through KpnI cleavage was not sufficient to centralize this bend site (Figure 5a). To visualize the shape of the fragments containing bends, the 2D projection of the 528 bp EcoRI/EcoRI, 485 bp ScaI/EcoRI and 534 bp ClaI/DraI fragments was performed (Figure 5d).

In the projected figure, the DG values vary from 30.1 to 49.8 kcal/mol and the roll angle values vary from )1.50 to 1.01 along the fragments. The negative roll values indicate that the bent DNA sites considered in this work display the ability to form a narrow minor groove and the low values of DG indicate a duplex instability on these regions. These observations are consistent with the idea that sequences around the bent sites contribute to the shape of the analyzed fragment. Structural features in the origin region containing the bending sites To identify other sequence elements that are common in replication origins, the 4289 bp region ()7893 to )3604 position) which encompasses the

79

Figure 5. Circular permutation analysis. (a) Representation of the 706, 651 and 682 bp permuted fragments resulting from cleavage with the repeat restriction enzymes of pBendblue polylinker, B, BglII; D, DraI; K, KpnI, N, NruI; P, PvuII; X, XhoI. The gray bars represent the inserts and the thin lines represent the 103 bp polylinker region of the vector. The positions of the bent sites b1, b2, b3 and b4 are indicated by diamonds. (b) Migration of the restriction fragments in 6% PA. The permuted fragments are indicated by arrows in the gels. The additional fragments shown in the DraI cleavages are from the vector. (c) Graphics showing the relative mobility of each fragment plotted against the distance from the 5¢ end cleavage site of the first enzyme used in the polylinker to the midpoint of the fragment generated by each of the restriction enzymes used. (d) 2D projection of the 3D path showing the variation of helical stability and minor groove width along the 528, 485 and 534 bp inserts. In the projected figures, the DG or free energy of the duplex (30.1–49.8 kcal/mol; circles of increasing size) and the roll angle variation ()1.50 to 1.01; gray circles of decreasing intensity) were calculated as described in Materials and methods. The region around the bent DNA sites shows low DG (minor circles) and negative roll angle (dark gray circles).

80 bend sites and the most active replication origin region, was analyzed by graphics of roll angle variation, AT percentage. DG values and MARpotential (Figure 6) as described in Material and methods. The area in the vicinity of the most active ori displays multiple features that are characteristic of eukaryotic origins of replication. The mean overall AT fragment content is 65.04% and for the

most active ori is 67.67%. Consequently, low values of predicted duplex stability, DG, were found for the AT rich regions. All bending sites showed negative roll angle values indicating that they are curved toward the minor groove. Potential S/ MARs were identified in two regions ()6513 and )4083 positions). These putative S/MARs encompass an AT rich region with negative roll angles, and the last one, )4083 position, overlaps with the most active ori (dotted line).

Discussion R. americana C3 puff DNA amplicon contains bent DNA sites

Figure 6. Structural features of the 4289 bp fragment ()7893 to )3604 position) which contains the most active region of replication origin and the bent DNA sites (b1–b4). The roll angle, AT percentage, DG values and ENDS ratio were calculated and graphed as a function of position in a 120 bp window width and a 10 bp step using the parameters described in Materials and methods. MAR prediction was generated by the MAR-Wiz tool with a 200 bp window width and a 10 bp step. The most active region of replication origin, indicated by dotted lines, shows 75% AT, negative values of roll angle, low DG values and two predicted MARs at the )6513 and )4083 positions. Restriction site: H, HindIII.

The replication zone of the C3-22 gene begins about 2 kb upstream the transcription unit and extends for about 6 kb. Recently, bending DNA sites have been detected in the promoter region, transcription unit and downstream region of the C3-22 gene (position )205 at +4240 bp) (Gouveia, F. de S., personal communication). Here, we mapped by computer analysis at least four major bend sites at )6753 (b1), )5433 (b2), )5133 (b3) and )4473 (b4) positions, with ENDS ratio values up to 1.15. With these results we can observe that these bending sites coincided with the replication zone, and are in a 2 kb upstream region of the transcription unit. The role of bending sites in replication and/or transcription in this amplicon needs to be investigated. Analysis of electrophoretic behavior revealed that the restriction fragments in the region between the )7893 and )1084 nucleotide position upstream from the C3-22 transcription unit show reduced electrophoretic mobility in PA, with R-values between 1.14 and 1.26. The sequence analysis of the four bend sites revealed the predominance of stretches of multiple poly (dA) and poly (dT) tracts around the bend centers, showing a periodicity of 10 bp and multiples of them. Origin regions from many organisms contain stretches of dAÆdT tracts known to be the cause of intrinsic DNA curvature. The origin region of the amplified dihydrofolate reductase (DHFR) gene in Chinese hamster ovary cells (CHOC 400 strain) presents stable bent elements in a 280 bp fragment, at 3342–3622 nucleotides, with five dA tracts spaced in each 10 bp [32].

81 Altman and Fanning [20] demonstrated that DR12 cells transfected with ori-b DNA segments lacking the bend sites show little replication initiation activity, indicating that bent DNA segments are important for origin activity. Sua´rezLopes et al. [56] identified the presence of a significant intrinsic DNA curvature near the WDV LIR (Large Intergenic Region of Wheat Dwarf Geminivirus) stem-loop, suggesting that it might serve as a binding site for factors involved in viral replication. In ARS (Autonomously Replicating Sequence) from Saccharomyces cerevisiae and in origins of replication from bacteria, DNA bending has been implicated as a structural element to facilitate the necessary DNA–protein interactions. Du et al. [33] have shown through experimental analysis and ENDS ratio calculation that restriction fragments of the 5¢NTS rRNA genes’ replication origin in Tetrahymena thermophila contain segments of bent DNA. The authors suggest that these bend sites may serve to facilitate initiator protein interactions or the association of the origin with the nuclear matrix. The proteins involved in initiation of DNA replication in yeast also appear to be used to promote replication in multicellular eukaryotes [57]. These include the six origin recognition proteins composing the origin recognition complex (ORC), which bind to specific sites and determine where replication begins. Specific regions that bind ORC have been identified only in two fly loci, which promote repeated rounds of DNA replication: the amplification origin in the chorion locus on chromosome 3 of Drosophila melanogaster [58] and the amplification origin in the II/9A locus of Sciara coprophila [34]. In this work it was proposed that putative bend sites in the upstream region of the II/9A locus replication origin might attract histones, leaving the origin region free to bind the ORC complex. Sequence elements in the C3 amplicon Computer analysis of the 4289 bp sequence in the replication zone of the C3 amplicon revealed that the mean overall AT fragment content is 65.04%. The major concentration of adenine (a mean of 67.67%) was found in the most active ori. Thermodynamic analyses of the 4289 bp sequence revealed that an overall mean DG for unwinding is 216 kcal/mol, ranging from 199 to 245 kcal/mol.

In the most active ori the DG values range from 201 to 230 kcal/mol. These DG values are similar to that found in the S. cerevisiae ARS1 DNA fragment: 207–227 kcal/mol [59]. Regions of low helical stability caused by the AT rich tracts facilitate the initial unwinding of the DNA molecule and are often found in the vicinity of replication origins [60, 61]. AT tracts can serve as binding sites for nuclear matrix-associated proteins and DUE (DNA unwinding elements) at the site where DNA unwinding begins [62]. The most prevalent feature of S/MARs is the presence of AT rich tracts [63]. In the 4289 bp fragment, a peak value of up to 0.6 of potential matrix attachment sites, predicted by the MAR-Wiz program, align with the experimentally determined ori. The 4289 bp fragment showed negative roll angles, predominantly in the most active ori that possibly will cause the narrowing of the helix minor groove. Putative SARs (Scaffold Attachment Regions) colocalizing with eukaryotic origins are generally AT rich and present a narrow minor groove [64]. Vernis et al. [59] reported that the region of low roll angles matches the mapped SAR in S. cerevisiae. Replication origins occur in close proximity to (or in association with) the nuclear matrix and the DNA is always associated with the nuclear matrix while it is replicating [65]. To verify the theoretical data, experimental approaches are necessary to determine whether the replication zone of the R. americana C3-22 gene is associated with the nuclear matrix. Acknowledgements A. Fiorini and F. de S. Gouveia received a fellowship from CAPES (Conselho de Aperfeic¸oamento do Pessoal do Ensino Superior). The software Map15a and 3D15m1 were kindly give by Philippe Pasero from Institute of Molecular Genetics, Montpellier, France. This work was supported by grants from CNPq, TWAS and FINEP/Fundac¸a˜o Arauca´ria. This paper is dedicated to Dr. Francisco Jeronimo Salles Lara (in memoriam). References 1. Koo HS, Wu HM & Crothers DM (1986) Nature 320: 501–506.

82 2. Perez-Martin J, Rojo F & de Lorenzo V (1994) Microbiol. Rev. 58: 268–290. 3. Gabrielian AE, Landsman D & Bolshoy A (1999) Silico Biol. 1: 183–196. 4. Jauregui R, Abreu-Goodger C, Moreno-Hagelsieb G, Collado-Vides J & Merino E (2003) Nucleic Acids Res. 31: 6770–6777. 5. Delabre ML, Pasero P, Marilley M & Bougis PE (1995) Biochemistry 34: 6729–6739. 6. Nair TM (1998) FEBS Lett. 422: 94–98. 7. de Souza ON & Ornstein RL (1998) Biopolymers 46: 403– 415. 8. Roux-Rouquie M & Marilley M (2000) Nucleic Acids Res. 28: 3433–3441. 9. Bash RC, Vargason JM, Cornejo S, Ho S & Lohr D (2001) J. Biol. Chem. 276: 861–866. 10. Fiorini A, Basso LR Jr., Pac¸o´-Larson ML & Fernandez MA (2001) J. Cell Biochem. 83: 1–13. 11. Milot E, Belmaaza A, Wallenburg JC, Gusew N, Bradley WE & Chartrand P (1992) EMBO J. 11: 5063–5070. 12. Wada-Kiyama Y, Kuwabara K, Sakuma Y, Onishi Y, Trifonov EN & Kiyama R (1999) FEBS Lett. 444: 117–124. 13. Virstedt J, Berge T, Henderson RM, Waring MJ & Travers AA (2004) J. Struct. Biol. 148: 66–85. 14. Homberger HP (1989) Chromosoma (Berl) 98: 99–104. 15. Boulikas T (1993) J. Cell Biochem. 52: 14–22. 16. Fukuda Y (1999) Plant Mol. Biol. 39: 1051–1062. 17. Fukuda Y (2000) Plant Mol. Biol. 44: 91–98. 18. DePamphilis ML (2000) Struct. Biol. 129: 186–197. 19. Biamonti G, Paixa˜o S, Montecucco A, Peverali FA, Riva S & Falaschi A (2003) Chromosome Res. 11: 403–412. 20. Altman AL & Fanning E (2004) Mol. Cell. Biol. 24: 4138– 4150. 21. Zahn K & Blattner FR (1987) Science 236: 416–422. 22. Calladine CR & Drew HR (1997) Understanding DNA: The Molecule and How It Works, 2nd ed., Academic Press, London. 23. Hosid S, Trifonov EN & Bolshoy A (2004) BMC Mol. Biol. 5: 14. [http://www.biomedcentral.com/1471-2199/5/ 14]. 24. Mukherjee S, Patel I & Bastia D (1985) Cell 43: 189–197. 25. Koepsel RR & Khan SA (1986) Science 233: 1316–1318. 26. Dong XN, Rouillard KP, Womble DD & Rownd RH (1989) J. Bacteriol. 171: 703–707. 27. Deb S, DeLucia AL, Koff A, Tsui S & Tegtmeyer P (1986) Mol. Cell. Biol. 6: 4578–4584. 28. Ryder K, Silver S, DeLucia AL, Fanning E & Tegtmeyer P (1986) Cell 44: 719–725. 29. Snyder M, Buchman AR & Davis RW (1986) Nature 324: 87–89. 30. Williams JS, Eckdahl TT & Anderson JN (1988) Mol. Cell. Biol. 8: 2763–2769. 31. Linial M & Shlomai J (1988) Nucleic Acids Res. 16: 6477– 6492. 32. Caddle MS, Dailey L & Heintz NH (1990) Mol. Cell Biol. 10: 6236–6243. 33. Du C, Sanzgiri RP, Shaiu W-L, Choi J-K, Hou Z, Benbow RM & Dobbs DL (1995) Nucleic Acids Res. 23: 1766– 1774.

34. Gerbi SA, Strezoska Z & Waggener JM (2002) J. Struct. Biol. 140: 17–30. 35. Kobayashi T, Rein T & DePamphilis ML (1998) Mol. Cell. Biol. 18: 3266–3277. 36. Altman AL & Fanning E (2001) Mol. Cell Biol. 21: 1098– 1110. 37. Anglana M, Apiou F, Bensimon A & Debatisse M (2003) Cell 114: 385–394. 38. Debatisse M, Toledo F & Anglana M (2004) Cell Cycle 3: 19–21. 39. Tower J (2004) Annu. Rev. Genet. 38: 273–304. 40. de Toledo SM & Lara FJ (1978) Biochem. Biophys. Res. Commun. 85: 160–166. 41. Laicine EM, Alves MA, de Almeida JC, Rizzo E, Albernaz WC & Sauaia H (1984) Chromosoma 89: 280–284. 42. Yokosawa J, Soares MAM, Dijkwel PA, Stocker AJ, Hamlin JL & Lara FJS (1999) Chromosoma 108: 291– 301. 43. Del Sal G, Manfioletti G & Schneider C (1989) Biotechniques 7: 514–520. 44. Sambrook J & Russell DW (2001) Molecular cloning: a laboratory manual, 3rd ed., Cold Spring Harbor Laboratory Press, New York. 45. Marini JC, Levene SD, Crothers DM & Englund PT (1982) Proc. Natl. Acad. Sci. USA 79: 7664–7668. 46. Diekmann S & Wang JC (1985) J. Mol. Biol. 186: 1–11. 47. Wu HM & Crothers DM (1984) Nature 308: 509–513. 48. Kim J, Zwieb C, Wu C & Adhya S (1989) Gene 85: 15–23. 49. Sperbeck SJ & Wistow GJ (1998) Biotechniques 24: 66–68. 50. Eckdahl TT & Anderson JN (1987) Nucleic Acids Res. 15: 8531–8545. 51. Ulanovsky LE & Trifonov EN (1987) Nature 326: 720– 722. 52. Bolshoy A, McNamara P, Harrington RE & Trifonov EN (1991) Proc. Natl. Acad. Sci. USA 88: 2312–2316. 53. Pasero P, Sjakste N, Blettry C, Got C & Marilley M (1993) Nucleic Acids Res. 21: 4703–4710. 54. Marilley M & Pasero P (1996) Nucleic Acids Res. 24: 2204–2211. 55. Kramer JA, Singh GB & Krawetz SA (1996) Genomics 33: 305–308. 56. Suarez-Lopez P, Martinez-Salas E, Hernandez P & Gutierrez C (1995) Virology 208: 303–311. 57. Me´ndez J & Stillman B (2003) BioEssays 25: 1158–1167. 58. Austin RJ, Orr-Weaver TL & Bell SP (1999) Genes Dev. 13: 2639–2649. 59. Vernis L, Chasles M, Pasero P, Lepingle A, Gaillardin C & Fournier P (1999) Mol. Biol. Cell. 10: 757–769. 60. Umek RM & Kowalski D (1988) Cell 52: 559–567. 61. Eckdahl TT & Anderson JN (1990) Nucleic Acids Res. 18: 1609–1612. 62. DePamphilis ML (1993) Annu. Rev. Biochem. 62: 29–63. 63. Boulikas T (1995) Int. Rev. Cytol. 162: 279–388. 64. Roberge M & Gasser SM (1992) Mol. Microbiol. 6: 419– 423. 65. Brylawski BP, Cohen SM, Cordeiro-Stone M, Schell MJ & Kaufman DG (2000) Crit. Rev. Eukaryot. Gene Expr. 10: 91–99.