I ANALYTICAL IIOCHEMISTRY
`
`Methods in the Biological Sciences
`
`IT’—
`
`1
`
`MTX1017
`
`

`

`ANALYTICAL BIOCHEMISTRY
`
`Volume 313, Number 1, February 1, 2003
`
`© 2003 E lseviet· Science (USA). All righ ts reserved.
`Tllis journal and the individual contributions contained in it are protected under copy1·ight by Elsevier Science
`(USA), and the following terms and conditions apply to their use:
`
`Photoco pying
`Single photocopies of single articles may be ma de for personal usc as a llowed by national copyright laws. Per(cid:173)
`mission of the Publisher and payment of a fee arc required for all other photocopying, including multiple or sys(cid:173)
`tematic copying, copying for advertising or pt·omotional purposes, resale, and all forms of document delivery.
`Special rates are available for educational i11stitutions that wish to make photocopies for nonp1·ofit educational
`classroom use.
`Permissions may be sought directly from El sevier Science Global Rights Department, P.O. Box 800, Oxford OX5
`lDX, UK; phone: (+44) 1865 843830; fax: (+44) 1865 853333; e-mail: permissions@elsevier.co.uk. You may also
`contact Global Rights directly through Elsevie1·'s home page (http://www.elseiver.com/homepage/supportl) by
`selecting "Obtaining Permissions."
`In the USA, u sers may clear permissions and make payments through the Cop)'l;ght Clearance Center, Inc., 222
`Rosewood Drive, Danve1·s, MA 01923, USA; phone: (978) 750-8400; fax: (978) 750-4744; and in the UK th1·ough the
`Cop)'l·ight Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Cow-t Road, London WlP OLP ,
`UK; phone +44 (0)20 7631 5555; fax: +44 (0)20 7631 5500. Other countries may have a local reprograpluc rights
`agency for payments.
`
`Deriv a t ive Wo rks
`Subscribers may reproduce tables of contents ot· p1·eparc lists of articles including abstracts for internal circulation
`within thei1· institutions. Permission of the Publisher is required for resale or distribution outside the in stitution.
`Permission of the Publisher is required for all other derivative works, including compilations and translations.
`
`E lectronic Storage O l ' Usa ge
`Permission of the Publisher is required to store or use electronically any material contained in this jorn·nal,
`including any article or part of an article.
`Except as outlined above, no part of this publication may be reproduced, stored in a retrieval system, OI' trans(cid:173)
`mitted in any form or by any means, electronic, mechanical, photocopy'illg, recording, or othmwise, without prior
`written permission of the Publisher.
`Address permission 1·equests to the Global Rights Department at the mail, fax, and e-mail addresses noted above.
`
`Notice
`No responsibility is assumed by the Publisher for any injury and/01· damage to pm·sons or property as a matter of
`product liability, negligence, or otherwise. or from any use OJ' operation of any methods, products, instl'uctions,
`or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular,
`independent ve1ification of diagnoses and drug dosages should be made.
`Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this pub(cid:173)
`lication does not constitute a guar antee or endorsement of the quality or value of such product O l' of the claims
`made of it by its manufactlli'er.
`
`0003-2697/03 $30.00
`
`Publicatio n infm·mation: Analytical BiochcmistiJ' (lSSN 0003-2697) is published semimonthly by Academic Press, an
`imprint of Elsevier Science, 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA. For 2003, Volumes 312-323, 24
`issues, are scheduled fo•· publication. Subscription prices are available upon request from the Publisher, from the Regional
`Sales Offices nearest you, or via the Elsevier Sdence Web site, www.elsevier.com. Subsci·iption orders are ente•·ed for the
`calendar year and are payable in advance. POSTMASTER: Send changes of add1·ess to Elsevier Science, Regional Sales
`Oilice, Customer Support Department, PO Box 945, New York, NY 10159-0945, USA. Periodicals postage paid at Orlando,
`FL, and at additional mailing offices.
`
`Orde t·s, claims, and product inq uit'ies: Please contact the Customer Support Department aL Lhe Regional Sales Office
`nearest you. New Yot·k: Elsevier Science, PO Box 945, New York, NY 10159-0945, USA; phone: (+1) (212) 633 3730 [toll(cid:173)
`free number for North American customers: 1-888-4ES-INFO (437-4636)]; fax: (+1) (212) 633 3680; e-mail: usinfo-f@else(cid:173)
`vier.com. Amsterda m: Elsevier Science, PO Box 211, 1000 AE Amsterdam, The Netherlands; phone: (+31) 20 4853757; fax:
`(+31) 20 4853432; e-mail: nlinfo-f®clsevier.com. Tokyo: Elsevier Science, 9-15 Higashi-Azabu 1-chome, Minato-ku, Tokyo
`106-0044, Japan; phone: (-t81) (3) 5561 5033; fax: (+81) (3) 5561 5047; e-mail: info@clscvier.co.jp. Singapore: Elsevier
`Science, 31Giliney Road, #08-01 Winsland House I, Singapore 239519; phone: (+65) 6349 0200; fax: (+65) (1733 1510; e-mail:
`asiainfo@elsevier.com.sg. R io de J a neiro: Elsevie1· Science, Rua Sete de Setembro 111/16 Andar, 20050-002 Centro, Rio de
`Janciro-RJ, Brazil; phone: (+55) (21) 509 5340; !>1x: (-t55) (21) 507 1991; e-mail: elsevier@campus.com.br. [Note Latin America:
`~'or orders, claims and help desk information, please contact the Regional Sales Office in New York as listed above.]
`
`§ The paper used in this publication meets the requirements of ANSIIN ISO Z39.48-1992 (Permanence of Paper)
`
`2
`
`

`

`Available online at www.sciencedirect.com
`
`SCIENCE @DIRECT ®
`
`Analytical Biochemistry 3 13 (2003) 128- 132
`
`ACADEMIC
`PRESS
`
`ANALYTICAL
`BIOCHEMISTRY
`
`www.elsevicr.com/locatc/yabio
`
`High-performance liquid chromatography purification of 26-bp
`serial analysis of gene expression ditags results in higher yields,
`longer concatemers, and substantial time savings
`
`Mette Damgaard Nielsen,a·*· 1 Mark Millichip,b and Knud Josefsena. l
`
`" Bart/to/in lll.~titullet, Unit•ersity Hospital. Bal'llwlinsgade 2. 1356 Copenhagen K. Denmark
`b Polymer Laboratories. £5sex Rd., Clturclt Srrel/on. Shropshire S Y6 6A X, UK
`
`Received 14 August 2002
`
`Abstract
`
`fn contrast to DNA chips, serial analysis of gene expression (SAGE) is not dependent on genes having been previously identified
`for their monitoring. Although useful. the method can be technically challenging, and particularly the last steps including con(cid:173)
`catenation and cloning may result in less than optimal results. We propose that many of the encountered problems can be attributed
`to the purification of the 26-bp ditags by polyacrylamide gel electrophoresis. Low yields, gel contaminants. potential exposure to
`degrading enzymes during handling and lengthy separation all disfavor the method. We introduce purification of 26-bp ditags bv
`reverse-phase high-performance liquid chromatography (HPLC) using polystyrene/divinylbenzene columns and tetraethylammo·
`nium acetate buffer with acetonitrile as mobile phase. The method is fast and gives excellent results. Ditags purified by HPLC readily
`ligate to high-molecular-weight concatcmers leading to their emcicnt cloning. The method should substantially l~tci lirare the con·
`struction of SAGE libraries.
`© 2003 Elsevier Science (USA). ;\II rights reserved.
`
`Keyll'ords: HPLC: SAGE: Ditags: Polystyrene/divinylbenzene: Purification
`
`Despite the popularity of DNA microa rrays, SAGE2
`is widely used in differential gene expression analysis as
`this techn ique holds unique strengths [1.2]. First, SAGE
`is based on the serial sequencing of concatenated eDNA
`tags obtained through the isolation of uniq ue short
`transcript tags from all expressed genes in a cell popu(cid:173)
`lation of interest. SAGE data hence represents absol ute
`mR NA expression levels and can readily be compared to
`an ever-growing number of SAGE databases (www.sa(cid:173)
`genet.org; www. ncbi.nlm. nih .gov/SAGE; www-dsv.cea.
`fr/t bema/get/sade. ht ml; http://genome-www4.sta tlford .edu/
`egi-bin/SGD/SAGE/quetySAGE; http://bioinfo.amc.uva.nl/
`HTM-bin/index.cgi/) or to data sets obtained from
`
`• Corresponding author. Fax: +45-33938566.
`E-nwil address: melledn@rh.dk (M . Damgaard N iel sen).
`1 Contributed equally to the work.
`1 Abbreviatiom used: SAGE, serial analysis or gene expression; PS/
`DVB. polystyrcnc/divinylbenzenc: TEAA. triethylamminium acetate:
`ACN . acetonitrile: PCI. phenol:chloroform:isoamylalchohol.
`
`corresponding cell populations under different physi·
`ological conditions. Second,
`in contrast
`to micro·
`arrays, SAGE is not dependent on genes having been
`previously characterized. SAGE is, however, techni(cid:173)
`cally dema nding as it consists of a series of critical
`steps. The final steps in the generation of libraries are
`especially challenging, since transcript-tag concatemers
`in adequate quan tity and length can be hard to obtain
`and since difficulties in cloning concatemers of satis·
`fying size have been noticed [3,4]. To achieve good
`concatenation it is impo rtant to efficient ly remove the
`liberated link ers as their presence in the ditag liga·
`tion reacti on will block any further polymerization of
`the concatemers and impair their subsequent cloning
`[4].
`Trad itionally, in preparation for the ligation of ditags.
`the latter have been purified by polyacryla mide gel ex·
`traction. Nonetheless, this method is not particular!}
`efficient, since excision of bands of defined sizes does not
`exclude the presence of other molecular weight species
`
`0003-2697103/$- see front matter © 2003 Elsevier Scicnc.:e (USA). All rights reserved .
`Pll : S0003-2697(02)00511 - 0
`
`3
`
`

`

`ill. Dmugaard Nid\C'II el a/. I Aua~rlical Biachemi11rr 313 (]003 ) 118 /31
`
`129
`
`and since gel contaminants including acrylic acid are
`known to be potent inhibitors of enzyme reactions [II).
`Other disadvantages include the temperature sensitivi ty
`of the 26-bp short ditags [6], which requires cooled gel
`running conditions, and loss of material duri ng the
`process as general experience shows that complete re(cid:173)
`covery from polyacrylamide gels is difficult to achieve.
`Finally, the many handling steps increa se the likelihood
`of DNase contamination. fn summary, PAGE purifica (cid:173)
`tion, in add ition to not being especially emcienl in itself.
`could be a contributing factor to subsequent ligation and
`cloning problems.
`We describe a purification method of 26-bp SAGE
`ditags based on HPLC that circumvents all of these
`problems. We show that th is procedure substantially
`enhances the quantity and length of concatemers, lead(cid:173)
`ing to their efficient cloning.
`
`Methods
`
`SA GE
`
`In our work we use a combination of SAGE adapted
`for small sa mples [7] and the Invitrogen SAGE protocol
`(http://www. i nvi trogen .com/Conten t/W o rld/sage_ma n.
`pd f.pdf) with two modifications [3,5]. The 102-bp ditags
`were amplincd in 200 x 50-~LL PCRs using l ~LL of I: 100
`dilut ion of ditag ligation reaction, isolated by 121X•
`PAGE, and purified using QIAquick silica membrane
`columns (Quiagen, Hi lden, Germany) [5]. The 26-bp
`the
`eDNA ditags were subsequently released from
`flan king linkers through digestion with N/aiii (New
`England Biolabs, Frankfurt am Main, Germany) in two
`150-~tL reactions using a total of 240 U of enzyme for
`1.5 h at 37 °C. The reactions were pooled and cooled on
`ice, and the entire reaction mixture was loaded directly on
`Lhe HPLC column.
`
`HPLC
`
`A Hewlet Packard 1100 HPLC system equipped with
`a 150 x 2. 1-mm column containing polystyrenc/divinyl(cid:173)
`benzcne (PS/DVB) reverse-phase col umn (5).lm particle
`size. 100 A pore; Polymer Labora tories, Church Strct(cid:173)
`ton. Shropshire, UK) was used. The column was
`thermostated at l0 °C and the fl ow rate was 0.2 mL!min .
`Buffer A was I 00 mM triethylammjnium acetate (TE
`AA; Applichem, Darmstadt, Germany), l mM ethylene
`diamine tetraacctic acid (EDTA ; Sigma , St. Louis, MO,
`USA), pH 7.0; buffer B was buffer A+ 90(Yo acetonitrile
`(ACN; HPLC grade. Fisher Scientific, Leicestershire,
`UK). Samples were loaded in 6.25% ACN and eluted
`using a gradient of 0. 125%/min ACN over 90 min. De(cid:173)
`tection was performed using a diode array detector at
`280 nm with 360-nm background subtraction . A 25-bp
`
`DNA ladder (Invitrogen , Carlsbad, CA, USA) was used
`for initial calibration or the method and a double(cid:173)
`stranded 26-bp test-ditag (5'-TTGATGT ACAGGCCG
`CCTTCCGCATG-3', 5'-CGGAAGGCGGCCTGTAC
`ATCAACATG-3'; TAG Copenhagen A/S, Denmark)
`was used to mimic the elution time of ditags. At the end
`of each run, the column was nushed with an ACN
`gradient of 6.25 90% over 30 min , followed by 30 min of
`restabilization in 6.25% ACN.
`
`Concatenation (~! H P LC-purified 26-bp ditags
`
`The HPLC-purifiecl ditags (approximately 0.6 ~tg)
`were extracted by phenol:chloroform:isoamylalchohol
`(PC I, 25:24: I ; Applichem, Darmstadt, Germany) and
`ethan ol precipi ta ted. After resuspen sion in 6 ~tL LoTE
`(3 mM Tris- HCI pH 7.5, 0.2 mM EDTA) concatena(cid:173)
`tion was performed in a total volume of I 0 ~tL using
`10 U of T4 DNA ligase (High Concentration; Invi(cid:173)
`trogen) at
`l6 °C for 3h. The resulting conca temers
`were analyzed on 8% PAGE, excised in size sections,
`and el uted overnight at 4 oc followed by 65 °C, 15 min.
`One ).lg of pZErO vector (rnvitrogen) was digested in a
`total vol ume of 25 ~t L using 15 U or Sphl (Invit rogen)
`for 25 min at 37 °C, purified by PCl extraction , ethanol
`precipitated , and resuspended in 60 ~tL LoT£. The
`purined concatemers were cloned into approximately
`15 ng of pZErO vector and 116 of the reaction mixture
`was transformed into 50 ~tL DH I OB bacteria (Invitro(cid:173)
`gen) using electroporation at 2.2 kV, 400Q (Gene
`Pulser; Bio-Rad, Life Science Research, Hercules, CA,
`USA). Transformants were plated onto low-salt LB
`plates containing Zeocin (Invitrogen). isopro pylthio-~­
`o -galactosicle,
`and
`5-bromo-4-ehloro- 3- i ndolyl-~-o­
`galctoside (X-GAL) for selection of positive clones
`[ 12].
`
`Results
`
`To circumvent the disadva ntages of gel purification
`we developed a method to purify 26-bp SAGE ditags
`using 1:-IPLC. To optimize the method, 2).lg of a 25-bp
`ladder was loaded on the column. Pea ks were collected,
`precipitated. and analyzed by 12% PAGE using I x
`TBE, pH 8.3, as buffer. As shown in Fig. I the lower(cid:173)
`molecula r-weight DNA fra gments resolved excellently,
`whereas fragments above I 00 bp eluted collectively in
`one peak. PS/DYB columns of different pore a~1 d par(cid:173)
`ticle sizes (300 A/5 ~tm , I 000 A/8 pm, and 4000 A/8 ~tm)
`produced similar results with better resolut ion for the
`higher-molecular-weight DNA fragments (data not
`shown). Since the resolution using the I 00 A pore 5).lm
`column was excellent for 25-bp fragments, which is also
`the size of the liberated eDNA ditags, this column was
`chosen for fu rther work.
`
`4
`
`

`

`130
`
`M. Damgaard Nit: I sen er a/. I Analytical Biorhemislly 313 ( 2003) 128-/32
`
`Fig. I. Two micrograms of a 25-bp double-stranded DNA ladder was fractionated o n a 100 t\15 fL111 reverse-phase column using buller A ( I 00 mM
`T EAA, pH 7.0), a nd a grad ient elution of buller B (buiTcr A+ 90% ACN) at 0. !25'Y.,fmin. Peaks were collected and a nalyzed by PAGE. Peaks
`corresponding to elution times 74, 77, 83. and 88 min were loaded in lanes 2. 3. 6, a nd 7, respectively. The remaining lanes correspond to the in(cid:173)
`termediary peaks. As control. 0.7 pg o f the 25-bp ladder was loaded directly in lane I.
`
`For purification of 26-bp ditags, the en tire 300-~tL
`crude restriction digestion reaction mixture of I 02-bp
`PCR-am plified ditags was loaded directly on the column
`and analyzed in a si ngle run. The peaks were collected
`and a sma ll sample of each (1110 of the elutio n volume)
`was a nalyzed by 121% PAGE. As seen in Fig. 2. 26-bp
`ditags eluted at 74min, whereas the two liberated "'40-
`bp linkers eluted distinctively in two separate peaks at
`78 and 8 1 min. The peaks contained approximately 0.6,
`I , a nd l ~tg of D NA, respectively, as evaluated by OD
`integration of the peak area. The 26-bp ditags peak was
`wider than the linker peaks, which could be due to
`sequence or size heterogeneity. For comparison, the
`(sequentially homogeneous) double-stranded 26-bp test(cid:173)
`ditag eluted in a much sharper peak at the same elution
`time (da ta not shown).
`influence of the purification
`To
`investigate
`the
`scheme on subseq uent SAGE protocol steps, the effect
`of TEAA on ligase was first examined. Increasing
`amo unts of TEAA were added to 0.5 ~tg HindTII-di(cid:173)
`gested lambda phage DNA (Fermentas, Hanover, MD,
`USA) containing 5U of T4 DNA ligase in a
`1 0-~tL
`
`reaction. No effects were visible in concentrations up
`to 0.5 M (data no t shown) as evaluated by agarose gel
`electrophoresis. Further, T EAA is read ily sol uble in
`ethanol. suggesting
`thereby
`that on ly
`insignificant
`amounts would be carried over to the subsequent li(cid:173)
`gation reaction. Based on these fi ndings we concluded
`that TEAA would not impair the ligation reaction . We
`next investiga ted the ligation of HPLC-purified 26-bp
`clitags . We next investigated the liga tion of 26-bp di(cid:173)
`tags. Approximately 0.6 ~tg of purified ditags was li(cid:173)
`ga ted as described. As visua lized by 81% PAGE
`analysis (Fig. 3) the resulting concatcmers were both
`much longer and present in higher yields than con(cid:173)
`ca temcrs previously obtained from PAGE-purified 26-
`bp ditags originating from the same amount of starting
`material.
`Not surprisingly, the higher quality concatemers also
`produced clones with longer inserts. I n o ur earlier trials
`with PAGE-purified ditags we did not o btai n clones
`with concatemer inserts larger than around 200 bp. To
`investigate the cloning efficiency of the HPLC-purified
`ditags, the inserts of clones generated from two different
`
`2
`
`3
`
`4
`
`5
`
`bp
`
`50
`
`20
`
`10
`
`Fig. 2. Nla lll restriction digest ion of PC R-amplified 102-bp ditags was purified and analyzed as in Fig. 1. Approximately 4 ftg of ditags was loaded
`directly in 300pL of reaction mixture. The content of peaks at 74, 78, and 8 1 min were analyzed in gel lanes 2. 3. ~nd 4 . Lanes I and 5 were 10- a nd
`25-bp ladders, respectively. Major peaks corresponded to the sizes of liberated ditags a nd the two linkers A a nd B.
`
`5
`
`

`

`,\/. Damgaard Nielsener a!. f rlfla~rricaf Biodll!misrry 1 13 (2001 ) 128 112
`
`131
`
`2
`
`3
`
`4
`
`bp
`
`!:00
`
`125
`
`50
`
`25
`
`Fig. 3. Results of concatenation of26-bp d itags purified by 12"/., PAGE
`(left ) and IIPLC (right). Equa l amounts of starting material were used
`in both cases. The use of HPLC-purificd ditags resulted in conc.:atemers
`in higher recovery and of substan tially higher molecular weight.
`
`conca temer size ranges were analyzed. Eight clo nes from
`each o f the ranges 300- 800 bp (small) and 800 bp 1.5 kb
`(large) were analyzed. Four clones in each group were
`white, two were light blue, and two were dark blue. PCR
`amplification of inserts using M 13 forward and reverse
`primers revealed tha t all clones contained inserts and
`that the inserts were around 500 bp for the small-size
`concatemer s and 150 bp for the large ones, rega rdless of
`colony colo•~ This inverse relatio nship between concat(cid:173)
`emer and insert sizes is a well-known phenomeno n in
`SAGE [3,4]. We did not find a ny empty clo nes, which
`fu rther suggests that special selection to exclude clones
`without insert can be omitted.
`
`Discussion
`
`We here introduce a reliable, a utomated , ch romato(cid:173)
`graphic purificatio n method for 26-bp SAGE ditags.
`HPLC purification represents a vast improvement over
`conventional purification methods. Not o nly does it
`result in substantially higher yield and length of con(cid:173)
`catemcrs, it a lso reduces hand ling time. Others ha ve
`previously demonstra ted the separation of DNA frag(cid:173)
`ments by ion-pair reverse-phase HPLC (8- 10]. but no
`satisfying separa tio ns of seq uences as short as 26 bp
`were shown. This was, however, successfu lly d emon(cid:173)
`strated here, thro ugh the use of PS/DVB reverse-phase
`colum ns.
`Ditags have traditionally been purified by PAGE, but
`it is well known that the subsequent concatemer poly(cid:173)
`merization often docs not produce optimal results. We
`propose that these difficulties a re ca used by contamina n ts
`from the polyacrylamide as these inhibit subsequent en(cid:173)
`zyme reactions (5]. A lthough such contaminants can be
`
`this
`ion exchange chromatography (5]
`removed by
`method is not applicable to 26-bp ditags since the sma ll
`fragments do not elute from the column. Rever se-p hase
`HPLC does not suffer fro m this d rawback and provides
`rapid and accurate purifica tion.
`W e observed that a small fraction of the loaded D N/\
`was retained o n the column fo llowing the gradient elu(cid:173)
`tion, although it was easily removed by flushing the
`column with h igh-concentration ACN. It is o bvio usly
`a bsolutely essential that the column is empty before a
`new sample is loaded, which is why we recommend
`raising the ACN concentra ti on to 90% at the end of each
`run.
`T he P S/DVB reverse-p hase media provided good
`resolution in the required DNA fragmen t range. The
`elution of liberated lin ke rs from the ditags occurred as
`two distinct peaks, demonstra ting that the column is
`capable of fractionating double-stranded o ligonucleo(cid:173)
`tides of as little as 2 bp in length difference. The wid(cid:173)
`ening of the peak containing the heterogeneous ditags
`compared to the sharp peak o f the sequentia lly hom o(cid:173)
`geneous 26-bp test-d itags suggests that the column also
`fractionates according to DNA sequence. A more likely
`explanatio n, on the other hand , could be the tendency of
`the type lls restriction enzyme, BsmFJ , used for the
`tagging reaction, to liberate tags var yi ng with I bp, thus
`giving rise to d itags ranging in length from 26 to 28 bp .
`The widen ing of the peak con tai ni ng di tags is not a
`problem, however, since they still elute in an easily
`managea ble volume (~300 ~t L). The sample size was
`brought down through PCl extraction followed by
`ethanol precipitation, but the la tter is probably sufficient
`in itself.
`Based on the PAGE analysis of the elution peaks
`(Fig. 2), it appears that the 26-bp ditag peak is con(cid:173)
`taminated with a faint amou nt of 40-bp linker. The
`nature of this band has not been examined, as it does
`not interfere with the efficiency of the subsequen t con(cid:173)
`catenation of the pur ified ditags (Fig. 3).
`Handling of d itags is critically dependent on tem(cid:173)
`perature [6]. M ost HPLC equipment has thermostated
`columns, but the injectio n volumes used in this method
`require installation of a special injecti on loop, wh ich is
`o ften not cooled . However, we have not ta ken specia l
`precautions to stabilize the tempera tu re of the loop and
`have not observed any problems in the subsequent en(cid:173)
`zymatic reactio ns.
`
`References
`
`[I) V.E. Velculesc.:u. l... Zhang, B. Vogelstcin, K.W. Kinzler, Science
`270 (1995) 484-487.
`[2] K. Polyak , G.J. Riggins, J. Cl in. Oncol. 19 (200 1) 2948- 2958.
`[3] M. Kenzelma nn. K. M uhlemann. Nucleic Acids Res. 27 ( 1999)
`9 17- 918.
`[4] J. Powell. Nucleic Acids Res. 26 (1998) 3445- 3446.
`
`6
`
`

`

`132
`
`M. Damgaard Nielsen el a/. /llnalyliml Biochemisny 3/3 (2003) 128-132
`
`[5) J.M. Augelastro. L.P. KJimaschewski. O.V. Vitolo, Nucleic Acids
`Res. 28 (2000) E62.
`[6) E.H. Margulies, S.L. Kardia , J.W. Innis. Nucleic Acids Res. 29
`(200 1) £60~0.
`[7) L. Cheval, B. Virion. E. Billon, J.C. Aucle. J.M . Elalouf. A.
`Doucet, J. Nephrol. I5 (2002) SI70-83.
`[8) K.H. Hecker. S.M. Green, K. Kobayashi. J. Biochem. l3iophys.
`Methods 46 (2000) 83-93.
`
`[9) K .H. Hecker, B. Turpie, A. Kuklin. Biotechniques 26 (I999) 2I6-
`2 18. 222.
`[IOJ H . Oberacher, A. Krajete, W. Parson. C.G. Huber, J. Chroma(cid:173)
`togr. A 893 (2000) 23-35.
`[II] G.E. Etokebe, A. Spurkland, Biotechniques 29 (2000) 694,
`696.
`[12) J.M. Angclastro. E.J. Ryu, B. Torocsik, B.K. Fiske, L.A. Greene,
`Diotechniques 32 (2002) 484. 486.
`
`7
`
`