CA2161767A1 - In vitro transposition of artificial transposons - Google Patents

Abstract

We have developed efficient methods of creating artificial transposons and inserting these transposons into DNA targets in vitro, primarily for the purpose of mapping and sequencing DNA. A target DNA has been engineered to convert virtually any DNA sequence, or combination of sequences, into an artificial transposon; hence, custom transposons containing any desired feature can be easily designed and constructed. Such transposons are then efficiently inserted into DNA targets, in vitro, using the integrase activity present in yeast Ty1 virus-like particles. Primers complementary to the transposon termini can be used to sequence DNA flanking any transposon insertion.

Description

4n45Pcr~PP

IN VITRO TRANSPOSlTION OF ARTI~ICIAL TRANSPOSONS

BACKGROIJND OF T~IE INVI~TION
DNA seqllPnt~ing has helped l~o~ ;oni7e the way that genes and genomes are sh~di~, and has led to a greater undçrst~n~ing of most aspects of biology.
Nevertheless, with efforts underway to map and sequence the genomes of a varietyof org~nicmc; the need to improve the efficiency of DNA sequen~ing has never been greater (1). One of the major problems associaled with sequ~ncing large se~mPntc of DNA is ob~ining sequence inf,- ... ql ;on beyond the limits of a single primer e tPnci~nn event. Several techniques are ~;U~ y used to acquire sequenceswithin the interior of a DNA insert; these include: i) the synthesis of custom primers to "walk~ along a ~g....~nl of DNA (2, 3), ii) shotgun subcloning, whichil~s a high degree of l~~ for complP~, sequence leco~ (4), or iii) the cons~uction of ovc~lappulg P~o~ cl~ce dtl~fiOn clones (3, 5). Each of thesemPths~s is time~o~cllming~ idiosyncratic and ther~Çole difficult to aulolllale, and/or costly.
..7l;vcl~, l.,~n~us~Ie el~..enl~ have been adapted for DNA mapping and sequ~Pnc~ng. P~mplp-s incllldç: ~yâ (6), Tn5 (7), TnlO (8), as well ac deliv~cs of these and other trans~030l s. Although these approaches g~onerally offer great promise, the insertion step is ?elrolllled in vivo in E. coli; hence, t,~r,;,~,ilion may occur ir~to either the p1~mj~ target or the E. coli genome, comrli~ting the l~Co.~.~ of target insertions. An ~litinn~l ~iff~ .lq arises from host effects on insertion r~ndomn~ss, i.e., ~hotspots~ and ~coldspots~ of int~ ;on are often ob~l~ed in vivo (9).
The complete DNA h~ l;on reaction employed by certain retroviruses and rel~ol~ osons as part of their normal life cycles can be carried out c~ e~1~ in vitro (1~14) orr~ ;ng a possible ~llf~ e to in viw l.dns~oson insertion techniques for DNA sequenci~.
There is a need in the art for a simple, reliable technique for g~l~ting sets of DNA t~ latcs for seqll~nt ing any target. In particular there is a need for sets of DNA temrl~t~s which are amenable to au~l-d~i seq~en~ing with a single set of prim~s, SIJMMARY OF T~IE INVENIION
It is an object of the invention to provide meth~l~ for providing le~
for DNA Se~u~n-~.;ng It is another object of the invention to provide . . ~I.~s for sequen~ng such DNA t~ ..l lAt~s, It is yet anotll.,~ object of the invention to provide a kit for DNA
sequ~n~i~.
It is yet an.~lh~ object of the ill~ ion to provide an ar,tificial ll~;.~son.
It is still ano~ object of the i.l~.Lon to provide p1q~mi-1~ for p~hing artificial l-,.n~ n~
It is yet anotll~ object of the invention to provide methods for the gener~tion in vitro of ins~ ns into a target DNA mol~~
These and other objects of the invention are provided by one or more of the e..-bo~ of the ill~ ion dç~ il~ below. In one embodiment a method is proYided for ~epaling te...l.l~t~ s for DNA se~u .~in~,. The method co...l.. ;~s the steps of:
incub~t; ~g in vitro: (1) a population of a target DNA, said target DNA compri~ing a region of DNA to be sequen~d, (2) a retrovirAl or re~ nspo30n integra~e, and (3) an artificial transposon having two terrnini which are s.~bstldt~c for said ih~ ~, wh~cin the molar ratio of artificial Ll~ns~oson to target DNA is at least 1:1, to forrn a population of target DNAs with quasi-randomly ;nteE.AI~ insertions of the artificial transposon;
l.~r,srolllling host cells with the population of target DNAs with quasi-randomly in~.g,AI~ insertions of the artificial transposon;
s~l~tin~ those host cells which have been tran~ ,ed with a target DNA with an insertion of the artificial l-dhs~son;
icnl~ting target DNA with an insertion of the artificial transposon from those host cells which have been ll~lsr~.ll.ed with a target DNA with an insertion of the artificial transposon, said target DNA with an insertion of theartificial transposon being suitable for use as a DNA sequ~n~-ing temrl~tç.
In another embodiment a method is provided for sequen~ing DNA. The method comrri~s the steps of:
in~ bAI;ng in vitro (1) a population of a target DNA, said target DNA co--,~-;C;ng a region of DNA to be seque-nc~d, (2) a l~llu~ us or lclr~tl~s~sùll inl~g~ and (3) an artificial ~n~pos~n having two termini which are ;,..l,st~ s for said into.~, W~ ~ the molar ratio of ar~ficial t~ansposon to target DNA is at least 1:1, to form a ~pula~ion of target DNAs with quasi-randomly ;n~.~-~t~ insertions of the artificial ~ o3~n;
I I A r, C r~ .; ng host cells with the po~ulalion of target DNAs with quasi-r~An~lomly int~.~,Al~d insert;ons of the artificial l,~ ~n;
sel~in~ those host cells which have been ~ r~lllled with a target DNA with an insertion of the artificial tl~n~soQ;
i.colAI;ng target DNA with an insertion of the artificial transposon from those host cells which have been n~crvllllpd with a target DNA with an ion of the ar~ficial ~ n, said target DNA with an insertion of the artificial transposon being suila~le for use as a DNA se~ nc;ng te.ll~l~le, hybri-ii7ing to said isolated target DNA with an insertion of the artificial I .~r. ~oson a primer which is co...l)lP ..tonhry to a tf ~...;nus of the artificial transposon;
e~tending said primer to det~-rminP a nu~l~oti-le sequence of DNA
fl~nking said artificial transposon in said i~l~ted target DNA with an insertion of the artificial transposon.
In still another embodiment of the invention a method for seql~n~ing DNA
is provided. The method comrri~-~ the steps of:
providing a population of target DNAs with quasi-~ndc-mly inleg.dted insertions of an artificial trAn~oson, said artificial transposon having termini which are substrates for a retrovirus or a lellv~l~nsposon, said population of target DNAs having been formed by in vitro insertion of said artificial transposon into the target DNAs using a ,~ ovildl or ~ ull~nspo3011 in~ ~, and a molar ratio of artificial l-~n.~o~n to target DNA of at least 1:1;
hybri-1i7ing to individual target DNAs of said population a primer which is co---l)lP~--e~ to a tr- .~n.Js of the artificial trAn~ n;
e~te-n-ling said prima to det~- ...;ne a r...~ e sequence of target DNA fl~nking said artificial t.~r.s~oson.
In still anotl,~ embo~im~nt of the invention a kit for DNA s~u~ is provided. The kit cornpri~, an artificial ~ os~l having termini which are sul~s1.,.trs for a ~LIuvil~l or l~l.iln.~l~son ;n~g.~.~;
a l~llU~ l or l~o~ n~l o5Qn int~g~e;
a buffa for in vi~ro transposition of said artificial l~ o~n, said buffer having a pH of 6 to 8 and 1 to 50 mM of a divalent cation; and a pnmer which is compl~...c~.l;..~ to a te ...;....s of said artificial n;,~oson.
In an ~litionql e...bodill-ent of the invention an artificial ~pos~n is provided. The trqnq~oso~ consists of a linear DNA mo'ocllle compri~ing:
a marker gene;

a se~uence of yeast ~ ansl~oson Tyl, said sequence s~
from the group c4r~ tine of a U5 s4uence and a U3 soquence, said s.~.lu~
fl~nking said marker gene on its upstream end, said s~ucnce Con~ g of 4 to 11 bp of l.c.".;n~l sequences of said Tyl; and a sequence of yeast lc~ h~fi.~ n Tyl, said sequenc~e ~l~h~d from the group eonsi~tir~ of a US sequenc and a U3 sequence, said sequence fl~nking said marker gene on its dow~ ~ull end, said s~u ~ce co~cicti~ of 4 to 11 bp of te ...;n~l sequences of said Tyl.
In yet an ~l~ition~l embodiment of the invention a DNA molecule useful for genPr~ting artificial ll~r,s~sons is provided. The DNA moleol)le comrri.~P,s:
an origin of reE1li~tion;
a first sP~t~ble marker DNA;
two blunt-ended ~ n~oSOn termini of at least 4 bp each, said termini being s~Jl,sl~t~s for yeast l~!.,.n.~l~son Tyl i~ L~ said ~ os n termini fl~nkin~ a first restricti~n e~ ",c site useful for insertion of a second seleet^' le marker gene to form an ~ ial !.i.r.~03~n;
a second restrietion e~ ,c site fl~nkine said two ~ ~sn n tP-rmini~ c~cin llig~cti~n with said second restri~-t~ IIC libe.~tr5 a blunt-ended r.,.g~ nl having said l,~son termini at dther end of the r.~;,....-~ the fi~;",enl thereby lihe~ being an artificial l.,.r..~o~n.
In still another c.llbo~ 1 of the invention a meth~ for in vi~o g~ ;on of insertions into a target DNA is provided. The m~tho~l Co~ ;~S
the steps of:
incub~ in vi~ro (1) a pop -l~ti~n of a target DNA, (2) a ~u~l or ldlul,~sposon inl~g.,.~, and (3) an artificial ~ os.~n having termini which are ~bsl.~1f5 for said ;n~.".~., ~.h~ "n the molar ratio of artificial 1.; fi~l-03nn to target DNA is at least 1:1, to form a population of target DNAs with quasi-randomly ~ .p..~ed insertions of the ar~ficial ~ sl)oson;
ll~hsrollllillg a host cell with the population of target DNAs with quasi-randomly integldted insertions of the artificial ~ po30n;

Xting those host cells which have been transformed with a target DNA with an insertion of the artificial Ll~ns~son.
The in vitro ~ste.~ls of the present invention offer several ad~ge~s over in vivo tr~n~positiorl systems: i) special b~t~-n~l strains are not ~ uilc;d, ii) polc.,tial host effects are avoid~, and iii) an in vitro reaction is ~men~ble tobioch~mi~l ~ltp~tion and p~mP~Pr opl;...;,~;o~. Thus a simple and reliable method is provided for gen~ l;ng large ~ ou~L~ of sequence in~o~ tion, such as iS l'~UilCd for sequPn~ing of entire g~no,l,es of particular org~nicms.
BRIEF DESCRImON OF T~IE DRAWINGS
Flgure 1. O~ .. of artificial l~ s~-on insertion into pla~nid targets.
The basic steps involved in ge-n~ ;ng artificial l~n~posol~ insertions in target pl~mids are in~ t~ Note the following: DNA sequences to be de~ ...mPd (dashed line) trim~thoprim r~ir~nce (tri') gene (shaded bo~c); targetpl~mid (double circle); PART (primer island artificial I~An~Gson) (box); Tyl U3 termini (filled rec~nglPs).
Figure 2. pAT-l and pAT-2.
Fig 2A. The b~^~ne co~ .nn to pAT-1 and pAT-2 is shown to contain the yeast UR~3 gene, a b~r~ ;31 origin of r~li~tinn (ori) and a mnlti~l~ni~ site(mcs). pAT-2, co..li-;n.~ the PART insert, is depicted. Fig 2B. The PART
which is created upon digestion with ~inn I, is shown. It cort;-;n.~ the dhfr (dih~ urolate ~luc~-~,) gene (stippled), the pBLUESCRIPI mcs (white bo~ces), and Tyl U3 c~s..lt~s (filled ~ ~), as well as two unique primer sites for s~uen~in the DNA fl~nkir~ an insertion site. Fig 2C. The sequence at Tyl U3/
Xinn I C~ 5. The arrows in~i~te the ~:inn I cl~ ge site. The shaded areas in~lie~te Tyl U3 sequences (one on dther side of the arrows), while the entire se~lu~ue en~es a l~n;l;~-n site for ~nn I.
~gure 3. PART insertions in done p7~2.
The 8 kb insert of clone p7~2, co~ ;n~n~ a ~Eg-~ of yeast chromosome m, is shown along with the sites of 78 ind~ndent PART insertions (arrows).
The ori~ont~tion of L,~s~oson insertion is in~ ted: ( ~ ) Forward (the dhfr gene in the artificial t~ poson is transcribed left to right, or ( t ) Reverse. This region of chr. m c~nl~in~d on the insert incl~des the PGK 1 gene (black bo~), a glycine tRNA gene (black circle with allu~l,ead in-~ir~ting direction of iplion), a Tyl solo delta (stippled bo~) and the YCR16w locus (striped bo~c).
The PART insertion loc~tinn~ were d~t~.. n~d by sequenring one or both insertion jnnr,tion~, F1gure 4. Conceptual contig map.
The loc~tionc of the 78 PART insertions were used to construct a concc~tual contig map based on the following assumptions: i) two primer ertencions would be initi~t~ from each PART (one in each direction) and ii) eachexten~ion would lead to the ~.~,~ of 250 bp of useful DNA sequence information.
Fgure 5. Interval Sizes of PART insertions into p7C-2.
The size of intenrals b~h.oc~ individual insertions of PART into p7~2 (i.e., the llict~nr~ l~h.een ~ ~nt ins~ ns in bp) were gruu~d and the nul"bc~
of intervals falling within each group is ~rhir~lly l~,p~ t~d.
~gure C. Distribution of PART insertions in plasmid pWAFp.
Plasmid pWAFp ~ ;n.c a S kb insert of human DNA &n~4~l;"g the WAF-1 plvlllGt~,~. We ~ at~ PART inse,.Lo~c into this targd using an artificial L,~posvn ~ d by PCR and l;gPv-~Qn with Bbs I to e~n ~e U3 and U5 s~.~enccs at the u~.t~-. and dv~.--st~ ends of the ~ tos~n, l~1i~_1y.
Of 45 ~l~Lvns analyzed, 12 ..al)ped to the pBLUESCRIPI vector ~l~,lll.,~
(shown in black), 13 ~ p~i to the 1.5 kb Not I/Pst I r.~,..~n~ of the WA~F-l insert, 12 I-.Al~.d to the 2.5 kb P~t I r.~,,... nl of WAF-l (WAF-l sequences are ,colid white). Hence, insc~lions were ~ d from all regions of this target pl~mi.1, and the ins~Lvn rl~u~ ~v ranged from 4.1 insertions/kb to 10 insertions/kb targd DNA. This sd of insertions was then used to dilc-,lly ~.
greater than 90% of the WAF-l DNA se~u~ce.

~lgure 7. Distribution of insertions into yeast chromosome m.
An ~lir~ ns~o~n having one U3 and one U5 t~ s, each 4 pb in length, was ge.~ t~ by PCR, digested with Bbs I, and filled-in with Klenow fragment of DNA pol~ l~ I. Distribution of insertions are shown on a map of the clllo..~oso...~ m ~eet.,....-t of DNA co~ infd on the target pl~mi~
~igure 8A~B. The nucleotide sequence of pAT-1.
F~gure 9A-9B. The nucleotide sequence of pAT-2.
~Sgure 10. The nudeotide sequence of the PART from pAT-2.
~lgure 11. Sequence contig map for 8 kb region of cosmid F13544.
169 independellt AT-2 insertions were gen~ ~ted in the cosmid F13544 by in vitro in~ep. .~;on. A c~ll~tirJn of 43 insertions which were found to map to an 8 kb region by restrir-ti~ n l..Al ph~g were assembled and sequenced using p~
SD118 and 119 in cor,jwlclion with ABI Prism t~hnl logy. A contig map of the sequenr-ing project is indic~t~ Each a~ow rc~rescnls a single primer e~tPnQio~
event. ~n.~lh is a map of se~lu~ncc completion. Black areas in-lir~te sequence on both str~n~ ~h~s h~t~ d areas are on one strand only.
FSgure 12. Artificial transposons.
Eight di~f~ cial ~ c30ns, inr~ the AT-l s~u nce and structure, are shown. Each was derived from either pAT-l or pAT-2, and is ep~od from its plq~mi~ with the same Xmn I sl,~y used for these p1~mitl~

DETAn ~ DESCRIPIION OF T~; PREFFRR~l EMBODIMENIS
It is a dis~.~l~ of the present invention that a ll~ucs~losol insertion technique that is carried out e~ in vimo may be applied to a variety of problems, inclu~ine DNA sequencin~. This technique employs artificial spo~nc which are created using a pl~mi~ construct, and l~hovil~l or )t-~fi~l~03~ n int~ ~, which may be provided in the form of viral or virus-like particles (VLPs), which ..~ l;AIe s the ins~lLion of these ll~ul;,~osol s into target DNA ~ llPs.

`- 21 61 767 g We have developed new Ill~llods for creating artificial trans~osons and effi~ tly inserting these ll~.s~sons into DNA targets, in vi~o. There are three key aspects of the lJr~SS: i) tne in vitro in~ ;on reaction is highly P-ffici~ntgiving rise to ~ )u~n~s of inl~.glAt;on.C per re~lction; witn most plasmid targets, this çffi~ ncy apploaches one insertion per pho~h~liPst*r bond, ii) tne insertion process is sllfficiontly random that n~r..~l~soll in4~.EI~;rn~ occur throughout target p1q~mid s~quen~, and iii) vir~ually any DNA sequence or combination of sequences can, in principle, serve as an artificial ll~r,s~oson. These three featu~s co"lbine to make this an ~AL~ ly versatile method of ~rnc A~;ng lc~",bina~t DNA molecules.
Artificial l,~nsposons are ideal for DNA s~,en~ g i) a large nu,ll~r of transposon inær~ions can be easily ~æPmhl~ from a single in~.~lAl;on re ~.ti- n,allowing the reco~ of inærtions suitably spaced to f~iliht~ sequçn~ing of a DNA set,l~ -t, ii) the nAn~llocon can be ~gin~.red to contain desired fe~lules uæful for DNA mapping or soquçnl~;ng~ and iii) since each ll~ oson carries two unique primer sites, the m~cl~tide sDqu~ce fl~nking each insertion site can be rapidly and effi~ i~P.ntly det~F - Ill;nPd, A set of Fl~cmi~ bearing ar~ficial l.,.fi.c~oson insertions are Pq~i~lly uæful for ~ue~`ing b~ll~ all the pl~cmids can be æquenced in parallel using a defined pair of ~ This is in cP~ C~ to the inP.ffi~ nt "æriesll approach of pdmer walking, in which each se~ucncc is used to specify the ne~ct primer. Hence, ar~ficial ~ o~l~s are flexible and ~ 1C1Y
effiri~-nt for gPn~..AI;ng DNA s~u-~ ing O~ llpl~l~s uæful for both small and iarge-scale DNA æquPn~in~ projects.
There are three macrornol~cul~r co---~nP-u~ to the in Yitro int~
re~ction i) an ~lirl~ial tr~r~sposrm~ ii) lCllUVil~l or l~~ s~s~l~ inl~ ~;l,..~ and iii) a DNA target. These three col--l)o ents are mi~ed log~ll-.f in a reaction conl~ining the a~;ûlJliate buffa and cofactors. In the case of yeast r~hù~ oson Tyl, the reaction is briefly incubqted at 30 and 37- ~el~ and ~.. Ill;n~ by adding EDTA and heating to 65- ~'e1~ . Finally, the nucleic acids are phenol/chlo~ll., el~tr~ t~ and ethanol pl~ipita~ed. The reco~ef~d DNA is used to transform a host cell to drug re-~ict~n~ (or other s.-i~ble ~l~t~'lle .), allowing the id~ltifi~ti~n of target mo1~ul~s which have ~ived a lldr~s~osonin~ ;on(Fig.l).Asetofl.~n~son-bearingtargetDNAmolecules may then be used di~tly to obtain the DNA s~ucnces fl~nking the insertion sites,using two primers col~n~i~ to the ~ ~son ~nini; a col1~tion of such ir.sertions can be used for the effi~ent l~kOVel~ of DNA se~lu~nce info....~1;0nfrom the region of interest.
We have focused our initial efforts on developing a s~ific appli~tion of this t~r~no10gy, i.e., in vitro insertion of "primer island" artificial tran~osons (PARTs) into pl~mid targets for the p~ ose of DNA mapping and sequPn~ing In a~ldit;on to the feat~l~ s mentioned above (~ffi~ency of int,r~ n, r~n~omn.oc~
of insertion, and flexibility of ~ s~oson), this system has other advan~ges co.-.paled with P-icting methods, inChJ~ g: i) the in utro pr~tocol is simple and highly reliable, even in the hands of a novice, ii) the PART does not contain large t~rmin-q-1 repeats which, in TnS and Tnl~based ~st,.--s, hinder access to sequences flqnki~ the insertion j~ln~irn~, and iii) the ~acLon is carried out completely in vitro and thc~ful~ is ~--en~bl~ to biorhPmirq-1 q1t.orati~ n and p~ramPtf~r ~I;...i,,~t;on; this may be P~q~rq11y useful with 11m1~q1 DNA t,~ ...p1~t~ s such as those con~ g tandem s~ucnc~ r~peats, high GC content, or ~m~1n te -p1~1e topolog~ which might ~ iffi~11t targets.
Il..~lt~,~, I.~fi~l)o~n in~ ;on within targets was ~-.rr.. :en~l~ ~n~ m that ins~lions were ~ d from all regions of target DNAs. Hence, Tyl in~ nd ;n~eLI~;on in utro is, at a --;n;--------, a nearly--~n~iG~ process.
It may, in fact, be totally ~ Ol~U This will only b~l,~ clear upon tes~ng large nu,.,bc~ of targets co~.lAini11g diîîc.~DNA sequence Ç~lu~s. N_~_,L~less, our current results strongly support a model of quasi-~n-lom insertion with no app~nt major biases. In c~ntr~t, this feature is not gPn~lly obse.~ of other ll~nsposon S~IlIS adapted for DNA sequPncing; inctP~, ho~ and coldspots of inser~on ~l~qu~nlly lead to a non-random distribution of inser~ons n~ g these systems in~hle of ~ ing large s~.~mPn~ of DNA sequence, or high levels of w.,steful red~m~qnry in other regions. These problems have been c,~;u,llveilted in some ~ ."s with mutant h~n~sases which display altered target s~-ifir-ity (9). However, this approach provides only a limited rPlq--q-tion of l.~.n~ ~ifi~l target ~ifirity. It is known that host cell factors co"hibu~ to target q)erifirity in VTVO for both TnlO (9, 9a) and Tyl (28); such target s~erifirity is el;~".n~d by the use of in vitro s~."s as taught herein.
~ollu~ , the process of artificial ~n~OSOn int~ l;nn in vitro by r~hunldl and l~oll~nsposon il~ ~5, such as Tyl intP~e, displays r~ndom-like behavior (Fig. 2), making it ideal for the lJulllose of DNA sequen~ing. Quasi-random, accolding to the present invention, means that insertions can be ob~inedin virtually any sequence at a spacing of at least one ;nle~.,-l;nn per kb. In practice, in~eg~"t;t)rls have been o~ in~ at ~ ---J--- S~`-ing!C of as low as one in~l~lion per 500 bp, or even one int~L.,.I;nn per 400 bp. In contrast, large cold-spots have been found in targets of Tyl ll~;,~silion in vivo.
Rec?~lse our method of conshucling ar~ficial t~ s~osons is very v~q-tilp~
transpcwns conlAil-ing a variety of sE~u~-u~s can be con;,~u,,t~d for a nwll~r of spe~ific appli~q~i-nc For e~ample, ather ...~ can be ins~.hd into the mnltirlonin~ site (mcs) site of pAT-l, ;Q~ A;ng but not limited to yeast and ,.. z.. ~liqn drug-s~l~ble or auA~ù;~hic genes, gcn~-~l;~ marker c~c~s that can act as l.~nc~ nc Such ar~ficial l-~ncpos~nc can be used for ~marker qAfliti-n~, i.e., the insertion of a useful auAcn,ùphic marker into an acceptable region of a plqQ-niA of interest. For use in b~ - ;q or yeast, for e~cample, pAT-l d~i~di~.~s oon~ ing a variety of 5~1~r~ 1e ..-~ in the mcs can be consLIu~d~ and the marker of choice (duAollu~hic, drug re~;~t-q~ ~sor, etc.) can be added to a target pl~C...;A with a simple in vitro ;.~ n r~,ction Indeed, the products of a single in~ l;nr~ rea~ion can be viewed as an ";nlf~.~t;-~n library" col.~;~inh~g a c~llP~i~n of in~.lions, each clone c4nlA;ni~ a single insertion at a particular pho.~hf~ r bond. Should it be n~PA~I,~ an insertion at any sper-ific phûsl~hoAip~stp~r bond can be iAP~tifiP~ with coll~_n~;o~
library sc~æning methods, using a junction oligonucl~Potide as a probe. Hence, using a custom artificial ~ransposon, and applying the a~plul)liate sc~l~ing method, ~o~hin~nt mo'^~^-ules of a desired ~llu~ule can be recovered.
In ~l~lition to the artificial ~ ,oson, the other two co...l,one~-ts of the system, i.e., the integ~ ~ and the target, are also versatile. For e~ample, other int~ ~S or t~ansposases can effect an equivalent or nearly equivalent in vitro intP.~ ;oi. r~^tion. In ~d~lition~ mutant in~e".~rS are also useful. The s~ifiç
plu~lies of such integrases might together provide a wider ~ange of ;n~eæ-AI;on p~Ç~c,~ces ûr frequPn~i-ps. Also, rather than providing the integrase in the form of viral particles or VLPs, purified in~l~s can be used. These may display altered levels of activiq or stability, relative to VLP-~sori~t~d integrases.
The in vitro integ-At;nn reaction can employ a variety of DNA targets.
Pl~mi~s, in~ iing comil1s~ artificial chromosomes, as well as b~ctPriophage or viral vectors are useful. R~tPrinphage lambda DNA has been used as a target in similar re~fil n~ using Moloney murine lellhPmi~ virus (10) and Tyl integrases (11,12) provided in the form of viral particles.
The PART-based system for e~ DNA sequ-~ ;ng tçmrl~te~ can be readily applied to-the d~ p..lc~l of high ~U~g1~PUI~ massively p~llPl DNA-S~u~ ~ ~;fS The high degree of r~n~ornnP-QQ of insertion and the large fr31cfi~ n of clones e~n~,~t;.~ useful s~uc~ce data mean that a sh~lgu-~ ~pp~acl~
to s~ ;~ of large l~l..hin~n~ p1qcrni~s~ ;n~ ~ oQQmi~lS as well qs Pl and bqrt~i~l artificial clllo--.osol--Ps is feasible and highly suited to auk...,ation.
~2qn~Qm doubly drug resistant colonies can be s~lo-~l their DNA ~t~l, and fed dil~ctl~ into an au~u-..Jt d s~u~Y~;ng a~ us All of these steps are q..,f n~ ~^ to ai-~ n. R~cq~use a single set of optimized primPrQ can be used to so4u~ce an entire set of plqcmi~ derivatives, all of the steps can be done inp~r~llPl without op~l~lor intervention with regard to primer design and sP1~tion~
etc. Hence,althoughallific;dl u.~030n-f~ilit~tPdDNAse~lu~ ;~ispredicted to be very useful for small-scale s~u~ ;ng p~;ce~, it may be even more useful for massive projects such as the effort underway to map and sequence the human genome.

The artificial transposon which is employed according to the present invention ~nt~in.c a 3'-hydluAyl and is blunt-ended. Such mol~ P5 can be pl~ ed using restriction enzymes which make staggered cuts followed by a "filling-in" reaction with a DNA polymerase, such as Klenow fra~m.ont of DNA
poly,l,~.~ I. ~lle. ..~ ely, the artificial l.~nsposon can be pl~d by a PCR.
Typically the ends of PCR p~lucts require "I.;.. ;ng" to gen~ blunt ends.
Thus a restriction enzyme, such as ~nn I, which makes blunt-ended tern~ini can be used to trim a PCR product. Most simply, an artificial transposon Cûn~;~in~
in a plasmid can be icolqted from the pl-q~mitl with a restrirti- n enzyme, such as X~nn I, which makes blunt-ended termini. This provides a homogenous preparation of blunt-ended fr~m~nt~ in one step.
Tn~.~rq~ activity can be provided by virus-like particles, in the case of yeast ~~ s~oson Tyl, or by cellular nucleoplolein comple~ces in the case of o~ildlparticles. ~ltrrnqtively, l,u.;r~d ;n~ f, maybeused. Itisdesirable that the artificial ~-~n~l~son be added to the in vitro ll~r,s~s;ion incubqtion l,~lu~s as protein-fiw DNA ~ nc ~lth~ h some native I l u~n DNA may be present in the int~.&.~G pl~rA. .~ nc, typically such l,~;,~osons will not be geno-ti~lly mqrk~ and will be present in ci~lifi~ tly lower molar --.ounl~ than the artificial l.i.n~l~son DNA ~ A.n~A within a t,ar,;.~son's tennini may be any d~ir~ marker or even a cryptic s~u~ce. Antibiotic ~ ~ nr~ genes, useful for dther prokaryotes or eu!~ot~s are often useful. Aw~o~ hic ",~L -~ are also useful, e~qlly in yeast. Cis-acting regulatory e~ nlc, such as promoters, may also be desired to asc~b~n function of previously unknown regions flAnking an insertion. Marker DNAs also in~ll)des other non coding îedlu~ such as restri~ti~-n sites, primer binding (hybrilli7Ati~n) sites, etc.
The ratio of ar~ficial l-~n~l~son to target DNA has been found to be a ~ignifi~nt factor in the ~ffi~n~y of the reaction. Desirably the molar ratio will be at least 1:1, and more pl~dbly the molar ratio will be at least 2.5:1, 10:1 or 50: 1.

Host cells may be ~ rOl,.,o~ by any means known in the art, inclu~ling ~n~f~P~tion, ~ C~ n, el~1,~p.~ ;on~ etc. SelP~ n of l.,.r.cr~ P~ cells is typically and con~ iently carried out by a genetic sel~t~ means, ~1thol)gh genetic and biochPmi~l scl~ing methodc may also be employed.
In the case of Tyl l.~r~ ;t;o~, the use of the entire U3 or U5 lf- ~..;n~l sequences has been found to be ...~nP~ . Thus as little as 4 bp of t ....;~
sequence of U3 and/or U5 can be use~. (The Sf~uf~C~ of U3 and U5 are ~i~los~Pd in figure S of ~f~ncc 12.) While there is some evidence that other u~lated s~u~ces may be suilabl^ as a SUbst~t~ for intr~e enzymes to genelale single transposon-end joining pr~lucls (14), such sf~u~lces may not be suitable for g~n~".~ g the two l~ on-end, complete ;~tJ~ n product nf~f ~ for the present invention.
Primers which are employed for sequ~nl~ing accoldlng to the present invention are those which are known in the art for dideo~y-type sequen~ing These are typically s~n~ t;c, ~,ngle-s-~.,n~ oligon~,rlp~ Ps of about 12 60 bases in length. It is ~ ~b'F, acoo~g to the present invention that the primPrs for se~luen~;ng each flank of the in~.~l t~ be unique. Th~rcfol~, if the two tl~ls~oson termini are id~ 1, which they can be, the primer compk~ ;ty must e~tend into or be wholly derived from the "marker region~ so that each primer only hyb.;A;,- ~. to a single end of the t~n~l~oson~ "compk ..~r~ yto a t~ ...;n..s of an artificial tl~os~n" are those oligo..-,cl~ s which are 12to 60 bases in length which are derived from the t~minql appioA;...~t~y 150 bp of the artificial l.~n~ A~irner s~u~c~s which are ol)lil~d for DNA
s~uenl~;ng can easily be ~ d into the ar~ficial l-~n~
Viral particles, accol~,h~g to the present invention are nll~loo~olein comple~ces which are i~lqted from cellular - n~ of inr~ cells. In the case of yeast lcll0~ ~3nn Tyl, the particles are known as virus-like particles. An inle~ .~ activiq can be purified from such particles using protein pllrifi-~qtion techniques known in the art. While Tyl is e~emrlified in this q~ q i~n, it is believed that its closely related yeast ~LI~nsposon Ty2 will be equally useful.

In q~ ition, retroviral and other int~ldses may also be used according to the preænt invention. Avi. n myel~lq~tn*~ virus (AMV) in~-~ can be used to ...~i~P the conc~.~d in~f~ ;nn of a n artificial ll~na~osol into a target DNA30). ~llrine lP~ Pmiq virus (I~V) and human imml-no~Pficiency virus (HIV) ~oVil~ in~ s m~i-q-tP quasi-random insertion of qrtificial Llarls~osons into target DNAs (31). The 3-D structure of HIV-l int~g.,.~ core dom. in has been shown to be similar to the barte~ ~a~aase~ MuA (32). Ihus b~ t .~n~poa~s could also be used in a similar m--anner.
It has been found that divalent cations are n~ for l-..n~oS;I;on Suitable con~ ;Qns of ma~n~-~;u... or mqngqn.ose ions range from about 1 to about 50 mM. Preferably the con~ l;on is belween about S and 45 mM. The pH range which is s~)it~ble for in utro trAn.q osition is broad, from pH 6 to 8, and may desirably be from pH 7 to pH 8.
In ~lition to the appli~tion of PART t~hnology to the sequlon~ing of DNA, there are a n.. ~- of other appli~ tioll~ which are possible, owing to the high effi~ enl-y and randomness of insertion of PARTs. Some of these are oullin~below.
1. DNA sequencing and mapping i) Small-scale DNA s~uf~nci~g.
~ u~ . A 3.5 kb se~ nl of DNA is cloned into a pl--cmiA cloning vcctor. The investi~tnrs wish to obtain the CG E'-t-~ nue-l~title s~uen~ of this3.5 kb insert, on both strands using polymerase-based (Sanger) dideo~cy se~ Y~ g. PARTins~llionsare g~r~At~Yl throughoutthe p~ mi-l inuitrv. The c~ c!n iS s~.,..~.~d by restri--tinn mapping to d~t ....ne whether individual PART insertions are located in the pl--~mid bac~one or the inært, and a collection of target r-~miti~ b~g insertions every 100 200 bp in the insert is ~co.~.
Each PART is then used to s~uc~ce the DNA on both sides of the insertion, using unique primers homologous to the termini of the PART. Since ;.tandald dideo~ty sequen---i~ pr~locols lead to the ~ of 20a 300 bp (or more) useful scquence infû, ~ ;on, the entire sequence of the 3.5 kb insert is lecove~ed, on both st~n-ls.

ii) Idrge-scale sequ~Pnri~.
F~mr~e: A yeast ar~ficial chromosome (YAC), bact~ l ar~ficial chromosome (BAC), or other vehicle used for the pr~tion of large S~ ~t`i of DNA co~lAinc a large ~g...~-t of human DNA that ~ uil~s DNA sequenoe analysis. ~ ming that a 400 kb YAC is used, the YAC is resolved on a pulsed field gel cast with low-melting point agar, and eYc-ic~ PART insertions are gen~ ~ in vitro within the YAC. A ~i~lj7f~l PART de~ivaliv-e, conl;~ni~ a sP-lP~t~hle yeast marker is used to enable the facile l~co~ of PART insertions by transrol",ing the crllectinn into yeast by pro~pl~cl fusion, with s.~ .JPnt selection for complcm~Pnt~tion of an auxo~l,hy. PART insertions are ~co~
throughout the YAC in this l"anncr. Each PART insertion is then used to l~:U~.,.sequence from the fl~nking DNA in both directions by cycle sequPncing~ using a thPrmost~hle polymP-r~e YACs bearing PART insertions are shotgun sequenced until the entire sequence is recovered. The origin~l lin~ge of the se~lu~c~ is ...~in~h~ed throughout the ~JlUCedul~, making data ~c~imil~ti-n simpler than most large-scale sequPn~ing mPthods Finally, many aspects of this process are ~mPn~hle to ~U~ tinn.
iii) DNA M~Min~.
Using PART il~lio~s such as those described above, a PART map could be cor,sllu~l in a DNA s~ of interest. Since the PART CQn~inc a ~
of useful restriGtinn sites (~bp and 8-bp cutters), the loc~tinn of the in~.lions relative to the P.u~ int~ of the insert could be de~ ;nPd by cutting the clone with an cr~ c such as Not I, and l~g the products on the appf~liate gel. The sizes of the pru.luc~ would yield i~o~ lion about the lo~tion of the PART
insertion relative to the ends and other sites such as known genes or Not I sites.
The sequence info.~ ;on lccu~e~ed from such a PART insertion could then be coll~,laled with a map position. This approach enables the rapid ~i~mPnt of a sequence tag to a map position, which would be a useful in~. Ill?A;Ate on the way to co...l)leting the entire s~ucnce, esp~~ y if an entire genome is being sequenced. Another advantage is that the orif~in~l linkage of the various map positions is ...~inl~;n~d th~oughout the mapping pr~lur~. ~lt~rnqtively, PCR
mapping str~q-t~es can be used to map the positi-~n of the insertion, using one PCR
primer coll~nf~i~ to a l~n~ n end and one primer coll~pQn~ to a known position in the target plq~mid. The size of the res~ltqnt PCR pl~lUCt~
allows the insert positi~ n and c.. ;~ n to be ~et~ - ...in~A

2. Gene mappingby int~6~ u~tion.
FYqmr~ A yeast gene has been cloned as part of a large, e.g., 15 kb DNA insert on a plq~mi~ The investig~qtor wishes to know where, within this 15 kb, the gene is lorqt~. The clone was originqlly i~lqted by complc~ l;on of a mutant phenotype in yeast; hence, a r,.n- l;onal assay for the pl~ ce of ~e gene exists. A set of PART insertions is- made into the target pl-q~mi~ and these arethen transformed into yeast; non-compl~ nt;~ clones should contain insertions into the gene of interest. A s~k~ - yeast gene (e.g., UR~3, TRPI or HIS~) could be incol~ld~ed into the artificial transposon, both s;~ iry~g the ori~pnqls~l~tiol~ in yeast for clones ...~ ;n;n~ a l~n~l~so~ ion, and allowing the facile il~Pntifi~ -qtion of gene disrupter clones which could be later used dil~lly to knock out the gene of interest in the host genome.

3. Introduction of any functional or non-functional DNA cis element, sequence, or combination of sequences into another se~nent of DNA.
i) 12Pstrirtion sites for .~ making del~ n~) adding new DNA
fr~mP.ntc/sequenCeS.
~ f~t~ .tinn e~,y..le,s are mllll;~v.~osc tools. By i~ g a site for a particular er~yl~le at a desired lo-~qti-)n, the site could be used for ,"~ ing,making ~rl~ n.~ or adding ~ n f~mPnt~ to the target DNA.
F~ An artificial l,~l;.~3~n oQnl~ining two Not I rÇstricti~n sites fl~nking a sele~1Al~lr marker is ~1 into the target pl~cmj~l in vitro. Miniprep DNAs are scf~ed by restriction m~rring to locate an artificial llA~ls~ n insertion in the desired region. ~ ..AI;.rely, an ins~n library con~;ning artificial ~ ~so~ insertions throughout the target clone is sc~cned with a junction oligonucleotide to identify an insertion at a particular phosphodiester bond. Once a suitably-FositionP~ nsposon is i~PntifiP~, the plasmid is cleaved with Not I, thus removing the majoriq of the transposon, and g~--...,.I;ng ends with a Not I rPstri---ti- n site. Since many sites flank the sel~--t-qhlG marker in pAT-l and pAT-2, this approach could be adapted for use with any pair of erlLyl,les that would lead to the removal of the sPlect~l~le gene and allow the subs~ucn~ cloning of an insert at the site. This general apl~ach offers an qltr-..~;vc to creating a restriction en~Qnl)rl~ce site by the method of site d~led mutq~n,cic F~qmrl, 2: A yeast ar~ficial chromosome (YAC) contqining 800 kb of human DNA is used as a target to ~ ar~ficial ll~u~rûsùn insertions. Upon recovery of insertions, one is l"apped to a position ne. r a site thought to contain no funnti.-l-ql genes. Since the ~ficial tr. nsposon contq-ins a single Not I site a nd the chromosome lacks Not I sites, the unique site could be used to inært a novelgene into this loc-~ ti..~n.
ii) E~ulllûl~S~ enhqncc.rs, ~....in~lo~ introns, exons.
F-qmr'- An artificial ~ n~posnn is created which con~ c the third e~con of gene W which is known to encode a stretch of 99 prolines followed by 33 hicti(1ina and then 11 lylusil~es. Normal ,-.~.. Aliqn 5' splice donor, 3' splice acceptor, and branch acceptor sites re il~co~ ed into the ~ n~oson at the app~ ale pos-l;nnc for correct sFli~i~, along with a ~~ 1c marker. The lli n~oson is in~ "-~ into gene X on a plq~mi~l . nd the plqcmi-1 s~bsc.lu~rtly trqncf~P~ ~ into ,.. z.. ~liqn cells in culture. The e~on is found to be appluy~incol~l~ into the l.~n~-.;bed mRNA of gene X, with precise P.I~;c:.,,, of all non-e~on s~u~l ces. The protein eh.--..i~h~ of the region en~ d by this e~on is now studied in the new protein c,-~nt-~t iii) Drug ~!1P~t ~l or au~ollu~hic ........ ~ useful in e ~ 1 and non~lY.;.. ~r~ ûl~n;c~llc inr,l~ b~t~ plants, yeast, insects, Dr~
worms, rodents, l~ nc, ~ lc in general.
"Marker swap~ or "Marker~ tio~ r.s~osons.
Goal: introduce or e~cl-~nge genetic l.~h~ls in a vector of interest, using the inte~,ldLion reaction rather than restriction enzymes. Tran~osons similar to the PART but cQI~t~imng dirr~l drug ~c;.~nf~ (chlorarnphenicol, kanamycin) or yeast Q~l~table ...~.L~.~ (UR~13, TR~l, HIS3, LEU2) ~n the l,~nsl?oson tern~ini could be in~.i~ted into a target pl~c..,i~ of choice. The res--lt~n~ plasmids could be ~k,c~ for the -~uisilion of the new marker and then if desired, be sc~ned for loss of a p~ ~ ;Cl;ng marker.
F~mrle You have a pl~Q~nid that conl~inc a marker for ~mr~ in reQiQt~nce as well as a gene of interest. For an llpco.~ g ~ n~, yOu desire that the pl~Qmid contain a chlo.i....rh~l~ical re-QiC~n~ marker, and require that the pl~Qmi~ he lacking the ~mricillin gene. Thus, the end goal is to h-ave a single pl~Qmi-l carrying your gene of interest, a chl~ h~ir~l reQiQt~n~ marker, and no ~mricillin recict~nr,e marker. To aG~ rlich this, you ~o~ an in vi~ro on with an artificial l~im~son conl~;n;ng a chl~ mphenicol gene, and elect pl~Qmi~lQ that are chlol ...ph~ eC:g~nt Ne~ct, you replica plate to ~mpi~-illin col~t;in.-~ plates, and id~lif~ chlo~ .,rh~l-ir,ol re-Qict~nt1~mr:--illin sensitive clones. The new marker is found to have ;ntfg.i1tPA within the Amp marker.
iv) Genes. Any gene of interest could be cloned into a pAT d~iv~tive and direc~y in~d as a ~.~fi~ n into a DNA target.
E~sample: A gene ll f-~p q wants to build a variety of new adenovirus cor~ uc~ to test as d~ icl~s for the cystic fibrosis l-i-fiC-.~ .b.;~n~
regulator (CFI~) gene, which is the human gcne l~o~s;l,le for cystic fibrosis.
Since both the adenovirus g~llle and the CFTR cDNA are both quite large, Sl~ g;f-5 based on lY~;~ 5 are not easily itlPntifiP~ Tnct~P~, the gene th~ ;s~ clones the C~-lK cDNA dnven by the CFTR promoter into a pAT
derivative C~l ~ illg a sP~ marker, and inserts the resll1t~nt ar~ficial oson c~l~ing the CFI~ gene into the ad~lovil~ vector. Thus, various constructs are rapidly built and tested.

v) Any func~io~al or non-fim-^-tio..~^l DNA
DNA ~p~mpnts compri~d of any m)-^-l-^Qtidç s~uw ce or co.-~l;;n-~;on of sequences, could be envisione~ to be inCC,l~)ldtCd into an artificial h~la~oson,thus beco~ g, mPn~blP to ~c4l~.k;n~ n with a target via an ;n~.~ )n rP~.r-.ti-)n~ i) Codon insertion mllt~genecic Restriction sites for a rare cutting restr ctiQn cr~l,le (e.g. SrfI, cutting GCCCGGGC) can be pos;l;on~i just inside the termini of the arlificial l~ oson, but fl~s~ki~ the s~lP~^-t~ e marker (e.g. dhfr). The restri~^-ti~n sites can be positinnPd such that, after dPlPtinn of the marker con~-n~ (dhfr in the ~ rlA) Srff fr~mPnt, there would be a net insertion of an in'~.g~l llulll~ of codons into the target pl~mi~l, rP-~Illting from the new bases introduced (these would consist of the target site dllpli~ti~n~ artificial transposon le .n;n~l base pairs, and the restri-^-tion site, plus one or two ad~iti~n^~l base pairs as n~.SSA~ to ensure the proper reading frame). Following insertion of such an artificial l.~n~loson intoa target pl, cmi~ or cosmid of interest, the population of insertion mutant r~mitl~
or co~mi~ls ;ould be digested en masse with Srff, diluted and self-ligated. These deleted pi~cmi~$ would then be ~ rv d into host OEllS, resllti~ in a popul~tion of codon insertion ~ ; nl~ These ciodon ~lio,~ r.t~ could then be used to study ~lut~ ~r fim~^-ti~n(s) are r~d~d in ~e target DNA ~ n?lly.
The restri-^,tion site would again be very helpfi~l for rapid .-~ ~ of the codoninsertion. O~er m~ho~ for codon in~lion ~u~gell~is are taught in the art (33, 34).

4. "Car~ along" t~nsposi~on.
An artificial L~i~n~o~n c~ries a drug s~ le ...~-L --/or ~n~ which allow srl~ti.^n of l.~n~l 0~ 4n~ining DNA taArget. The l.,.r.~ n also cont~;n.c other DNA sequences ^dj^^~nt to the marker (such âs a gene). Hence, botn the drug marker and the gene of interest are il,w-luced upon int~ n of an artificial ~,~ns~)oson with such a structure.

. Fusion protein con~tl u.~.
An artificial L~ oson is dç-~ign~d such that, upon inse~tion into an open reading frame of a fim-tionql gene, a fusion protein would be produced. The fusion would be compriced of a portion of the originql coding region of the i~nctiQnql gene, as well as a 1~ l,ul~. which could be used to identify such active fusion p~leins.
F~...1,1F An artificial ~-s~oson is created that c~nt~q-in~ the beta g,qlqct~si~l~q~ gene. The reading frame is open from the t~ ...in~ of the li.n~osQl~ through the beta g~l~qrtocidqoe gene. Upon int~ lion in a frame in a target gene, a fusion protein is produced that shows beta ~ ctosidq~ activity.
6. Transgenic constmcts.
A drug-s~l~tqble marker useful in the organism under study is introduced into a desired region of a gene or DNA within a cloning vector, for the llltimqte pul~ose of introducing the s~.. nnl of DNA into the host genome. This general approach has been r~pol~d for bact~iq~ yeast, dr~s~phila, C. ele~,qn~ and mouse,as well as other mqmmql~, and in~ludes h.~gl~ knockouts such as those ~ol~d by M. C~l ecchi's lab.
EJcample 1: A l~ch~ wishes to e.;.. ne a 20 kb s~ment of mouse DNA for pos~ plu~ tcr activity both in cultured cells and in the conte~ct of t_eor~Pnicm. An artificial tr~nQ~pOSO~ COI ~;.ining a lepûller gene such as Chlorq-n~rh~nic~l acetyl 1- --.~f. -~ (CAI~, lu~ e, or ~-gql^^t~si~lqQ~ could bein~ ~d into t_e 2~ kb region, and scl~d by restriction Illa~lng. NeAt, the in~,lions could be tested for ~A~l~on in cell culture or muscle injection t~nQi~nt assays. Finally, constructs sh~ A~les~;on could be used to g~?.,.~.
transgenic qnimql~ Such qnimql~ could be used to study the ~ ,ion conf~l~d by the ~r~ll,ot~r, by assaying l~rr activiq in various tissues or develop states.
F~ P- 2: An artificial ~ csson is created which Cont~inQ a human l.~n~ )lional enh-qnt~pr ~lem~nt that functionQ only in heart muscle tissue during e. rly heart development. By inserting copies of this transposon in the upstream, downsh~ , and intron regions of a gene of interest (cloned on a r~cmi~), constructs are ~ ~ where the gene would pot~ lly be lf~.llat~ by the enh~ncer in a tissue-specific and ~Illpoldl manner. These constructs are used tog~ te !.,.n~enic ~nim~lc where this gene would be eAln~sed in this m~nner.
F~ 3: Tl~nsgenic hlocLoul constructs. An artificial har~os~n co~ 1~; n;~ a NEO gene is created and inl~ t~l into a plqcmi~ clone C~~ g the S' portion of a gene of interest. The insertions are sc,~ncd, and a single insertion occllmng in the first e~on of the gene, just downstream of the ~n~lqtinn start codon AUG, is identifif~ The resnlting construct is used directly to L -o~L--u~ the gene by gf~nf~ g a transgenic animal by ES t~hnology. A second version would include the addition of a co~-lf-~l~t~t-le marLer at the 3' end of the construct to diL~er,~ate belween ho~ lQgous and non-homologous insertions.
This coun~f .~ l~t~hle marker could be carried on a second artificial 1.; n~llo~n This general approach has been ~le~ihe~ by C; l~hi and colleagues to gen~
nLnockout mice" lacLing the function of a particular gene.

E~aunples Constmction of pAT-1 pAT-l (pSDS44) and pAT-2 (pSD545) were co~sL-ucb~d as follows. First, the pl~mi~ pRS316 (ref. 15; a d~ e of pBLUESCRIPT, S~tag-pnp~) was mo~1ifiçd to elimin~te the ~mri~-illin re~i~t~nc~ (amp'~ gene. This was acc~mrli~h~d by lig~tir~ together two fra~mPnt~ of pRS316 (a 2.1 kb Ssp I
r.~,...." and a 2.1 kb Bsa IlSsp I fi~lllc~ ), thus c~ling the plq~mi~l pSD528 which lacks a functi~n~l bla gene; this pl~mid can be p,u~ in the pyrimi~linP-requiling E. coli strain MH1066 since the yeast UR~3 gene on this construct complements the b~cteri~l pyrF aw-o~oph~ (16). pAT-1 and pAT-2 were constructed from plasmid pSD528 by replacing the pBLUESCRIPT
multi~ ing site (mcs) (from the unique ~n I site to the unique Sac I site) with polymerase chain reaction (PCR) adapters cont~ining the appr~p,iale sequences tocreate the structure indicated in Fig. 2. These PCR adapters were generated using primers SDl 12 (JB661~ (5'- AAAA-GCTGGG-TACCGA-ACATGTT-CTCGAGGTCGACGGTATCG3') and SD113 (~662) (5'~ GAATTGGA-GCTCGAAC-ATGTTCACCGC-GGTGG-CGGCCGCTC-3') with pl~smi~s pBLUESCRIPT and pSD511 as te-..pl~t~s The reC~lting PCR products were digested with }~n I and Sac I, and ligated to Kpn VSac I- ligest~ pSD528 to generate pAT-1 and pAT-2. The ~llu~ules of ~hese corsllu~ were conf;....~ by restriction ~ .;ng and s~ ~ ce analysis.

In vi~ro reaction conditions.
A typical in utro DNA inl~g~t;~ n was carried out in a 20~L1 reaction volume, and c~nt~in~d the following. 100-500 ng artificial l. ..-~poson (0.8 kb), 1 ~Lg CsCl-purified pl~mi~ target (a 10 to 1 molar ratio of l.~ .os~ n to target), 2~110X reaction buffer (150 mM MgCl2, 100 mM Tris HCl, pH 7.5, 100 mM
KCl, and 10 mM DTT), S ~Ll 20% [w/v] PEG 8000, 2 ~1 VLPs, and water to 20 ~1. The reac~on was inc~b~l~ at 30 Celsius for 30 ~-.;n~t., followed by 37-Celsius for 10 .~;n~u~s, and then was t~-- .--;.-~I,~d by adding i.o ~ o.s M EDTA
and heating to 65 Celsius for 20 ~ JO~S Finally, the nucleic acids were phenol/chlolof~ ~ ethanol pl~;p;l~t'~d C~ll~d by cc~t~;r~g~l;on, washed with 70% ethqn~l, and ~ s~cd in 10 ILi TE (10 mM Tns, pH 8.0, 1 mM EDTA). 1 ~1 was used to h~rlsÇul.,l 6 ~1 DHlOB E. coli (Gibco/BRL) to drug l~ - C~n~ by el~1l0pc .~ )n.

PCR, sequencing, primers, pla~id constructions, CsCl preps.
The PCR was carried out using reagents dll~-n~d from Perkin Elmer, a des~nhed (17). DNA se~uY~-;~ was carried out using Sequenase (IJSB), and analyzed as described (18). Custom oligonucleotide primf~rs were ob~in~ from Operon Te~hnologies, Inc. (~lsmf~, C~lif~rnia). The two ~im~rs used for sequen~ing from within the PART were SDlll (JB563) (5'-GACACTCTGTTA-TTACAAATCG3')andSD110(1B532) (5'-GGTGATCCCTGAGCAGGTGG-3').
The in~.dlion site of each PART insertion was ~ in~d using either one or both of these primers, and analyzed with the aid of the Wi~coll~in GCG package.
Pl~mi~s were constructed using standard DNA cloning m~.thods (19), and were pllrifi~ from E.coli cultures by either STEI Imnip~ (20) or ~ ine lysis followed by CsCl b~n~ing (21).

~ion of artificial ll 5~SQ~C from pAT-1 and dc.;-..li~cs.
20~g of CsCl-purified pl~mi~ DNA was ~igPst~d with 50 units of Xinn I
(13Oehringer M~nnhiem) for 4 hours at 37- CPIC;1m The reSulting f~m~nt~ were .~p~. ;.l~ on a 1% agaroselTBE gel, and the ~ r.~l1oson fr~m~nt was electroeluted from the gel using an IBI ele.,~ u~;on device.

Reco.~. ~ of donec ca~ .~sQ-~ insertions using ampicillin/trimethop~n pL~tes.
E. coli clones ca~nng plq~mi-l~ with tmn~poson inærtions were identifi~
by s~lecti-n on M9 minimql plates (22) c~nl~in;ng 1.0 rnM !l--zl--;nP HCl, 50 ~g/ml ampicillin (Amp) and 100 f~g/ml trim~hoprim (Tri; Sigma). After one to two days ~ "~l~l;~ n at 37- Cel~ the majoriq of co!oni~ growing on M9/Amp/Tri plates c4fi~inPd plqo~ni~iS with a 1. .n~l~30n inser~ion. Dilutions of the l,~hSrO....~;on were l~ ulind~ plated on LB plates con~A;ning 50 ~g/ml Amp (22); this control l~lcu~ d the nwll~ of t rget pl-q~mi~s suc~ rully carried through the procedure. When c~...p~r~d to the number of colonies on M9/Amp/Tri plates, the rl~ of l~ns~os~ n insertion could be c;~ n~
(rlclluwl~ of insertion = t#-colonies on M9/Amp/Tri plates] / [# c410ni~q~s on LB/Amp plates]). A po~li._ control plqcmitl pSD511, c4nlA;l~in~ both AmpR and TnR "~ outinely gave rise to equivalent numbers of colonies on LB/Amp (50 ug/ml), M9/Tri (100 ug/ml), or M9/Amp/Tri (50/lOOug/ml) plates under these con-liti,ln~, Transfolmation of E. col .
The two strains llarlsro~ d r~uli~lely in this work were DHSa (23) and DHlOB (24). DHSa vas pncp~d for elec~ropoldtion as described (25), and el~c1.0cG...~ nl DHlOB cells were yu chas~d from Gibco/BRL. Tl~nsÇo....-al;on by elec~loyo.AI;~n was accollll)l;~hrAA for both strains using a Biorad Genepulser with 1 mm ~u~lt~s and the following sett~ c-Ar-AcitAnre 25 ~LFD; voltage: 1.8 kV; and re-~ictAnc~ 200 ohms. Using pUCl9 or pBLUESCRIPT as a t~st p~ mid, freshly-prepared ele~tlocolllycte.~t DH5a generally showed transformation effiri~nri~s of 107 - 108 c~lonirsl~g DNA, whc.c~s ckcllocol~ t~t DHlOB
pulch~d from BRlJGibco generally showed efficirnri~ of S X 108 to 5 X 109 colonies/~g DNA.

VLP preparation.
VLPs were p cp~d from yeast cultures as described (26). F~ ti5~n~ from the final sucrose gr~lient cQnlA;~ , intr.gTa~e activity were aliquoted and frozen at -70 Celsius where they were stable for more than 6 m-nthc.

In ~tro integration of "p~imer island" 1~ ~ons into a doned segment of yeast chromosome m c~ :ed on a pla~id target.
We ne~ct gP~ dl~ PART ii~s~Lons in vitro using various pl-~cmi~1 targets.
One of the pli}~ test clones con~ t~d of a pRS200 backbone (a d~iv~i~_ of pBL~JESCRIPI~ with an 8.0 kb insert that spans bp 136,155 to 144,333 of yeast chrom~some m; this pl~cmi~l is called p76-2. With a single in vitro in~ dl;on ~tinn, we l~ appnJ,;~ tA~1~ 13,000 PART insertions in p7~2 ~Table 1).

Table 1. Rec~ of PART insertions into clone 76-2.
Rxn EDTAa Total Total insertionFrcquencyof transforrnantsb pl~smi~s~n~yosi~iond -- 1. - O O
2. 3.1 X 108 4.5 X lo8 3. - 3.1 X lo8 1.3 X 1044.2 X 10-5 4. + 5.7X108 S.OXlo2 9.1X10-7 Rcaction 1) negativc transformation cont~l (no DNA added); 2) ~ ;Li~c trahS[~ ldt;On control (pSD5 I l, which con~ins both AmpR and TriR markcrs); 3) complctc intcgration reaction using p7~2 as thc targct; 4) same as reaction 3, but EDTA was added (inhibits integrasc activity).
a +, EDTA addcd to 25 mM
b. Total numbcr of AmpR transformants c. Total numbcr of AmpR~TriR transformants d. Number of t~r.sposilions into targct ptasmid (AmpR/TriR colonics) dividcd by the total numbcr of transformants (Amp colonics) By m~llring the nu,llb~r of coloniP~ tl~fol"led to ampicillin re-~ict~ne~
vs. colllbined trimPth~prim and ~mp^illin re-ci~t~n-,e, we det~ lined that the rl~uen~ of tl~ ~los~n ins~ion l~CO.~ was appro~im~tPly 4.2 X lo-5 (i.e., 1 insertion per 2.4 X 104 target m~ lPs Table 1). Although this frequency is not likely to lC~ t the upper limits of opl;...;,,.l;<n, it is suffici~ntly high that a large number of insertion events are readily recovered, while suffici~-ntly low that a single target is gPnP~lly limited to a single ~ ~son insertion (two l,~s~oson insertions in a single target might be useful for some pUllJOSeS~ but would render the mc'eculP useless as a s~u-Y-c;~ template).
Analysis of 156 r~n<lomly chosen AmpR/TriRcoloniPs in~ t~ that PART
insertions occurred into all areas of the pl~mi-1 target, inclu-ling both the pRS200 backbone (6.0 kb) and the 8.0 kb chromosome m insert, as ~etPrmined by restriction mapping and/or s~ucnce analysis (Table 2).

Table 2. Examination of TriR/AmpR colonies ~rom a single in vitto integration reaction.
qo Total number of TriR clone;s exarnined 156 100 # minipreps recovercd 153 98 .

# casily-identifi~blc inscnions . 134 86 In insert 78 50 In vcctor 56 36 Other 19. 12 double insertions/cotransformantsa 13 8 unknown plqsmid map 5 3 notransposon I c I

L Thls class co~t~inc some plasmids that apparently had lwo indcpendcnt inscrtions in thc target as detcrmined by restriction mapping, and othcrc with DNA sequcncc that was readablc to the insertion junction, at which point two supcrimposed sequences werc obser~cd.

More than 86~o of these 156 clones (134) had easily-identifi~ PART
insertions; of these, 78 (50%) were in the cloned 8 kb insert, while 56 (36%) were in the vector. A small L~lc~,~ge of the clones were found to have two s~lperimposed restriction maps/and or s~u~nces~ There are several likely pl~n~l;nns for this result, includin~ the pos~ihility that two ~ mi~s ll~srul"led a single E. coli clone, or that two ~ ~son in3~lions OCCU11.d into a single pl~mi~ target; the available evidence in~irat~Ps that most of these clones are P~pl~in.od by such mP~h~ni~m~ Hence, a small portion of clones recovered from an in vitro int .~ I;on reaction would not be s ~i~hl~ for direct DNA sequ~nce analysis for this reason (12 % in this ~ rl~7 T~hle 2). LLkewise, vector insertions would not be useful for seq~nPnrir~ the insert. N~._.ll,eless, one ofevery t~o Ampa/Tria coloniPs analyzed from this single reaction could be used directly to obtain DNA s~u~nce from the cloned insert. Ful!l..-... o.~, analysisof only 156 Il~iniplq)S led to the assembly of 78 useful insertions in an 8 kb insert, coll~onding to an e~ led ~ trihl)tion of roughly one insertion per 100 bp.
The distribution of individual insertions of the ar~ficial l.~fi ,l)o~n relativeto ~jacPnt insertions is shown in Table 3.

-~ - 30 -~ T~iBLE 3 Tabulation of PART insertion data from p!~s~id target p76-2 ~on Insert~ point di~nce to Plasn~d ~1 p76~2 S~pr~ne ~on~
(~ m nu.~
S-pru~c cnd 1361SS
lSl 136394 ~ 239 107 13~460 R 3S

135 - 13668S F 7~

124 137192 ~ 27 89 13~811 F lS4 - 147 1378~9 R - 68 138503 ~ lS8 63 138587 F . 6 152 138702 F &4 110 1387~0P 18 32 138147 R 2.7 112 13928~ ~ 377 103 140n94 R 348 57 141024 ~ 1 2 141o74 R S0 58 141?6S F 132 31 14232? R 12 ?8 142180 F 86 l4n26 R 46 12~ 142382 R lS6 3 1425Sl R 169 27 143616 F .283 39 1438S6 ~ 240 Sl 143921 F 6S
13 144076 F lSS
66 14412~ F Sl 3-pnune cnd 144333 206 S~ti~i~ on insatlons n.78 Mean in~nral ~ ,)cc = 102.3 ~/- 88.1 Ins~ons~ f~ cach 1 1~ of targe~
Nl~mb of inshons Region of targct pes 1~ u~et DNA
1,6.1~5 to 137,000 13 1J-~OOO tO 138,~00 9 13~:,000 to 139,040 17 13Ç,OOOto 140,0X3 14 140.000 to 141,000 6 141,000~o 142,000 10 142,000 to 143,0~0 9 143,000 to 144,0~0 6 144,000 to 144.333 6 Mcan numb of insenions pcr ~b target DNA . 10.2 +~- 3.
n r~ 34 (44%) Rc~rse 44 (56%) Since the entire yeast chromosome m sequence has been previously determin~ (27), we could easily identify the precise sites of transposon in~P~ ion by de~~ ih~.ng the nucleotide sequences at the insertion junctions. Tnde~ the 78PART insertions were found to be distributed throughout the entire 8 kb insert (Fig. 3). A little less than half of these insertions were in the rOlw~ orilont~tion (34/78 or 44%), in~ ting a slight ori~nt~tion bias for this target. However, since primer e~n~iol~ can be initi~t~ into the sequences fl~nking the insertion on both sides irrespective of the PART ori~nt~tion, an or ent~qti~ n bias does not affect the utility of the PART insertion for ~ull~ses of DNA sequen~i~ The mean di~t~nce~ belween adjac~-nt insertions was 102.3 +/- 88.1 overall. Only sixof the intervals were greater than 250 bp, and the largest of these was only 377bp. Hence, the vast majority of the intervals bel~n~n ~ nt lldnsyoson insertions were well below the ~ t~nce that can be reached with an average primer eYten~ion under sequen~ing con~lition~ A ~lopelly of Tyl ~ o~r~e is that it creates ch~ra~t~ori~tic S bp target sequence ~ plit~tion~ fl~nking the insertion site upon int~ tion (10-12, 28). As e-l~cled, S bp target site dupli~ti-n~ were found at each PART in~P~ site e~min~ (only a small portion of the insertions were -sequenced at both ends in this e-~mpll~-). No on~ or ~ ge...Pnt~ were ob~ed.
A cQnn~Pptu~l primer e~tpnsion contig map based on our results is shown in Fig. ~. We have made the ~llmption that each primer e~tPn~ion would lead to the suc~P-~ CO~rCd~ of 250 bp of useful sequence inform~tion- 100% of the sequence would be recovered on one strand or the other using the 78 PART
insertions shown in Fig. 3. Only 6 gaps (3 on the top strand, and 3 on the bottom; each < 150 bp) would exist. But because the two initial primer P.tPn~ion~
fl~nking such a gap would cross in the middle on opposile str~n~s, uninlc ~)led DNA s~u~ l~ce would be l~cuver~d on one strand or the other. Nevc.Lheless, the gaps on the 1GII~q;n;I1g strand could be closed with either~ ition~l PART
insertions in the n~ s~l ~ regions, ;dP-ntifiPd with a~?pr~pliate restriction mapping, ii) custom primPr~, or iii) longer sequ~pnning runs. Of course, we have made the assumption that only 250 bp of sequence information can be recovered from a single primer e t~n~ n; in fact, greater than 400 is routinely obtained with aulo,llalGd sequencers, and 800 to 1000 is bo~lning possible with a~ lllalGd sequenc~.~ in development. Hence, if the mean readable sequence is e~t~on~d to 400 bp, 100% of the sequence could be easily recovered using fewer than 78 PART insertions.

Other targets tested.
In ~dition to clone 7~2 conl;~;n;~ a DNA insert from yeast chromosome m, we have tested other pl~mid targets. These pl~mi-l~ had a variety of backbone structures and carried various cloned inserts (Table 3). Ihe bacLl~onesint~luded pUCl9 and pBLUESCRIPr as well as others, and the DNA inserts origin~ted from dirrerGnt species inclu~ling yeast and human. In each case, results similar to those shown for clone 7~2 were obtained: i) insertions were ~I~apped to all regions of these targets, ii) a large number of insertions was readily recovered from reactions using each target, and iii) reco~cd insertions cQ~ict~ntly served as succe-~cful se~l..en~ s. M~1G~, in two cases other than p7~2 (pCAR143 and pWAF-l; table 3), this system was used to l~CO~ 90-100% of the nucleotide scquGIlce from clones with previously unknown sequences. Hence, in vi~o in~ of artificial tran~osons is e-l~t~ to work well with most or all pl~mi~ targets, making it both a generally useful sequen in~ tool and a general method of ;~ g.AI;ng new DNA se~uenc~s into pl~mi-l targets to genGlatG rGc~l"bi~ t DNA molecules.
MAPPlNG AND SEQUENCING COSMID DNA USING ARTIFICIAL
TR~NSPOSONS.
We have ~lemon~tr~ted that artificial transposons can be çffi~i~ntly inleglaled into a wide variety of pl~mi~ targets in utro using ~yl inlegl~se. Our data in-li~, that cosmi-ls can also serve as targets for int~.~ration using the same protocol as that used for pl~mi~lc, Hence, DNA rnapping, sequçncing and functional genetic analysis can be ~roln~ed directly on large (30-50 kb) DNA

2 1 6 1 7 6~

inserts pro~qt~ in cosmid cloning vectors. Tnese results cor,fi~ that the targetfor artifici~l t ~ oson inærtion is fle~cible; in p. ;r..~ , any DNA mo~cculP- could serve as a target for the integl~lion of artificial l~n~5 ~nc in vitro. The resl-lting reco...hinqnt could be used either to analyze the regions surrounding the inærtion, or for any otner purpose generally provided by recombinant DNA m~lpcllles~
inclllding but not limited to function-ql genetic analysis and ~ecolllbinallt DNA
Png;nrY~ g Supportive Data.
1. AT-2 insertions have been genPrqte~ in four dirr~l~nt cocmi~s using the same meth-Ylc used to generate AT-2 insertions in rqcmi~s. These include three cosmids obt~ned from the Lawrence Li~.lllûl~, Genome Center, P23932, P13544, and P20080, each co~cicting of a Lawnst cloning vector and an insert of appro~rimqttoly 30 to 50 ~h derived from Human chlulllosolllc 19, as well as oneadrlition-q-l cosmi~, JEDI-C, also c~l~ing an inært of ap~ y 30 to 50 lmh.2. RPstricti~n mapping. Insertions nlapped to all regions of target c4cmi~lC
~u~ ing a quasi-random model for in~c,;.l;on a was oh-s~l~rcd for plqcmi-lc.
3. AT-2/cosmid ,~c4...hin~nt~ were svcc~ rully used as sequ~-ing t~lllplat~s with ABI Prism cycle se~v~nc-;~ te~ nolc~y. More than 100 cosmid recombinant. (in~h~ 17 from a previously lm~ * - ;~ c4smi~ F23932) have been evaluated as s4u~-nn;~ t~ pl~les and lhe majonq (>9096) gave r~-1~hv1e s4u~nce of 300 to 600 bp for each primer e~tpncion with high levels of ~,y ( > 95 %)-4. 8 kb of previously rh~r~ctP~i7~d sequence of the cosmid F13544 was r~analyzed with AT-2 insertions and Prism sequpn~ing te~hn~l~gy (see Figure 11). All available data in-li~te that this method is fully c~p~l~le of recovering ~cl~t~o sequence infollllation co...~ hvle with other state-of-the-art methods.
Thus, cosmi~s can he analy_ed with ar~ficial ~ ;,~n t~hnol~y at both the structural and sequence levels. It is predicted that cosmid l xo...bin~ could also be used for fi-netion~l genetic analysis. The advantages of direct analysis of DNA inserts prop~g~t~ as large l~...bin~nt cosmid mo~ s are as follows. 1) -Direct analysis allows the ori~inql lin~ge of the insert to be ~ in~ throughout the analysis, avoiding the problems -q~ qt~ with de~llo~u~g linkage, e.g. as in shotgun sequen~-in~ 2) direct analysis allows the navigation of n~iffiCult'' DNAinserts cont~ining complex repeat structures and 3) map and sequence inform-qti-~n from a single tr~q~n~;poson insertion can be used in concert to permit ~imrlified sequence q~sem~ly sch~om~s VARYING T~IE NUCLEOTIDE SEQUENCE AND STRUCTURE O~ IE
ARTIFICL~L TRANSPOSON.
Our initial ~ nl~ were pc.rolllled with the artificial ll~ns~oson AT-2.
Our results suggested that the sequ~cc and design of the artificial lldn~OSOll were likely to be flexible. We have now tested this hypothesis by de-~igning and constructing artificial ~ osons with a varieq of sequences and features. Like AT-2, these artificial trans~osons were constructed in pAT-l or pAT-2 vectors orderivatives (Figure 12), relying upon the same mlllticloning site for construction of these plq~mid5, and the sarne Xmn I restriction s~ eg~ for p~ ;on of the ll~SpOSOIl from the vector (in each case, the artificial ~ fis~Josol~ bears the sarne rel~tionchir to its parent pl~mj~l that AT-2 bears to pAT-2). The results of ourstudies in-li~tç that, indeed, the æquence of the artificial ll~s~o30n can be varied ~IJbsl~ lly while ~;n;.~ ;nn activity. Thus, any desired feature can, in principle, be incol~l~led into an ar~ficial !.~n~l o50l using mtoth~ls available for P~gi~ ;ng pl~cmi~ls or linear DNA m~'^cules The following artificial ~solls have been constructed and, where in~ tP~, have also been tested for ion or otherwiæ.
1. AT-2. The artificial ~ oson AT-2 co~ inc at its termini 4 bp of Tyl U3 I~...,n~l æquences (S'-AACA-3'); s~e~.."n~l primer sites SD110, 111, 118 and 119 used for PCR or æquen~ing; s~t~ . ..,in~l restri~ti5~n sites for mapping and Pngin~.;ng; a drug-select~ble dhfr r~te conferring resict~nc~ to the antibiotic trim~thoprim in E. coli. AT-2 was constructed in the pl~cmid pAT-2.
2. AT-2-TRPl This transposon is i~llo.nti-~l to AT-2 with the exception that the yeast auxotrophic marker TRPl has been added at the unique Hind m site present in pAT-2. The oveiall transposon is appr~ ximAtely 1.6 kb in length. TheTRPl marker is s~k~ 1e in both b~t~iA and yeast. AT-2-TRPl transposes in vitro using the m.q.th~s established for AT-2. Insertions were found to 'oe quasi-randomly distributed. Following i~legla~on into pl~cmi-l targets and tran~rv....A~ion into yeast, the locations of fi-n. tionAlly active regions on the target pl~cmi~ were -~pp~l by ~s~lional inactivation. For C ~ le, in one plAcmi~
target conlAining the yeast URA3 and LYS2 genes (pSD553), AT-2-TRPl insertions were found to inactivate these genes upon insertion within their openreading frames, leading to a Ura- or Lys- phenotype in yeast (Table 4). When insertions OC~;Ur1ed outside of these genes in the same target, however, the plasmids were still capable of yielding a Ura+, Lys+ phenotype in yeast. In all cases, a Trp+ phenotype due to the TRPl marker on the transposon was observed in yeast.

? Table 4 AT2-TRPl /pSD553 l~,llbinants Pl~ y~s Recombin~nt Amp Tmp Ura Lys Trp AT2-TRPl ins. site R R - + + URA3 ORF
2 R R + + + LYS2 3' UTR
3 R R + - + LYS2 ORF
4 R R + + + vector (bet~een Amp and CEN) R R + - + ND
6 R R + - + LYS2 ORF
7 R R + - + ND
8 R R + - + ND
9 R R + + + ND
R R + + + ND
11 R R + + + ND
12 R R - + + URA3 ORF
13 R R - + + ND
14 R R + - + ND
R R + - + ND
16 R R + + + ND
17 R R + - + ND
18 R R + + + ND
19 R R + + + ND
R R + - + ND
553 +C R S + +
554 +C R S - - +

Lcgcnd Results of r '- ~ analy~s of pSDSS3 I~ in yeast Theresultsof' analyrisof20 '~E ~~ AT-2-TRPl .~ ' of pSMS3arel ' J
The .~- ' were fir t e ~ in virro, and ~ ed in E coli by rielection for I ~
After mapping sites of ~ each ,~ ' was I r ~~ into the yeast ritrain yPH499 (ura352, lys801, trplD63) and platod on yntketic media lacking uracil, Iysine, or IlJpt ,*
Finally, L. ' were replica plated to each media and their p~ ,e scored R = re istad;
S = ~ensitive; + = grow~ on media laclcing the pecifiet nutrient; - = no grow~ lne dtes of dx insertion events d~ - ~ by sequence 1~ - .; indicated in the last column 3. AT-2-LacZ. This transposon is identical to AT-2 with the exception that the LacZ marker has been inserted between the unique Sal I and Xho I sites of pAT-2. The overall transposon is appro~im~tely 4 kb in length. AT-2-LacZ
tr~n~posp~ in vitro with the mPth~s established for AT-2. When insertion occurs in-frame with an open reading frame present on the target, the resnltin~
recombin~nt encodes a fusion protein which can be ass~d for function in the a~p~pliate host using an in~li~tor substrate such as X-gal. We have tested this approach on an 8 kb segmPnt of yeast chromosome m, and AT-2-LacZ accurately predicted the location of a known gene present on the clone. Thus, ar~ficial transposons can be used to functionally map the location of genes by making ~oller fusion proteins.
4. AT-2-neo. This trans~oson is idPnti~l to AT-2 with the exception of the ~d-lition of a neo c~ette at the unique Hind m site in pAT-2. This ll~s~)osoll has not been tested funr-tion~lly.

5. AT-3. This transposon was derived from pAT-l by adding a cassette ~.n~ling the neo gene at the unique Bam HI site of pAT-l. This neo cassette confers re~i~Pn~ to G418 in yeast and k~ .yc n in b~etçri~ AT-3 l,~r,s~G3es in vitro with nt th~ est^' 1i~h~ for AT-2. The ~ .nl~ n of the neo c~set~e is left to right, with the unique Not I site of AT-3 on the left, and the unique Xho I site on the right, of the C~ t~..

6. AT4. This l~ oson is id~nti~l to AT-3 with the eYt~ption that the neo c~eette is in the opposite G.;.~ ;on AT~ !.,~ o~s in vitro with the me.th<~s est~ h~d for AT-3.

7. AT-S. This ~ n~oson was de~ig~d to contain the bla (~mp;cillin re~ict~n~) gene and is otherwise ident;-~l to AT-3. AT-5 has been de-~ign~ but not built nor tested.
These results collec~vely in-li~ that the cis sequences of the artificial .osi)n can be varied l,A~nsi~rely while r~; ;ning l~fisl ositi~)n function and quasi-random inl~.g.~tion in vitro. Thus, trans~osons with custom Çe~lul~s can be constructed and used for a variety of pul~oses. These features include both functional and non-functior~l DNA sequences, primer sites, rest~ietilm sites, and otherwise useful sequences.

Æ~CES
1. Smith, L.M. (1993) Science 262, 530-531.
2. ~t~hlra, K., Rossi, J.J., and Wallace, R.B. (1984) Ann. Rev. Biochem. 53, 323-356.
3. Sambrook, J., Fritch, E.F., and M~ni~ti~, T. (1989) Molecular Cloning A
Laboratory Manual, Second Edidon. Cold Spring Harbor Labol~t~l~ Press, Cold Spring Harbor, NY. pp 13.2-13.104.
4. Sulston, J., Du, Z., Thomas, K;, Wilson, R., Hillier, L, Staden, R., and etc. (1992) Nature 35C, 37-41.
5. ~usub~l, F.M., Brent, R., KinE~ton, R.E., Moore, D.D., S~ m~nJ J.G., Smith, J.A., and Struhl, K. (1989) Current Protocols in Molecular Biology 1, 7.2.1- 7.2.20.
6. Sl.~t~....An, M., ~milton, B.A., Mayeda, C.A., Simon, M.I., M~uwi E.M., and P ~1~7791'" M.J. (1991) Proc. Natl. Acad. Sci. USA 88, 1247-1250.
7. Phadnis, S.H., Huang, H.V., and Berg, D.E. (1989) Proc. Natl. Acad. Sci.
USA 86, 5908-5912.

8. Way, J.C., Davis, M.A., Mori~to, D., Roberts, D.E., and Kl~n~r, N.
(1984) Gene 32, 369-379.

9. Kl~n~r, N., Bender, J., and Gû~-s...~n, S. (1991) Methods Er~ymol.
204, 139-180.
9a. Lee, F.Y.,Butler, D., andKl~n~r, N. (1987) Proc. Natl. Acad. Sci.
USA 84, 7876- .

10. Brown, P.O., Bo~ an, B., Varmus, H.E., and Bishop, J.M. (1987) Cell 49, 347-356.

11. F;~hir~er, D.J. and Boeke, J.D. (1988) Cell 54, 955-966.

12. Fi~llin~er, D.J. and Boeke, J.D. (1990) Genes Dev. 4, 324-330.

13. B~ , L. and Boeke, J.D. (1994) Mol. Cell. Biol., in press.

14. B~A;~ lAn, L. and Boeke, J.D. (1994) Mol. Cell. Biol., in press.

15. .8ikors~ R.S., and Hieter, P. (1989) Gene~ics 1~2, 19-27.

16. 8iknr~ R.S., and Boeke, J.D. (1991) Methods En~ymol. 194, 302-318.

17. Innis, M.A., and ~;to.lf~n~, D.H. (1990) ~: PCR Protocols A Guide to Me~hods and Applications. ~c~demic Press, Inc., San Diego, CA pp 3-12.

18. Sanger, F., Nicklen, S., and Coulson, A.R. (1977) Proc. Natl. Acad. Sci.
USA 74, 5463-5467.

19. Sambrook, J., Fritch, E.F., and ~niAti~, T. (1989) Molecular Cloning A
Laboratory Manual, Second Edidon. Cold Spring Harbor Labolat~ly Press, Cold Spring Harbor, NY. pp 1.53-1.110.

20. ~olm~s, D.S., and Quigley, M. (1981) Anal. Biochem. 114, 193-197.

21. Au~lbel, F.M., Brent, R., Kingctc-n, R.E., Moore, D.D., Sei~1mAn~ J.G., Smith, J.A., and Struhl, K. (1989) Current Protocols in Molecular Biology 1, 1.7.1-1.7.11.

22. M~niAti~, T., Fritsch, E.F., and Sambrook, J. (1982) Molecular Cloning A Laboratory Manual. Cold Spring Harbor LabGld~l~ Press, Cold Spring Harbor, NY. pp68-69.

23. ~An~hAn, D. (1983) J. Mol. Biol. 166, 557-580.

24. Calvin, N.M., and Hanawalt, P.C. (1988) J. Bacteriol. 170, 2796-2801.

25 ~usubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., .s~ m~n~ J.G., Smith, J.A., and Struhl, K. (1989) Current Protocols in Molecular Biology 1, 1.8.4-1.8.8.

26. n~;~....~n, L.T., Mor~ n, G.M., F;~hin~er, D.J., Merbs, S.L., Gabriel, A., and Boeke, J.D. (1994) Gene, in press.

27. Oliver, S.G., van der Aart, Q.J.M., Agostoni Carbone, M.L., Aigle, M., Albergl~ina, L., and etc. (1992) Nature 357, 3846.

28. Ji, H., Moore, D.P., Blomberg, M.A., n~;t~ n~ L.T., Voytas, D.F., Natsoulis, G., and Boeke, J.D. (1993) Cell 73, 1007-1018.

29. Rll~hm~n, F.D., and Craigie, R. (1991) Proc. Natl. Acad. Sci. USA 88, 1339-1343.

30. Vora, A. C., Mc~ord, M., Fitzgerald, M. L., Inman, R. B. and Grandgenett, D. P. (1994) Nucleic Acids. Res. 22, 44544461.

31. Brown, P. O., Singh, I. and Crowley, R. (1995) Genetic Fooll..;n~;ng:
using a re~oviral integrase to study gene function (abstract). Retroviral Tnt~g.~s lU~e~ing, 1/19/95 Washinglon, D.C.

32. Dyda, F., T~ n, A. B., Jenkins, T. M., F~g~lm~n, A., (~igi.o, R. and Davies, D. R. (1994) Science 266,1981-1986.

33. Barany, F. (1985) Proc. Na~l. Acad. Sci. USA 82:
4202-4206.

34. Boeke, J. D. (1981) Molec. Gen. Genet. 181: 288-191.

Y~QUL.._h LISTING

(1) ~r'N~RAT lh~ORMATION:
(i) APPLICANT: The Johns Hopkin~ University ~ii) TITLE OF l~hh~ION: In Vitro Tran~po~ition of Artificial Tranupo~on~
(iii) NUMBER OF ~u~ S: 7 (iv) CORk~S~ONDEN OE ADDRESS:
'A'l AnD~æS~: Banner, Birch, McKie, and Beckett B~ STREET: 1001 G Street, N.W.
C CITY: Wa~hington D STATE: D.C.
E COuh.~s: U.S.A.
F ZIP: 20001 (v) CO.~u~ R~AnARrF FORM:
,'A' MEDIUM TYPE: Floppy disk B COMPUTER: IBM PC compatible ,C, OPERATING SYSTEM: PC-DOS/MS-DOS
DJ SOFTWARE: PatentIn PDleA~e ~1.0, Ver~ion ~1.25 (vi) ~UKKhh. APPLICATION DATA:
(A) APPLI Q TION NUMBER:
(B) FILING DATE: 02-MAR-1995 (C) CLASSIFICATION:
(viii) A,~O'~.rs/AGENT lN~ ATION:
(A) NAMæ: ~agan, Sarah A.
(B) REGISTRATION NUMBER: 32,141 (C) ~r~n~NCE/DOCRET NUMBER: 01107.49245 ( iX ) T~T-~ ~ ~ Ur lCATION lN~ O~.TION:
(A) TELEPHONE: 202.508.9100 (B) TELEFAX: 202.508.9299 (C) TELEX: 197430 BBMB UT

(2) lN~ O~.TION FOR SEQ ID NO:1:
( i ) S~UL.._~ CHARACTERISTICS:
'A' LENGTH: 4164 ba~e pair~
B TYPE: nucleic acid ,C STRANDEDNESS: double ID TOPOLOGY: circular (ii) MOLECULE TYPE: DNA (gen~ ;c) (iii) ~srO~n~ICAL: NO
(iv) ANTI-SENSE: NO

~vii) IMMEDIATE SOURCE:
(B) CLONE: pAT-1 (Xi) ~LyUL.._~ DESCRIPTION: SEQ ID NO:1:
TCGCGCGTTT CGGTGATGAC GGTr-AAAAr,C TCTr-ArAr~T GCAGCTCCCG r-Ar-ACGGTCA 60 CAG~..~.~. GTAAGCGGAT GCCGGGAGCA GACAAGCCCG T QGGGCGCG Tr-Ar~}Gb~G 120 A~r~TAcr~r AG~.... AA TTCAATTCAT CA---.---- TTA.. ~- TTTTGATTTC 240 G~... ~... -`, AAA ..... GATTCGGTAA TCTCC~AAr-A GAAGGAAGAA Cr-AAr,r-AAr,G 300 _ _ ___ __AGrArAr-~T TAGATTGGTA TATATACGCA TATGTAGTGT Tr-AAr-AAArA TGAAATTGCC 360 CAGTATTCTT AAcc~AAcTG rAr~cAArAA AAACCTGCAG rAAArrAArA TAAATCATGT 420 CGAAAGCTAC ATATAAr,r-~A CGTGCTGCTA CTCATCCTAG .C~.~..GCT GCCAAGCTAT 480 TTAATATCAT GrArr-AAAAG CAAArAAAr,T TGTGTGCTTC ATTGGATGTT CGTACCACCA 540 AGGAATTACT GGAGTTAGTT GAAGCATTAG GTCCrAAAAT ..-...ACTA AAAAr~ Q TG 600 CCAAGTACAA ~----ACTC TTCr-~Ar-Ar-A GAAAATTTGC TGA Q TTGGT AATArAr,TCA 720 AATTGCAGTA CTCTGCGGGT GTATArAr-AA TAGCAGAATG GGrAr-ArATT ACGAATGCAC 780 ACGGl~,l~. GGGCC Q GGT ATTGTTAGCG GTTTGAAG Q GGCGGCAGAA GAAGTAACAA 840 AGGAACCTAG AGGC~-l..G ATGTTAG QG AATTGT QTG Q AGGGCTCC CTATCTACTG 900 rArAATATAC TAAGGGTACT GTTGA Q TTG cr-~Ar-~r7cGA r~AAr-ATTTT GTTATCGGCT 960 TTATTGCT Q AArA~A Q TG GGTGr~Ar-~r- ATGAAGGTTA CGA-..GG--~ ATTATGA Q C 1020 CCG6.~.aGG TTTAGATGAC AAr,Gr-Ar-~r~G Q ~GGvl~A A QGTATAGA ACCGTGGATG 1080 Al~l~G.~.C TArAr,r-ATCT GA Q TTATTA TTGTTGGAAG AGGACTATTT GCAAAGGGAA 1140 AAATTAGAGC TTCAATTTAA TTATAT QGT TATTACCCTA TGCGGl~-aA AATAccGcAc 1320 AGATGCGTAA CrArAAAATA CC~CAT Q GG AAATTGTAAA CGTTAATATT TTGTTAAAAT 1380 TCGCGTTAAA -~--AA AT Q GCT QT TTTTTAACCA ATAGGCCGAA ATOGGCAAAA 1440 lCC~llATAA ATr-~AAAr-~A TAGACCr-~r-~ TAGGGTTGAG l~7~ ~C~A GTTTGGAACA 1500 AGAGTC Q CT ATTAAAr-~Ar- GTGGACTC Q ACGT QAAGG Gcr-AAAAAr-c GTCTAT QGG 1560 AAG Q CTAAA TCGGAACCCT AAACGCACCC CCCGA m AG AGCTTGACGG GGAAAGCCGG 1680 CGAACGTGGC r-~C-AAAGr-PA GGr-AAr-AAAG CGAAAGGAGC GGGCGCTAGG GCGCTGG Q A 1740 GTGTAGCGGT Q CGCTGCCC GTAACr~CrA CACCCGCCGC GCTTAATGCG CCGCTA Q GG 1800 GCGC~.CGCG C Q TTCGC Q TT QGGCTGC GCAACTGTTG GGAAr~}G~r-A TCGGTGCGGG 1860 C~.C..CGCT ATTACGC Q G CTGGCGAAAG GGGGATGTGC TGCAAGGCGA TTAAGTTGGG 1920 TAACGCCAGG G...iCCCAG TrArr-ACGTT GTAAAACC-AC GGC Q GTGAA TTGTAATACG 1980 ACTCACTATA GGGCGAATTG GAGCTCGAAC ATGTT QCCG CGGTGGCGGC CG~.~lAGAA 2040 CTAGTGGATC CCCCGGGCTG QGGAATTCG ATAT QAGCT TATCrATACC GTCGACCTCG 2100 Ar~ArATGTT CGGTAC QGC '~ ~L~Ccc TTTAGTGAGG GTTAATTCCG AGCTTGGCGT 2160 AATCATGGTC ATAGCTGTTT C~.~.~.~AA ATTGTTATCC GCT Q CAATT cr~r~r-AArA 2220 TACGAGCCGG AAGrATAAAG TGTAAAGCCT GGGGTGCCTA ATGAGTGAGG TAACT Q CAT 2280 TAATTGCGTT GCGCT Q CTG CCCGCTTTCC AGTCGGGAAA C~-~-C4.GC Q GCTG Q TT 2340 AATGAATCGG C Q ACGCGCG GG~ GCC~ GTTTGCGTAT .GGGC~.~. TCCGCTTCCT 2400 CGCT Q CTGA CTCGCTGCGC .CGG.C~-.C GGCTGCGGCG AGCGGTAT Q GCTCACTCAA 2460 AGGCGGTAAT ACGGTTATCC ArAr-AAT QG GGr-ATAAr~C ArrAAArAAr ATGTGAG Q A 2520 TCCGCCCCCC TGACGAG Q T r~r~AAAATc GACGCT QAG T QGAGGTGG cr-~AAcccr-A 2640 CAGGACTATA AArATArrAr~ GC~...CCCC CTGGAAGCTC C~.C~.GOGC .~.C~.~..C 2700 CGACCCTGCC GCTTACC~C' TAC~.~.CCG C~... CCC TTCGC~AGC GTGGCGCTTT 2760 CT Q TAGCTC ACGCTGTAGG TATCTCAGTT CGG.G-AGGT CGTTCGCTCC AAGCTGGGCT 2820 G.~G~ACGA ACCCCC~4.l Q GCCCGACC GCTGCGCCTT A.CCG~.AAC TA.~G-~..G 2880 AGTCrAAr,CC GGTAAr-~ Q C GACTTATCGC Q CTGGCAGC AGCCACTGGT AA Q GGATTA 2940 GCAGAGCGAG GTATGTAGGC GGTGCTACAG AG..C..GAA GTGGTGGCCT AACTACGGCT 3000 GAGTTGGTAG ~.~-..GATCC GGrAAArAAA C Q CCGCTGG TAGCGGTGGT l~ -- 3120 G QAGCAGCA GATTACGCGC Ar-A~AAAAAAG GATCT QAGA AGA C~..G A~ A 3180 CGGGGTCTGA CGCTCAGTGG AAcr-~AAAcT CACGTTAAGG GATTTTGGTC ATGAGATTAT 3240 ~AAAAAr~,~T CTT Q CCTAG A.Cu..--AA ATTAAAAATG AAGTTTTAAA TCAATCTAAA 3300 CAGCGATCTG TCTATTTCGT TCATCCATAG TTGCCTGACT CCCCG.CG~G TAGATAAr,TA 3420 CGATACGGGA GGGCTTAC Q ~C.GGCCC~A GTGCTGCAAT ~,ATPCCr-~TT ATTGAAGCAT 3480 TTATCAGGGT TA..~.~. A TGAGCGGATA Q TATTTGAA TGTATTTAGA AAAATAAP~ 3540 AATAGGGGTT CCGCGCACAT ~.CCC~AA AGTGCCACCT GGG.C~..~l CAT Q CGTGC 3600 TATAAAAATA, ATTATAATTT A-A-A~ A A~ATAAATAT ATA-AATTA-A-A AA~ArAAA~T 3660 AAA~AAi~A ATTAAAGAAA AAATAGTTTT ~...CC~A AGATGTAAAA GACTCTAGGG 3720 GGATCGCCAA rAAATAr-TAC C m TATCTT GC. C~G CTCTCAGGTA TTAATGCCGA 3780 A..~... AT u.~. ~.G TA~-AAGACr~ r~rArr-AAAA .C~-.GATT TTA Q TTTTA 3840 CTTATCGTTA ATCGAATGTA TATCTATTTA ATCTGCTTTT C..~lC.AAT AAATATATAT 3900 GTAAAGTACG ~.... ~.. G AAA -.---A AAC~... ~... TA......... TCTT Q TTCC 3960 GTAACTCTTC TACu..~... ATTTACTTTC TAAAATCCAA ATArAAA~rA TAAAAATAAA 4020 TA-A~r~r-A~-~ GTAAATTCCC AAATTATTCC ATCATTAAAA GATACGAGGC GCGTGTAAGT 4080 TACAGGCAAG CGA.CC~.CC TAA~-A-A--AcrA TTATTATCAT GACATTAACC TATAAAAA~A 4140 (2) lN~ ATIoN FOR SEQ ID NO:2:
(i) SEQUEN OE CHARACTERISTICS:
~A'l LENGTH: 4933 ba~e pair~
(B TYPE: n~lcle1c acid (C STRA~n~FSS: double (D,~ TOPOLOGY: circular (ii) MOLECULE TYPE: DNA (genomic) (iii) ~ru-nh~lCAL: NO
~iv) ANTI-SENSE: NO

(vii) IMNEDIATE SOUR OE :
(B) CLONE: pAT-2 (Xi) 5~U~N~ DESCRIPTION: SEQ ID NO:2:

CAG~.~.~. GTAAGCGGAT GCCGGGAGCA GACAAGCCCG TCAGGGCGCG TCAGCGGGTG 120 ACCATArrAC AGC~ AA TT Q ATTCAT Q .......... TTA.. `... I TTTTGATTTC 240 G6..-~--.G AAA...--.. GA..CG~.AA TCTCCr-AArA r,AAr,rAArAA CGAAGGAAGG 300 AGrArAr-A5-T TAGATTGGTA TATATArGCA TATGTAGTGT Tn-AAr-AAAr-A TGAAATTGCC 360 CAGTATTCTT AArcrAAcTG rArArAArAA AAACCTGCAG r-AAACGAAr,A TAAAT Q TGT 420 CGAAAGCTAC ATATAAGr-AA CGTGCTGCTA CT QTCCTAG .C~.~..GCT GC Q AGCTAT 480 TTAATATCAT GrAcr-AAAAG r'AAArAAACT TGTGTGCTTC ATTGGATGTT CGTACCACCA 540 AGGAATTACT GGAGTTAGTT GAAG Q TTAG GTCCrAAAAT --,...ACTA AAAArArATG 600 CCA~GTACAA ----.ACTC TTCr-AAr-ACA GAAAATTTGC TGACATTGGT AATAr~r,TCA 720 AATTGCAGTA CTCTGCGGGT GTATArAr-AA TAGCAGAATG GGrAr-ArATT ACGAATGCAC 780 AGGAACCTAG AGGC~....G ATGTTAGCAG AATTGTCATG CAAGGGCTCC CTATCTACTG 900 rArAATATAC TAAGGGTACT GTTGA Q TTG Cr~Ar-AGC~CA rAAAr-ATTTT GTTATCGGCT 960 TTATTGCTCA AAGAr-ArATG GGTG~-An ATGAAGGTTA CGA.~G~7..G ATTATGA Q C 1020 CCG6.~.GGG m AGATGAC AAGC~ CG Q..GG~.`A A QGTATAGA ACCGTGGATG 1080 A.,.~,.~.C TAC~r,GATCT GA Q TTATTA ~ ~GAAG AGGACTATTT G QAAGGGAA 1140 GGGATGCTAA GGTAr~-,GGT GAACGTTA Q GAAAAG Q GG CTGGC~ Q TATTTGAGAA 1200 AAATTAGAGC TT QATTTAA TTATAT QGT TATTACCCTA TGC~G.~..,A AATA5C~ Q C 1320 AGATGCGTAA Gr-Ar-AAAATA CCG Q T QGG AAATTGTAAA CGTTAATATT TTGTTAAAAT 1380 .CGC~..AAA ----~--AA ATC~GCT Q T TTTTTAAC Q ATAr~CCr-AA ATCCGCAAAA 1440 .CC~..ATAA ATrAAAAr-AA TACArCr-AGA TAGGGTTGAG .`..~.IC~A GTTTGGAACA 1500 AGAGTC Q CT ATTAAAnAAr GTGGACTC Q ACGTCAAAGG Gcr-AAAAA5c GTCTAT Q GG 1560 GCGATGGCCC ACTACGTGAA CCAT Q CCCT AAT Q AGTTT -...GGG~,.CG AGGTGCCGTA 1620 AAG QCTAAA TCGr-AACCCT AAACGCACCC CCCGATTTAG AGCTTGACGG GGAAAGCCGG 1680 CGAACGTGGC r-Ar-AAAGGAA GcrAAG-AAA~ CGAAAGGAGC GGGCGCTAGG GCGCTGGCAA 1740 GTGTAGCGGT Q CGCTGCGC GTAACr-ArrA Q CCCGCCGC GCTTAATGCG CCGCTACAGG 1800 C~ CGCT ATTACGCCAG CTGGCGAAAG GGGGATGTGC TGCAAGGCGA TTAAGTTGGG 1920 TAACGCCAGG G. . . CC~AG TCACGACGTT GTAAAA~C GGCCAGTGAA TTGTAATACG 1980 CTAGTGGATC CTGCAAGCAG GATAG~GGC ATGCACGATT TGTAATAA~A GA~ G 2100 TATTTTTAAA GAAAGTCTAT TTAATArA~G TGATTATATT AATTAACGGT AAGCATCAGC 2160 GGGTGACAAA ACGAGCATGC TTACTAATAA AATGTTAACC TCT~-AGGAA~- AATTGTGAAA 2220 CTATCACTAA TGGTAGCTAT ATC~-~A~-AAT GGAGTTATCG GGAATGGCCC TGATATTCCA 2280 TGGAGTGCCA AAGGTGAACA G~C~.~..- AAAGCTATTA CCTAT~AC~-A ATGGCTGTTG 2340 ACACGTT Q A GTTTTACATC TGACAATGAG AACGTATTGA T~ ~C~ATc AATTAAA~AT 2460 GCTTTAACCA ACCTAAA~-AA AATAACGGAT CATGTCATTG m CAGGTGG TGGGGAGATA 2520 TA~AAAAGCC TGATCGAT Q AGTAr-AT~A CTA~-ATATAT CTA~ATAr-A Q TCGAGCCG 2580 GAAGGTGATG TTTACTTTCC TGAAATCCCC AG Q ATTTTA GGCCAGTTTT TACC~AAGAC 2640 TTCGCCTCTA A~ATAAATTA TAGTTAC Q A ATCTGGCAAA AGGGTTAACA AGTGGCAG Q 2700 ACGGATTCGC AAAC~ A CGC~,,~, GC~AAAAGCC GCGC QGGTT TGCGATCCGC 2760 TGTGC QGGC GTTAGGCGTC ATATGAAGAT ~C~G~ `ATC CCTGAG QGG TGGCGGAAAC 2820 ATTGGATGCT GAGAATTCGA TAT QAGCTT ATCrATPCCG TCGACCTCGA GAA Q TGTTC 2880 GGTAC QGCT -~.-..CC~. TTAGTGAGGG TTAATTCCGA G~GGC~A ATCATGGT Q 2940 TAG~ C ~ GAAA TTGTTATCCG CT Q QATTC CA~A~AA~AT ACGAGCCGGA 3000 AG~-ATAP~-T GTAAAGCCTG GG~GC~AA TGAGTGAGGT AACT Q Q TT AATTGCGTTG 3060 CGCT Q CTGC CCG~C~A GTCGGGAA~AC ~-~-C~.GCC AGCTGQTTA ATGAATCGGC 3120 Q A~iX~C~G GGAGAGGCGG TTTGCGTATT GGGCGCTCTT CCG~C~C GCTCACTGAC 3180 TCGCTGCGCT ~G~C~..~G GCTGCGGCGA GCGGTAT QG CT Q CTCAAA GGCGGTAATA 3240 CGGTTATC Q QGAAT QGG G~-ATAACG Q G~-~AA~-AA~-A TGTGAG Q AA AGGCCAGCAA 3300 AAGGC QGGA ACCGTAAAAA GGCCGCGTTG CTGGCGTTTT TC QTAGGCT CCGCCCCCC, 3360 ~-~C~-AGrATC A~AAAAATCG ACGCTCAAGT QGAGGTGGC ~AAArCC~A~ AGGACTATAA 3420 A~TPC~AGG CG...CCCCC TGGAAGCTCC CTCGTGCGCT ~,CC,~,,CC GACCCTGCCG 3480 CTTACCGGAT AC~.~.CCGC ~..,~,CC~, TCGGGAAGCG TGGCGCTTTC TCATAGCTCA 3540 CCCCCCG~C AGCCCGACCG CTGCGCCTTA ~CC~AACT A~C~ GA GTCCAACCCG 3660 GTAArA~ACG ACTTATCGCC ACTGGCAGCA GCCACTGGTA ACAGGATTAG CAGAGCGAGG 3720 TATGTAGGCG GTGCTACAGA G~ GAAG .~.GGC~A ACTACGGCTA CACTAGAAGG 3780 TCTTGATCCG G~AAA~AAAC CACCG~ AGOGG.GG~ G CAAGCAG Q G 3900 ATTACGCGCA ~-AAAAAAAGG ATCTCAAGAA GA~C~GA ~ AC GGGGTCTGAC 3960 GCTCAGTGGA A~r~AAAcTc ACGTTAAGGG ATTTTGGTCA TGAGATTATC AAAAAGGATC 4020 TTCACCTAGA C~ AAA TTAAAAATGA AGTTTTAAAT CAATCTAAAG TATATATGAG 4080 CTA.~C~l~ CATCCATAGT TGCCTGACTC CCC~C~,~ AGATAACTAC GATACGGGAG 4200 GGCTTACCAT CTGGCCCCAG TGCTGCAATG ATA-cc~-ATTA TTGAAGCATT TATCAGGGTT 4260 A~ AT GAGCG~-ATA~ ATATTTGAAT GTATTTAGAA AAATAAA~AA ATAGGGGTTC 4320 CGCGCACATT TCCCC~-AAAA GTGC Q CCTG G~C~,,~C ATCACGTGCT ATAAAAATAA 4380 TTATAATTTA AA~ AA TATAAATATA TAAATTAAAA ATA~-AAAGTA AAAAAAGAAA 4440 TTAAAGAAAA AATAGTTTTT ~..~CCGAA GATGTAAAAG ACTCTAGGGG GATCGCCAAC 4500 AAATACTACC TTTTATCTTG ~ C~GC TCTCAGGTAT TAATGCCGAA ~,,,~ATC 4560 ~ ~, A~AA~Ac~A~ A~cr-AAAAT C~.~ATTT TA QTTTTAC TTATCGTTAA 4620 TCGAATGTAT ATCTA m AA TC~CCT$1TC .~ AATA AATATATATG TAAAGTACGC 4680 ~ `A AA~ ,AA AC~ A~ ,,, CTTCATTCCG TAA~ 4740 AC~ A TTTACTTTCT AAAATCCAAA TA~AAAA~T AAAAATAAAT AAA~ACAGAr, 4800 TAAATTCC Q AATTATTCCA T Q TTAAAAG ATA~Jr-~GGCG CGTGTAAGTT ACAGGCAAGC 4860 GA~CCG~C~ AAGAAAc~AT TATTATCATG ACATTAACCT ATAAAAATAG GCGTATCACG 4920 AGGCC~C GTC 4933 (2) l~rOh~3ATION FOR SEQ ID NO:3:
(i) ~yUL OE CHARACTERISTICS:
A LENGTH: 864 ba~e pair~
B TYPE: nucleic acid C, ST~A~-~3~ SS: double D TOPOLOGY: 1 inear (ii) MOLECULE TYPE: DNA (genomic) (iii) ~rO.~h.ICAL: NO

` 2161767 (iv) ANTI-SENSE: NO

(vii) IMMEDIATE SOURCE:
(B) CLONE: PART

(xi) ~yuL CE DESCRIPTION: SEQ ID NO:3:
TGTT Q CCGC GGTGGCGGCC GCTCTAGAAC TAGTGGATCC TG~a~-~Ar-G ATA~-ACGGCA 60 TG Q CGATTT GTAATAA~G A~.~ ATTTTTA-A-AG AAAGTCTATT TAATACAAr,T 120 GATTATATTA ATTAACGGTA AGCATCAGCG GGTGarAAAA- CGAGCATGCT TAcTAAT~AA 180 ATGTTAACCT CTr-Ar,r-AAr-~ ATTGTGAAAC TATCACTAAT GGTAGCTATA TCr-~A~-aATG 240 GAGTTATCGG GAATGGCCCT GATATTC Q T GGAGTGC Q A AGGTGAA QG ~C~ A 300 AAGCTATTAC CTATAACrAA TGG~ GG TTGrAcGr~A GACTTTTGAA TCAATGGGAG 360 CATTACCCA~ CCr-aAAGTAT GCG~-AA CACGTTCAAG TTTTACATCT GACAATGAGA 42û
ACGTATTGAT ~-..CCAT Q ATTAAAa~TG CTTTAACC~A CCTAAAr-~AA ATAACGGATC 480 ATGT QTTGT TT Q GGTGGT GGGr-AnATAT Ar~A~AGCCT GATCGATCAA GTA~aTACAr 540 Ta~ATATATC TarAATAr-Ac ATCGAGCCGG AAGGTGATGT TTA~ C~ GAAATCCCCA 6û0 GCAAT m AG GC QGTTTTT AccrAAr-AcT TCGCCTCTAA rATAAATTAT AGTTacraAA 660 TCTGCCAAAA GGGTTAA QA GTGGCAGCAA CGGATTCGCA AAC~-~-~AC GC~ G 720 Cr~AAA~CCG CCCr~-~TTT GCGATCCGCT GTGC Q GGCG TTAGG~- A TATGAAGATT 780 CG~GATCC CT~ GGT GGCGGAAA Q TTGGATGCTG AGAATTCGAT AT Q AGCTTA 840 TCr-ATArCGT CGACCTCGAG AA Q 864 (2) INFORMATION FOR SEQ ID NO:4:
yv~r -r CHARACTERISTICS:
~A LENGTH: 22 base pairs B TYPE: nucleic acid C STFA~K~ SS: single ~D~ TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA
(iii) ~rO-~-lCAL: NO
(iv) ANTI-SENSE: NO

(vii) IMMEDIATE SOURCE:
(B) CLONE: JB563 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

(2) INFORMATION FOR SEQ ID NO:5:
(i) SEQUENCE CHARACTERISTICS:
~A LENGTH: 20 ba~e pairs B TYPE: nucleic acid C STRANDEDNESS: ~ingle ,D, TOPOLOGY: linear ~iiJ MOLECULE TYPE: cDNA
(iii) hrro~ IcAL: NO
(iv) ANTI-SENSE: NO

(vii) IMMEDIATE SOURCE:
(B) CLONE: JB532 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

(2) INFORMATION FOR SEQ ID NO:6:
(i) SEQUENCE CHARACTERISTICS:
~A'l LENGTH: 42 ba~e pair~
B TYPE: nucleic acid ,C, STRANn~nNESS- single ~D, TOPOLOGY: linear ( ii ) M~T-T'CUT-T~ TYPE: cDNA
(iii) ~r~O.n~-lCAL: NO
(iv) ANTI-SENSE: NO

(vii) IMMEDIATE SOUR OE :
- (B) CLONE: JB661 (xi) ~UL OE DESCRIPTION: SEQ ID NO:6:
AAAAGCTGGG TACC~-AA~T G-.~.CGAGG TCGACGGTAT CG 42 (2) INFORMATION FOR SEQ ID NO:7:
UL.._~ CHARACTERISTICS:
(A~l LENGTH: 43 base pairs (B~ TYPE: nucleic acid (C STFU~NDEDNESS: single (D, TOPOLOGY: linear ( ii ) ~T FC~T ~ TYPE: cDNA
O.~.ICAL: NO
(iv) ANTI-SENSE: NO

(vii) IMMEDIATE SOUR OE :
(B) CLONE: JB662 (xi) S~yu~.._~: DESCRIPTION: SEQ ID NO:7:
GCGAATTGGA GCTCGAA~AT GTTCACCGCG GTGGCGGCCG CTC 43 _

Claims (46)

1. A method for providing templates for DNA sequencing, comprising the steps of:
incubating in vitro: (1) a population of a target DNA, said target DNA comprising a region of DNA to be sequenced, (2) a retroviral or retrotransposon integrase, and (3) an artificial transposon having two termini which are substrates for said integrase, wherein the molar ratio of artificial transposon to target DNA is at least 1:1, to form a population of target DNAs with quasi-randomly integrated insertions of the artificial transposon;
transforming host cells with the population of target DNAs with quasi-randomly integrated insertions of the artificial transposon;
selecting those host cells which have been transformed with a target DNA with an insertion of the artificial transposon;
isolating target DNA with an insertion of the artificial transposon from those host cells which have been transformed with a target DNA
with an insertion of the artificial transposon, said target DNA with an insertion of the artificial transposon being suitable for use as a DNA sequencing template.

2. The method of claim 1 wherein said integrase is yeast retrotransposon Ty1 integrase.

3. The method of claim 1 wherein said target DNA is a plasmid.

4. The method of claim 1 wherein said target DNA is a cosmid.

5. The method of claim 2 wherein said integrase is supplied as Ty1 virus-like particles.

6. The method of claim 2 wherein each of said termini contains Ty1 U3 sequences.

7. The method of claim 6 wherein said termini consist of 4 to 11 base pairs.

8. The method of claim 1 wherein said artificial transposon is provided by restriction digestion with an enzyme which generates blunt ends.

9. The method of claim 8 wherein said restriction enzyme is XmnI.

10. The method of claim 1 wherein said step of transforming is facilitated by electroporation.

11. The method of claim 1 wherein said molar ratio is at least 2.5:1.

12. A method for sequencing DNA, comprising the steps of:
incubating in vitro (1) a population of a target DNA, said target DNA comprising a region of DNA to be sequenced, (2) a retrovirus or retrotransposon integrase, and (3) an artificial transposon having two termini which are substrates for said integrase, wherein the molar ratio of artificial transposon to target DNA is at least 1:1, to form a population of target DNAs with quasi-randomly integrated insertions of the artificial transposon;

transforming host cells with the population of target DNAs with quasi-randomly integrated insertions of the artificial transposon;
selecting those host cells which have been transformed with a target DNA with an insertion of the artificial transposon;
isolating target DNA with an insertion of the artificial transposon from those host cells which have been transformed with a target DNA
with an insertion of the artificial transposon, said target DNA with an insertion of the artificial transposon being suitable for use as a DNA sequencing template;
hybridizing to said isolated target DNA with an insertion of the artificial transposon a primer which is complementary to a terminus of the artificial transposon;
extending said primer to determine a nucleotide sequence of DNA flanking said artificial transposon in said isolated target DNA with an insertion of the artificial transposon.

13. The method of claim 12 wherein said integrase is yeast retrotransposon Ty1 integrase.

14. The method of claim 12 wherein said target DNA is a plasmid.

15. The method of claim 12 wherein said target DNA is a cosmid.

16. The method of claim 13 wherein said integrase is supplied as Ty1 virus-like particles.

17. The method of claim 16 wherein each of said termini is derived from a Ty1 U3 sequence.

18. The method of claim 17 wherein said termini consist of 4 to 11 base pairs.

19. The method of claim 12 wherein said artificial transposon is probided by restriction digestion with an enzyme which generates blunt ends.

20. The method of claim 19 wherein said restriction enzyme is Xmn I.

21. The method of claim 12 wherein said molar ratio is at least 2.5:1.

22. The method of claim 12 wherein said step of transforming is facilitated by electroporation.

23. A method for sequencing DNA, comprising the steps of:
providing a population of target DNAs with quasi-randomly integrated insertions of an artificial transposon, said artificial transposon having termini which are substrates for a retrovirus or a retrotransposon, said population of target DNAs having been formed by in vitro insertion of said artificial transposon into the target DNAs using a retroviral or retrotransposon integrase and a molar ratio of artificial transposon to target DNA of at least 1:1;
hybridizing to individual target DNAs of said population a primer which is complementary to a terminus of the artificial transposon;
extending said primer to determine a nucleotide sequence of target DNA flanking said artificial transposon.

24. The method of claim 23 wherein the integrase is yeast retrotransposon Ty1 integrase.

25. The method of claim 23 wherein said target DNA is a plasmid.

26. The method of claim 23 wherein said target DNA is a cosmid.

27. The method of claim 24 wherein said integrase is supplied as Ty1 virus-like particles.

28. The method of claim 24 wherein each of said termini is derived from a Ty1 U3 sequence.

29. The method of claim 28 wherein said termini consist of 4 to 11 base pairs.

30. The method of claim 23 wherein said molar ratio is at least 2.5:1.

31. A kit for DNA sequencing, comprising:
an artificial transposon having termini which are substrates for a retroviral or retrotransposon integrase;
a retroviral or retrotransposon integrase;
a buffer for in vitro transposition of said artificial transposon, said buffer having a pH of 6 to 8 and 1 to 50 mM of a divalent cation; and a primer which is complementary to a terminus of said artificial transposon.

32. The kit of claim 31 wherein said integrase is yeast retrotransposon Ty1 integrase.

33. The kit of claim 32 wherein said integrase is supplied as Ty1 virus-like particles.

34. The kit of claim 32 wherein said artificial transposon is isolated by digestion with a restriction enzyme which creates blunt ends.

35. The kit of claim 34 wherein said restriction enzyme is Xmn I.

36. An artificial transposon consisting of an isolated, linear, blunt-ended DNA molecule comprising:
a marker DNA;
a sequence of yeast retrotransposon Ty1, said sequence selected from the group consisting of a U5 sequence and a U3 sequence, said sequence being upstream and flanking said marker gene, said sequence consisting of 4 to 11 bp of terminal sequences of said Ty1; and a sequence of yeast retrotransposon Ty1, said sequence selected from the group consisting of a U5 sequence and a U3 sequence, said sequence being downstream and flanking said marker gene, said sequence consisting of 4 to 11 bp of terminal sequences of said Ty1, wherein each of said sequences of yeast retrotranspson Ty1 are at the termini of said linear DNA
molecule.

37. The artificial transposon of claim 36 which is isolated by digestion of a DNA molecule containing said artificial transposon with a restriction enzyme which creates blunt ends when it cleaves DNA.

38. The artificial transposon of claim 37 wherein said restriction enzyme is Xmn I.

39. The artificial transposon of claim 36 wherein the marker DNA is an antibiotic resistance determinant.

40. The artificial transposon of claim 36 wherein the marker DNA is a dihydrofolate reductase gene (dhfr).

41. The artificial transposon of claim 36 wherein the marker DNA is a yeast auxotrophic marker.

42. The artificial transposon of claim 36 wherein each of the sequences flanking the marker DNA consist of the sequence 5'-AACA-3'.

43. The artificial transposon of claim 36 wherein each of the sequences flanking the marker gene are derived from U3 sequences.

44. A DNA molecule useful for generating artificial transposons, comprising:
an origin of replication;
a first selectable marker DNA;
two blunt-ended transposon termini of at least 4 bp each, said termini being substrates for yeast retrotransposon Ty1 integrase, said transposon termini flanking a first restriction enzyme site useful for insertion of a second selectable marker gene to form an artificial transposon;

a second restriction enzyme site flanking said two transposon termini, wherein digestion with said second restriction enzyme liberates a blunt-ended fragment having said transposon termini at either end of the fragment, the fragment thereby liberated being an artificial transposon.

45. A method for in vitro generation of insertions into a target DNA, comprising the steps of:
incubating in vitro (1) a population of a target DNA, (2) a retroviral or retrotransposon integrase, and (3) an artificial transposon having termini which are substrates for said integrase, wherein the molar ratio of artificial transposon to target DNA is at least 1:1, to form a population of target DNAs with quasi-randomly integrated insertions of the artificial transposon;
transforming a host cell with the population of target DNAs with quasi-randomly integrated insertions of the artificial transposon;
selecting those host cells which have been transformed with a target DNA with an insertion of the artificial transposon.

46. The method of claim 46 wherein the molar ratio of artificial transposon to target DNA is at least 2.5:1.