USOO5405775A United States Patent (19) 11 Patent Number: 5,405,775 Inouye et al. 45 Date of Patent: Apr. 11, 1995

54 RETRONS CODING FORHYBRID ATCC Catalog (1985), pp. 66-79. DNA/RNAMOLECULES Yee et al. Multicopy Single-Stranded DNA. Isolated from a Gram-Negative Bacterium ... etc. Cell 38, pp. 75 Inventors: Masayori Inouye; Sumiko Inouye, 203-209 (1984). both of Bridgewater, N.J. Dhundale et al. Distribution of Multicopy Single-S- 73) Assignee: The University of Medicine and tranded DNA amoung Myxobacterial and Related Spe Dentistry of New Jersey, Piscataway, cies J. Bacteriol. 164, pp. 914-917 (1985). N.J. Furuichi T. et al. Biosynthesis and Structure of Stable 21 Appl. No.: 518,749 Branched RNA Covalently . . . etc. Cell 48, pp.55-62 (1987). 22 Filed: May 2, 1990 Dhundale et al. Structure of msDNA from M. xanthus ... etc. Cell 51, pp. 1105-1112 (1987). Related U.S. Application Data Dhundale et al. Mutations that Affect Production of 63 Continuation-in-part of Ser. No. 315,432, Feb. 24, 1989, Branched RNA-Linked msDNA.M. xanthus ... etc. J. abandoned, Ser. No. 315,427, Feb. 24, 1989, Pat. No. Bacteriol., pp. 5620–5624 (1988). 5,079,151, Ser. No. 315,316, Feb. 24, 1989, Pat. No. Lin and Maas, Reverse Transcriptase-Dependent Syn 5,320,958, and Ser. No. 517,946, May 2, 1990. thesis of a Covalently Linked . . . etc. Cell 56, pp. 51. Int. Cl...... C12N 1/21; C12N 15/70, 891-904 (Mar. 1989). C12N 15/52 Primary Examiner-Richard A. Schwartz 52 U.S.C...... 435/252.33; 435/320.1; Assistant Examiner-James Ketter 536/23.2; 536/25.2 Attorney, Agent, or Firm-Weiser & Associates 58) Field of Search ...... 536/27, 33.1, 35.2, 536/23.2, 25.2; 435/91, 172.3, 172.1, 252.3, 57 ABSTRACT 252.33, 320.1 A multicopy single-stranded DNA (msDNA) synthe sizing system in E. coli is disclosed. The use of the 56) References Cited msDNA system to synthesize cDNA in vivo is dis PUBLICATIONS closed. Construction of synthetic msDNA is also dis Masui et al., “Multipurpose Expression Cloning Vehi closed. Also processes for gene amplification and for cles in Escherichia coli', in Experimental Manipulation of producing a stable RNA are disclosed. Gene Expression, Academic Press, New York, 1983, Inouye, ed., pp. 15-32. 29 Claims, 17 Drawing Sheets U.S. Patent Apr. 11, 1995 Sheet 1 of 17 5,405,775

cy c) N EcoRI Sn N Pst 1(a) S Pvu II v ( \ Q V U V S Bol I \ S \ ( V Vs Hindo I I I \ V V Q NL Hind I I I s Pst I (a) Pst I (b)

Hind III Pst I (b) S Pst I (c) s Pst I (d) ----- :EcoRI K-12.4% EcoRI

FIG. 1 U.S. Patent Apr. 11, 1995 Sheet 2 of 17 5,405,775

5

U Ao 50 A

FIG. 2A U.S. Patent Apr. 11, 1995 Sheet 3 of 17 5,405,775

FIG. 2B

•

U.S. Patent Apr. 11, 1995 Sheet 5 of 17 5,405,775

pINIII(lppp.)Ec67-RT 1O. 7kb

Xbo I BAP + Synthetic msDNA (196bp)

1 ppp-NS loc PO

-- RTV pINIII(lppp.ms10ORT 1.O. 9kb

FIG. 4

U.S. Patent Apr. 11, 1995 Sheet 7 of 17 5,405,775

Dro I # Synthetic msl)NA (196bp) Xbo I Xbo I Soc II SocII Km R IacP ox Xibal, Sac II pUCKms 1 OO anem-ame

+ ms C1, C2 FRAGMENTS

pUCKns 101

FIG. 6

U.S. Patent Apr. 11, 1995 Sheet 9 of 17 5,405,775

I -Sloc Z' pUC19-Ec67-20

FIG. B U.S. Patent Apr. 11, 1995 Sheet 10 of 17 5,405,775

1-yC Idc Z' pUC19-Ec67-20

TRANSCRIPTION OF Icic Z

5' 80 BASES 20 BASES 3' mRNA cDNA synthesis in vivo

msDNA 3' cDNA 5' st an ame a so - - w

mRNA Primer for PCR (P1, P2)

d KC Pf P2 ApCR

H P X U.S. Patent Apr. 11, 1995 Sheet 11 of 17 5,405,775

U.S. Patent Apr. 11, 1995 Sheet 12 of 17 5,405,775

U.S. Patent Apr. 11, 1995 Sheet 13 of 17 5,405,775

U.S. Patent Apr. 11, 1995 Sheet 14 of 17 5,405,775

U.S. Patent Apr. 11, 1995 Sheet 15 of 17 5,405,775

IGG CA TO AGA TAC GGA TTT ICA CTTCCT. IGA (AG 16 AIGACTAIG CIG CAT GAAAIC 60 GA IA ICATT GAG (AT CT CIT ICC ICA GAT (GCA GAA CIG GC GC TTT IGC ICA 120 IGT CAT GA IGT GA IGAAAA CA CIG (ATAAA G. GGC AGG CCI GGC GGG GATACGAGC 180 GCG (GC IAT CAC CAAAA IAG CAAAA IAC IIC IGGAAAACA GAA AGT ICA ACT GATAIC 240 ce RNA O2 TIC AIA AACAC CAT GIA CAC III GT GT ICT CAIC CACA GIG GIA AIG 300 AAG AT TIG 76 CIA (ATCO ICT AAA (AA (AACA CTTAGC CTT GT CAC CGG AAT IAC GAGGGGAAIC GCC. ICC CIA AAA ICC IIG ATT (AGAGC IATACG GCA GT GIG CIG IGC 360 (GT CCT (CT TAG CAS GATTTTAGGAA. IAA GIC 106 AIA 16 CCT CA (AC GAC ACG -- ol GAA GA CIG CCT GCA IGC GTT ICT (CI TGG CCT IIT TIC CIC IG GAT GAA GAA GAA AIG 420 CIT CICAC GGA CGT ACG CAA AGA CGA ACC CGAAAAAAG GAG ACC CIA CIT CTI CTT IAC K Ko DNA ACA AAA ACA ICT AAA CTTGAC GA CTTAGG GCI GCIACT ICA CGI GAA GAC TIG GCTAAA 480 T K T S K L D A L R A A T S R E D L. A. K. AIT IIA G4T AIT AAG IIG GIA IIT IIA ACT AAC GT CIA TAT AGAATCGGC ICG GAT AAT 540 I L. D 1 K L V F L J W W. L. Y. R. I. G. S D W CAA IAC ACT CAA TTT ACA AIA (GAAGAAA GAAAA GGG GIA AG ACTATT ICT GCA CT 600 0 Y T 0 F T 1 P K K G K G W R T | S A P abO ACA GAC CGG, TIGAAG GACAIC CAA CGA AGA AIA IGT GAC IIA CTI ICT GAT ICT AGA GAT 660 T D R L. K. D 1 0 R R N C D L L S D C R D CAGAIC TTT GCTATA AGAAAAITAGTAACAAC IAT ICC TTT GCI ITT GAG AGG GAAAA 720 E / F A 1 R K 1 S M N Y S F. G. F. E. R. G. K. at 100 ICA AIA AIC CIA AAT GCT IATAAG CAT AGAGCCAAA (AAAAAIA TIA AAIAIA GT CIT 780 S 1 I L N A y R H R G K Q 1 I L N J D H AG CAT IIT III GAAGC IITAT IIT G CoA GT AG. G. IAT III CTT ICC AAI (AG 840 K D F F. E. S. F. N. F. G R W R G Y F L S N 0 GATTTTTIA TIA AAT (CT GIG GIG GA ACG ACA CTTGAAAA GCT CA IGC IATAATGA 900 D F L L N P W W A T T L A K A A. C. Y. M. G. af30 FIG. 13A U.S. Patent Apr. 11, 1995 Sheet 16 of 17 5,405,775

ACC CIC CCC CAA CGA AGT CCA IGI ICT CCTATTAIC ICA AAT CIA ATT IGC AATATTAIG 960 T L P O G. S P C S P 1 1 S M L I C W 1 K GAI AIGAGA TIA GCTAAG CIG GCI AAAAAA IAI GA IGT ACT IATAGC AGA IAT GCT GAT 1020 D N R L. A K L A K K Y C C T Y S R Y AD a 200 GAI AIA ACA AIT ICT ACA AATAAA AAT ACA ITT CCG IIA GAA AIG GCIACT GIG CAA CCT 1030 D I T 1 S T N K N T F P L E N A T W G P GAA GGG GTI GTT TIG GAAAA GTI IIG GIA AAA GAA AIA GAA AAC ICT GGA TIC GAA AIA 1140 E. G. W. W. L. G K W L W K E.E / E. W. S G F E I AAI CAT ICA AAG ACTAG CTIAC IATAAG ACA IA AG (AA GAA GTAACGGGA CTTACA 1200 N D S K J R L J Y K I S R 0 E W T G L T a250 GTTAAC AGA AJC GTTAAT ATI GAT AGA TGT IAT IATAAAAAA ACT (GG GCG TIG GCA W M R / W N I D R C Y Y K K I R A L. A CA A 1 Y R T G E Y K W P D F M G W L V S

GT CIG GATAAA CTTGA6 GG AIG TTT GT TITATIGAT (AA GTT (ATAAG TTT AACAAT 1380 G. L. D. K L E G N F G F | D 0 W D K F W W AIA AAGAAAAAA CIG AACAAG CAA CCT GAT AGA TAT CIA IIG ACT AAT GOG ACT TIG CAT 1440 I K K K L. M K C P D R Y W L T W A T L H GGT TITAAA TIA AAG TIGAATGG (G GAA AAA GA IATACTAAA TTIATI IAC IATAAA 1500 G F K L. K. L. M. A. R. E. K. A. Y S K F | Y Y K 50 TTI ITT CAT GOC AAC ACC IGT (CTACGAIA AIT ACA GAA. GGG AAG ACT GAT CGG AIA IAT 1560 F F H G W T C P T I I I E G R T D R Y TIGAAG GCT GCT TIG CAT ICT TIG GAG ACA ICA IAT (CTGA6. TIG ITT AGAGAAAAAACA 1620 L. K. A A L H S L E T S Y P E L. F. R. E. K. T st400 GAT AGAAAAAGAAA GAAAAAAI CTTAAT AIA ITT AAA ICTAATGAAAAG ACCAAA IAT 1680 D S K K K E 1 N. L. M I F K S N E K T K Y Ap IIT IIA GAT CIT ICT GGG GG ACT GOGAT CIGAAAAAA IIT CIA GAGGA CGT TATAAA AAT 1740 F. L. D. L. S. G G T A D L. K. K. F. E R Y K W FIG. 13B U.S. Patent Apr. 11, 1995 Sheet 17 of 17 5,405,775

AAI IAI GCI ICI IAI IAI GCI ICT GTI (AAAA (ACCCA GIGATTAIGCTI CTTGATAAT 1600 N. Y. A S Y Y G S W P K 0 P W I W. W. L. D. W at 450 GAT ACA GCTCCAAG GAT TIA CTTAAI ITT CIG CGC AATAAA GTIAAA AGC IGC CA. GAC 1860 D I G P S D L. L N F L R N K V K S C P D GAT GIA ACT GAA AIGAGAAAGAIGAAA IATATT CAT CIT TIC IATAAT TIA IATAIA CTT 1920 D W T E M R K M K Y / H / F Y W L Y I W #500 CIC ACA CATIGAGT CCI ICC GGC GAA CAA ACT ICA AIG GAG GAT CIT TIC (CTAAA GAT 1930 L J P L S P S G E 0 I S. M ATI TIA GATAIC AAG ATI GAT GGTAAGAAA TIC AAC AAA 2 0.40 1 I D I K I D G K K F. M. K N D ACG GAA IAI GGG AAG CAT ATI ITT ICC AIG AGGGTT GIT AGA GATAAA T E Y C K H 1 F S W R V V R D K #550 GAT TITAAG GCA TTI IGT IGTATT ITT GAT GCTAIA AAA (AIAIA AAG GAA CAT IATAAA 2160 D F K A F C C D F D A 1 K D I K E H Y K TIA AIG, TIAAAIAGC IAA IA ACA GCC CIA ACG TA IGAAC CTA AG CIGATT ITT CGT2220 L. W. L. N. S. IAA AAT TIA (A7 GCT TIGAAT TGT AATATA TIA ICT TOAAG AT TIA TTIAAT ICC IGC 2280 AICCTT TIC ICTAAG GTATTAAT ICG TIC CIC AC AAC ACT AAA CIC GCT IIT ICC ACA 240 ICC COAAAC CCC (CTAAC ATTAIT (GG CATAATCO CAT CAT II CGIG (ACAC AIG 2400 CGC ICC CAT CAT CIC AIC GCGCC

FIG. 13C 5,405,775 1. 2 linked msDNA, and predicted that a reverse transcrip RETRONS CODING FOR HYBRD DNA/RNA tase (RT) is required for this reaction (2). MOLECULES Initial studies indicated that msDNA is not found in the common E. coli K-12 laboratory strain (1). To date, This is a continuation-in-part of patent application it has been observed that approximately 6% of all E. coli Ser. No. 07/315,432, filed Feb. 24, 1989, now aban isolates from clinical strains carry anmsDNA synthesiz doned, by Lampson et al. and also of two patent appli ing system. This synthesizing system has been classified cations co-filed on even date, Feb. 24, 1989, U.S. Ser. as a retron on the basis of rather surprising similarities No. 07/315,427, now U.S. Pat. No. 5,079,151, U.S. Ser. between themsDNA and retroviruses and retrotranspo No. 07/315,316 now U.S. Pat. No. 5,320,958 and U.S. O sons (8). Ser. No. 07/517,946, filed May 2, 1990 and relates to a The present invention provides for an E. colimsDNA pending patent application filed May 2, 1990, entitled synthesizing system. The invention also provides for its “Prokaryotic Reverse Transcriptase” by Masayori products and uses. Inouye and Sumiko Inouye which are incorporated herein by reference. 15 BACKGROUND ART The parent application discloses the presence of Bacterial reverse transcriptase and msDNA were msDNA in the clinical E. coli isolate, C1-1, the cloned initially discovered in Myxococcus xanthus and another Petton from the same strain and its nucleotide sequence. myxobacterium Stigmatella aurantiaca. The publica The instant continuation-in-part application addition tions noted here report on the myxobacteria discover ally discloses a retron from another E. coli clinical iso 20 ies. All such references are hereby incorporated by late, C1-23, a process for in vivo cDNA production in reference. E. coli and synthetic msDNA molecules. Additional Yee, T. and Inouye M. "Reexamination of the Ge disclosures of processes for gene amplification and pro nome Size of Myxobacteria, Including the Use of a New duction of stable RNA are also made. Large production Method for Genome Size Analysis', J. Bacteriol. 145, of proteins is made possible by the invention. 25 pp. 1257-1265 (1981), reports the discovery of a rapidly renaturing fraction of DNA found during the study of FIELD OF THE INVENTION Myxobacteria genome size. This invention relates to a prokaryotic msDNA (mul Yee, T. et al., “Multicopy Single-Stranded DNA ticopy single-stranded DNA) synthesizing system, also Isolated from a Gram-Negative Bacterium, Myxococcus known as the retron. The invention also relates to 30 xanthus, Cell 38, pp. 203-209 (1984), reports that the msDNAs and to their production and their use to syn rapidly renaturing DNA found in Myxococcus xanthus is thesize cDNA. The invention further relates to the use found as a satellite band upon polyacrylamide gel elec of one or more retron components in the production of trophoresis. This satellite DNA was called msDNA. various Ins)NAs. Myxococcus xanthus was found to contain 500 to 700 35 copies of msDNA per chromosome. The msDNA was BACKGROUND OF THE INVENTION cloned and sequenced. Its length and secondary struc A novel satellite DNA called msDNA (multicopy ture was determined. A similar satellite DNA was single-stranded DNA) was originally found in Myxococ found in the myxobacterium Stigmatella aurantiaca. The cus xanthus, a Gram-negative bacterium living in soil authors report that they were unable to detect any satel (1). It consists of a 162-base single-stranded DNA, the 5' lite DNA in Escherichia coli K-12. end of which is linked to a branched RNA (msdRNA) Furuichi, T. et al., “Branched RNA Covalently of 77 bases by a 2',5'-phosphodiester linkage at the 2' Linked to the 5' End of a Single-Stranded DNA in position of the 20th rC residue (2). There are approxi Stigmatella aurantiaca: Structure of msDNA", Cell 48, mately 700 copies of msDNA per genome. msDNA is pp. 47-53 (1987) and Furuichi, T. et al., "Biosynthesis widely distributed among various myxobacteria includ 45 and Structure of Stable Branched RNA Covalently ing the closely related Stigmatella aurantiaca which Linked to the 5' End of Multicopy Single-Stranded possesses an imsDNA, msDNA-Saló3. This molecule is DNA of Stigmatella aurantiaca”, Cell 48 pp 55-62 highly homologous to msDNA-Mx162 from M. xanthus (1987), showed that msDNA isolated from S. aurantiaca (3, 4). It is noteworthy that several M. xanthus strains, (type Sal63) contained a DNA portion that was linked independently isolated from different sites, all contain 50 to an RNA molecule (msdRNA) by a 2, 5'-phosphodi msDNA (5). Recently it was found that M. xanthus ester bond. The authors also reported that the coding contains another smaller species of msDNA called region for msdRNA (mst) is located downstream of the msDNA-Mx65 (6). In contrast to the close homology coding region for msDNA (msd). The coding regions between msDNA-Mx162 and msDNA-Saló3, there is were found to exist in opposite orientation with respect no primary sequence homology between msDNA 55 to each other with their 3' ends overlapping. Mx162 and the small molecule, msDNA-Mx65. How Dhundale, A. R. et al., “Distribution of Multicopy ever, it was found that msDNA-Mx65 does share key Single-Stranded DNA among Myxobacteria and Re secondary structures such as a branched rC residue, a lated Species', J. Bacteriol. 164, pp. 914-917 (1985), DNA-RNA hybrid at the 3' ends of the msDNA and examined how widely msDNA exists in various bacteria msdRNA, and stem-loop structures in RNA and DNA closely and distantly related to M. xanthus. msDNA strands. was found in other myxobacteria and nine indepen It has been further shown that msdRNA is derived dently isolated strains of M. xanthus. The authors report from a much longer precursor RNA (pre-msdRNA), msDNA to be found in certain gliding bacteria but not which can form a very stable stem-and-loop structure in others. (2). A novel mechanism for msDNA synthesis was 65 The references cited above do not disclose or suggest proposed, in which the stem-and-loop structure of pre that msdNA exists in E. coli. The publication of Yee et msdRNA serves as a primer for initiating msDNA syn al. in Cell 38, 203 (1984) indicates that msDNA was thesis as well as a template to form the branched RNA undetectable in E. coli K-12 strain. The present inven 5,405,775 3 4. tion encompasses recombinant DNA constructs encod bacteria producing msDNA can be used for the screen ing an E. colimsDNA synthesizing system and the com ing of antibodies and chemicals which block reverse ponents thereof. The unexpected discovery that about transcriptase activity. 6% of E. coli clinical isolates examined to date harbor msDNA enables the present invention. The present 5 DETAILED DESCRIPTION OF THE invention is thus a novel departure from the back INVENTION ground art. This invention relates to a prokaryotic msDNA syn thesizing system. This genetic system has been found in BRIEF DESCRIPTION OF THE FIGURES E. coli isolated from individuals with blood and urinary FIG. 1 shows a restriction map of an E. coli retron 10 tract infections and those that are apparently healthy. and flanking sequences. The DNA fragment encoding the whole genetic system FIGS. 2A and 2B show the complete primary and of msDNA has been classified as a retron since it ap proposed secondary structure of msDNA-Ec67 and pears to represent a primitive form of retroelement (8). msDNA-Ec74. It is proposed that the function of msDNA in the cell FIG. 3 shows the synthetic msDNA genes. 15 may be to serve as a primer to produce cDNA and the FIG. 4 shows the construction scheme for p retron may function as a transposable element. FIG. 1 NIII(lppp-5) ms100-RT. shows a restriction map of the retron. The DNA strand FIG. 5 shows synthetic msDNA ms100. of themsDNA molecule is coded for by the msd gene. FIG. 6 shows the construction scheme for The RNA molecule of msDNA is encoded by the misr pUCKms100 and puCKms101. 20 gene. The two genes are convergently situated (5' to 3) FIG. 7 shows synthetic msDNA ms101. such that their respective 3' ends overlap. The third FIG. 8 shows the puC19 derivative used for in vivo retron component is an open reading frame (ORF) lo cDNA production, pUC19-Ec67-20mer. cated upstream of msd and downstream of msr encod FIG. 9 shows the steps of in vivo cDNA synthesis. ing a protein with reverse transcriptase activity (9). It is FIG. 10 shows the detection of msDNA in a clinical 25 proposed that other ORFs may exist within the retron. isolate of E. coli These other ORFs may share sequence similarities with FIG. 11 shows the sequence determination of the retrovital proteins such as integrase, protease and gag branched RNA linked to msDNA. proteins. FIG. 12A and 12B show a DNA blot analysis of E. A population of E. coli clinical strains carry msDNA coli chromosomal DNA and analysis of msDNA syn- 30 synthesizing systems. At present, retrons have been thesis. found in approximately 6% (7 out of 113) of the clinical FIG. 13 shows the nucleotide sequence of an strains analyzed. It is contemplated that retrons exhibit msDNA-Ec67 retron. ing structural similarities exist in other genera of the SUMMARY OF THE INVENTION Enterobacteriaceae family. 35 Retrons from two E. coli clinical strains have been Methods and compositions are provided for produc sequenced. The RNA and DNA sequence of the tion of msDNA. The invention enables production of msDNAs produced by these retrons has also been deter natural and synthetic msDNA. mined. The complete primary and proposed secondary The invention provides for an imsDNA synthesizing structure of these molecules (Ec67 and 74) are shown in system. The three components of this system can be 40 FIG. 2. The numeric designation indicates the length of cloned in an E. coli expression vector as a unit or sepa the DNA molecule. Little sequence homology is ob rately. The source of these components may be natural served in both the RNA and DNA components of these or synthetic. The components can also be utilized as molecules. However, despite their primary sequence they exist on the prokaryotic chromosome. differences, E. coli msDNAs all share key functional The method of the invention provides for the utiliza- 45 common features which include a single-stranded DNA tion of the prokaryotic msDNA synthesizing system. with a stem-and-loop structure, a single-stranded RNA The synthesizing system (retron) has three components, with a stem-and-loop structure, a 2, 5'-phosphodiester msd, msrand an ORF. Transcription and translation of linkage between RNA and DNA, and a DNA-RNA the ORF region results in production of a protein hav hybrid at the 3' ends. ing reverse transcriptase activity. Transcription of the 50 The invention also relates to the use of retron compo msr region followed by DNA synthesis by reverse tran nents in the production of various msDNAs and reverse scriptase results in msDNA production. transcriptases. The retron of the invention can be natu The method of the invention provides for cDNA ral or synthetic. Two entire nsr-msd regions have been production within the cell. The invention provides an in synthesized using synthetic oligonucleotides and an vivo system to produce cDNA complementary to a 55 example is illustrated in FIGS. 3 and 4. The region was specific RNA transcript in E. coli. Upon insertion of a inserted into a pINIII vector (14) (as a form of double sequence complementary to the 3' end of a msDNA stranded DNA) such that a synthetic pre-msdRNA was molecule into a specific mRNA, cDNA to the mRNA is produced in response to the addition of a lac inducer. produced in vivo using msDNA as a primer. It is con The total gene length of approximately 200-bp was templated that cDNA could also be produced in vitro constructed by four units of double-stranded oligonu by providing an appropriate RNA, msDNA and reverse cleotides. The gene was inserted into the unique Xbal transcriptase. site of the vector. The RT gene was provided in cis by The invention also contemplates additional uses of inserting it into the same plasmid or in trans by inserting artificial retrons as tools in life sciences research. Arti it in a separate plasmid. It is thought that the retron can ficial msDNAs are contemplated to be useful for gene 65 be utilized on the chromosome or extra-chromoso amplification, mRNA stabilization and production of mally. It is contemplated that a naturally occurring ribozymes and antisense RNAs. Additionally, owing to retron can be altered through genetic engineering tech ease of detection of msDNA, it is contemplated that niques. 5,405,775 5 6 It is proposed that various artificial msDNAs can be ture at the 3' end may be able to prime cDNA synthesis constructed within the limitation of the requirements by itself if an RT gene is expressed in the cell. stated above. It is further proposed that the 3' end of Another contemplated approach is to use exoge msDNA can be variable, this part of the sequence can nously added synthetic oligonucleotide as primers be substituted with a complementary sequence to a which are complementary (antisense) to the mRNA. It specific mRNA. Such an imsDNA may be able to serve is proposed that cells permeabilized with organic sol as a primer for the production of cDNA for a specific vents will be useful for this method. mRNA. The system of the invention is useful in various appli msDNA-Ec67 retron is able to synthesize cDNA if cations. Since msDNA is produced in several hundred cells contain an mRNA which has a stretch of RNA O copies per retron, the system can be used for gene am sequence complementary to the 3' end of msDNA plification. This can be achieved by replacing the dou Ec67. A plasmid was constructed from puC19 (13) ble-stranded region of msDNA with another double which was able to produce an mRNA containing a . stranded DNA containing a gene. In the synthetic sequence complementary to the 5' end of msDNA-Ec67 msDNA depicted in FIG. 5, the stem-and-loop region (FIG. 8). The sequence contained the 15-base sequence 15 (double-stranded region) of msDNA can be removed by identical to the 3' end of msDNA-Ec67 such that the restriction enzyme digestion of the retron-containing RNA transcript from puC19 contains the 15-base se plasmid DNA with XhoI and SacII. A new DNA frag quence complementary to the 3' end of msDNA-Ec67 ment is then ligated to this site, which contains two at position 80. copies of a gene of interest either in head-to-head or in When E. coli harboring the Ecó7 retron capable of 20 tail-to-tail orientation. As a result, when this region is synthesizing msDNA-Ec67 is transformed with plasmid copied as a single-stranded DNA in a synthetic pUC19-Ec67-20mer, the 3'-end region of Ec67-msDNA msDNA, a secondary structure or a stem-and-loop forms DNA-RNA hybrids not only with the 3' end of structure is formed because of palindromic orientation msdRNA (as shown FIG. 2) but also with the RNA of the two copies of the gene. Thus, the gene of interest transcript from pUC19-Ec67-20-mer as shown in FIG. 25 is reconstructed in the stem structure. By this method of 9. Since the cells contain Ec67-RT, this enzyme starts to gene amplification, a large number of copies of the gene synthesize cDNA by extending the 3' end of msDNA (e.g., more than 4,000), can be produced. This is pro along the mRNA template. A single-stranded DNA is vided that the plasmid containing msDNA sequences is synthesized of 152 bases which consists of the 67-base maintained in E. coli at a copy number of 20 and that msDNA at the 5' end and the 85-base cDNA to the 5' 30 each plasmid produces 200 transcripts of themsDNA in end of the lac transcript at the 3' end. Identification of a steady state. Of course, since themsDNA structure is this cDNA was made by polymerase chain reaction not foreign to E. coli, the microorganism is particularly (PCR) (21). The results in FIG. 10A indicated a good well suited as the vehicle for gene multiplication. agreement with the predicted 150-bp cDNA depicted in In another application, themsDNA of the system of FIG. 9. An identical result was obtained with cells 35 the invention are used to produce stable RNA. A DNA transformed with a pINIII vector (14) which also con fragment can be inserted in the Xbal site located in the tained the same 20-bp sequence (FIG. 10, lanes 3 and 5). RNA structure (see FIG. 3A). When the resulting re The results described above are consistent with the tron is transcribed, RNA from the inserted DNA is cDNA structure depicted in FIG. 9, in which msDNA added in the Xbal site of msdRNA. When the inserted primes cDNA synthesis. To unambiguously prove this DNA contains an open-reading frame, then the newly model, the DNA sequence of the PCR product with formed msdRNA functions as an mRNA containing the pUC19-Ec67-20-mer (FIG. 10A, lane 2) was deter open-reading frame to produce a polypeptide. If the mined. FIG. 11 shows the DNA sequence of the junc same DNA fragment is inserted in the opposite orienta tion site, which clearly demonstrates that the 3' end of tion, the newly synthesized msdRNA contains an RNA Ec67-Ins)NA is connected to the cDNA of the lac 45 sequence complementary to the mRNA. Thus, it works transcript of pUC19 at the 85th position. as the antisense RNA against the mRNA, so that it can The present results unambiguously demonstrate that be used to regulate the expression of the gene for the cDNA to a specific RNA transcript can be synthesized mRNA. The RNA produced contains a Shine-Dal in E. coli cells by the method of the invention. This garno sequence, an initiation codon and the coding further indicates that cells are capable of producing 50 sequence. This mRNA is extremely stable because of cDNA if they contain RT and appropriate primers for a the 3' DNA-RNA hybrid structure, the secondary specific template. structure of the mRNA at the 3' end, and the branched cDNA detected in the present study seems to exist rG residue at the 5' end. All these structures are consid mostly as single-stranded DNA, since cDNA produc ered to protect the mRNA from degradation. tion was detected by PCR after RNaseA treatment but 55 Thus, the invention provides a very useful system not after treatment with S1 nuclease (See Example 5). whereby a large amount of a specific mRNA is pro Conversion of single-stranded cDNA to double duced in a cell, resulting in expression of a large quan stranded cDNA may, however, easily occur in the cells tity of a specific polypeptide from the cloned gene. The if appropriate primers are provided. It is contemplated industrial applications of the system are evident. High that an imsDNA-synthesizing system could be estab volume of a desired peptide can be comparatively inex lished in eukaryotic cells. It is further contemplated that pensively produced from the corresponding selected such a system may be used to obtain cDNA to a specific gene. Numerous valuable polypeptides can be produced RNA transcript in vivo or cDNA to polyadenylated like interferon, erythropoietin, plasminogen activators, mRNAs in vivo by properly engineering the 3' ends of antiplatelet aggregants, interlukin, growth and other msDNA. 65 hormones; and other biologically useful proteins. It is also proposed that E. coli RT can synthesize Another important application of the invention is that cDNA from an mRNA if an appropriate primer is pro themsDNA be used to construct ribozymes or antisense vided. An mRNA having a stable stem-and-loop struc RNAs or their combination. A DNA fragment can be 5,405,775 7 8 inserted in the Xbal site so that the msdRNA synthe with a urinary tract infection, three were found to con sized from this construct contains a so-called hammer tain msDNA. From patients with blood infections, 3 head structure which works as a ribozyme, i.e., a ribo strains were found to contain msDNA. In addition, nuclease which cleaves a specific RNA. Such a ribo msDNAs have been found in E. coli strains from appar zyme can be used to destroy a specific mRNA. The ently normal human stool samples; msDNA was not Xbal site shown in FIG. 3 (see also FIG. 5) can be observed in the E. coli K-12 strain, C600. utilized for this purpose. If a hammerhead structure from a plant viroid (15-17) can be formed in the EXAMPLE 2 msdRNA at this site, a ribozyme is formed in msdRNA, Nucleotide sequence of msDNA. which functions as a sequence specific ribonuclease. 10 Similarly, if an antisense RNA against a specific gene is To determine the base sequence of the DNA mole inserted at this site, themsDNA-antisense RNA may be cule, the RNA-DNA complex isolated from the clinical very effective in blocking the expression of a specific stain was labeled at the 3' end of the DNA molecule gene. It is thought that a better suppression effect may with AMV-RT and a-32PdATP. By adding ddCTP, occur upon combination of both ribozyme and antisense 15 ddTTP, and ddGTP to the reaction mixture, a single RNA within a single msdRNA. This approach leads to labeled adenine is added to the 3' end of the DNA mole a new method for constructing more effective antisense cule. RNA is removed with RNase A-T and the end RNA. The ribozyme functions to cleave selected other labeled DNA is subjected to the Maxam and Gilbert RNA molecules, e.g., specific vital RNAs. This antivi sequencing method (3). FIG. 2 shows that this msDNA ral approach can be usefully applied to a ribozyme spe 20 consists of a single-stranded DNA of 67 bases and that cific as anti-HIV agent. The practical applications are it can form a secondary hair-pinstructure. Accordingly, evident in this area. this msDNA has been denoted as msDNA-Ec67. It is also proposed that E. coli producing msdNA can The sequence of the RNA molecules was determined be used for the screening of antibodies and chemicals using the RNA-DNA complex purified from E. coli which block RT activity. It is thought that anti-RT 25 C1-1 as described in Example 1. As shown in FIG. 11, compounds will show stronger inhibitory effects on a large gap is observed in the RNA sequence "ladder'. msDNA synthesis than on chromosomal DNA synthe This gap is due to the DNA strand branched at the 2' SS. position of the 15th rC residue of the RNA strand The following examples illustrate the detection of which produces a shift in mobility of the sequence lad msDNA in E. coli, nucleotide sequencing of msDNA and cloning of themsDNA genetic locus. ThemsDNAs 30 der (see FIG. 2). The RNA consists of 58 bases with the and reverse transcriptases described herein are not lim DNA molecule branched at the Gresidue at position 15 ited to those specifically described herein. It can readily by a 2, 5'-phosphodiester linkage. The branched G be seen by those skilled in the art that various msDNA structure was determined as described for msDNAs molecules can be produced through synthetic means or from myxobacteria (5, 6). After RNase (A and T1) treat genetic engineering. 35 ment, msDNA retains a small oligoribonucleotide The following examples are only given for purposes linked to the 5' end of the DNA molecule due to the of illustration and not by way of limitation on the scope inability of RNases to cleave in the vicinity of the of the invention. branched linkage. The 5' end was labeled with y-32P ATP using T4 polynucleotide kinase and the labeled EXAMPLE 1. RNA molecule was detached from the DNA strand by Detection of msDNA in E. coli. a debranching enzyme purified from HeLa cells (5, 6). This small RNA was found to be a tetraribonucleotide Fifty independent E. coli urinary tract isolates identi which could be digested with RNase T1 to yield a la fied with the use of the API-20E identification system beled dinucleotide. Since RNase T1 could not cleave the (9) were examined for the presence of msDNA. Since 45 RNA molecule at the G residue before debranching msDNA contains a DNA-RNA duplex structure, the 3' enzyme treatment, it was concluded that the single end of the DNA molecule serves as an intramolecular stranded DNA is branched at the G residue via a 2, primer and the RNA molecule as a template for RT. 5'-phosphodiester linkage. In addition, partial RNase When RNA prepared from one of the clinical strains, E. U2 digestion cleaved the RNA molecule to yield a 32P coli C1-1, was labeled in this manner, two distinct, low 50 molecular weight bands of about 160 bases became labeled mono- and a 32P-labeled trinucleotide. Thus, the labeled with 32P and are shown in FIG. 10. If the la sequence of the tetranucleotide is 5A-G-A-(U or C). beled sample is digested with ribonuclease (RNase). A Based on these data, the complete structure of msDNA prior to loading on the gel, a single band corresponding Ec67 from E. coli C1-1 is presented in FIG. 2. Despite to 105 bases of single-stranded DNA is detected (lane 55 a lack of primary and structural homology, msldNA 4). This indicates that both bands in lane 3 contain a Ec67 displays all the unique features found in msDNAs single-stranded DNA of identical size. The two labeled from myxobacteria. These include a single-stranded bands observed prior to RNase treatment (lane 3) are DNA with a stem-and-loop structure, a single-stranded due to two species of msDNA comprised of a single RNA with a stem-and-loop structure, a 2, 5'-phos species of single-stranded DNA linked to RNA mole 60 phodiester linkage between the RNA and DNA, and a cules of two different sizes. Among the fifty clinical DNA-RNA hybrid at their 3' ends. This hybrid struc isolates screened, three other strains produced msDNA ture was confirmed by demonstrating sensitivity of the like molecules of varying size and quantity suggesting RNA molecule to RNaseH. extensive diversity among these molecules. EXAMPLE 3 In a similar experiment, RNA was extracted from 113 65 independent clinical isolates. Fifty were from patients Cloning of the locus formsDNA. with a urinary tract infection, and 63 from patients with In order to identify the DNA fragment which is re blood infections. Among the 50 strains from patients sponsible for msDNA synthesis in E. coli C1-1, DNA 5,405,775 10 blot hybridization (18) was carried out with various in length were synthesized. The appropriate pairs of restriction enzyme digests of total chromosomal DNA oligonucleotides were annealed by heating at 100 C. prepared from E. coli C1-1, using msDNA-Ec67 labeled for 5 minutes, then cooling at 30° C. for 30 minutes and with AMV-RT (the same preparation as shown in lane for 30 minutes in a refrigerator. An E. coli plNIII(lppp 3, FIG. 10) as a probe. For each lane, 3 ug of the DNA 5) expression vector (14) was digested with Xbal-EcoRI digest was applied to a 0.7% agarose gel. The result is and an Xbal-EcoRI fragment from the clinical E. coli shown in-FIG. 12A EcoRI (lane 1), HindIII (lane 2), strain C1-1 was inserted such that the RT gene under BamHI (lane 3), Pst (lane 4) and Bgll (lane 5) diges lipp-lac promoter control and used to transform E. coli. tions showed single band hybridization signals corre After identification of the clone, the 10.7-kb p sponding to 11.6, 2.0, approximately 22, 2.8 and 2.5 O NIII(lppp-5) Ecó7-RT plasmid DNA was isolated. The kilobase pairs (kb), respectively. The upper band ap 196-bp synthetic msDNA fragment was then inserted pearing in the EcoRI digestion is due to incomplete into the vector by digesting with Xbal, treating the digestion of the chromosomal DNA. Analysis of total vector ends with bacterial alkaline phosphatase and chromosomal DNA prepared form E. coli C1-1 by aga ligating the fragment into the site. The construction rose gel electrophoresis revealed that the strain contains 15 scheme is shown in FIG. 4. E. coli CL-83 was trans two plasmids of different size. However, neither plas formed with the plNIII(lppp-5) ms100-RT plasmid and mid hybridized with the 32P-labeled probe, indicating the production of msDNA determined as in Example 1. the fragments detected in FIG. 12A are derived from The results indicated that msDNA was produced. This chromosomal DNA. Furthermore, there is only one artificial msDNA was designated ms100 and is illus location for themsDNA-coding region on the chromo 20 trated in FIG. 5. some, since various restriction enzyme digestions gave A second synthetic msDNA, ms101, was expressed only one band of varying sizes. from the vector puCK19, a derivative of puC19 (13). The 11.6-kb EcoRI fragment and the 2.8-kb Pst pUC19 DNA was digested with Dra and the 2-kb fragment were each cloned into puC9 (9) and E. coli fragment isolated. The isolated fragment was ligated to CL83 (a recA transductant of strain.JM83), anmsDNA 25 free K-12 strain (lane 1, FIG. 12B) was transformed a 1.3-kb Hinf fragment from Tn5 encoding the kana with the plasmids. Cells transformed with the 11.6-kb mycin resistance gene. The resultant 3.3-kb plasmid, EcoRI clone (pC1-1E) were found to produce msDNA pUCK19, was digested with Xbal and the 196-bp syn (lane 2, FIG. 12B, whereas cells transformed with the thetic msDNA described above was inserted. The 2.8-kb Pst clone (pC1-1P) failed to produce any detect 30 pUCKns 100 construct was digested with XhoI and able msDNA (lane 3, FIG. 12B). A map of the 11.6-kb SacII which results in the excision of a 61-bp fragment fragment is shown in FIG. 1. DNA blot analysis of the from within thems100 region. A synthetic 45-mer dou fragment revealed that a 1.8-kb Psti-HindIII fragment ble-stranded oligonucleotide (shown in FIG. 3B as ms hybridized with the msDNA probe. When the DNA C12) was ligated into the vector yielding puCKms101 sequence of this fragment was determined, a region in which themsr-msd region is under lac control. The identical to the sequence of themsDNA molecule was 35 construction scheme is shown in FIG. 6. RT was pro discovered. The DNA sequence corresponding to the vided by transforming E. coli containing pljCKms100 sequence of msDNA is indicated by the enclosed box on or pUCKms101 with plNIII (lppp-5) Ec67-RT. msDNA the lower strand in FIG. 7 and the orientation is from production was detected in the cells containing these right to left. The location of this sequence is also indi constructs. ms101 is shown in FIG. 7. cated by a small arrow in FIG. 1. A sequence identical to that of the RNA linked to msDNA (see FIG. 2) was EXAMPLE 5 found downstream of the msDNA-coding region in In Vivo cDNA Production in E. coli opposite orientation and overlapping with the region by In order to test a cDNA production in E. coli, a plas 7 bases. This sequence is indicated by the enclosed box 45 mid was constructed which was able to produce an on the upper strand in FIG. 13 and the branched G mRNA containing a sequence complementary to the 5' residue is circled. Again, as in all themsDNAs found in end of msDNA-Ec67. The construction of this plasmid myxobacteria, there is an inverted repeat comprised of (pUC19-Ec67-20), in which a 20-bp sequence was a 13-base sequence immediately upstream of the added at the unique Xbal site of puC19 is illustrated in branched G residue (residue 250 to 262; sequence a2 in 50 FIG. 9. The 20-bp sequence contains a 15-base sequence FIG. 13) and a sequence at the 3' end shown by an identical to the 3' end of msDNA-Ec67 (see FIG. 2A) arrow in FIG. 13 (residue 368 to 380; sequence al). As a so that the RNA transcript from the lac promoter of result of this inverted repeat, a putative longer primary pUC19 contains the 15-base sequence complementary RNA transcript beginning upstream of the RNA coding to the 3' end of msDNA-Ec67 at the position 80 bases region and extending through the msDNA coding re 55 downstream of the 5' end of the transcript. gion would be able to self-anneal and form a stable If E. coli JA221 harboring the Ec67 retron (pC1 secondary structure, which is proposed to serve as the IEP5b), is transformed with plasmid pUC19-Ec67-20, primer as well as the template for themsDNA synthesis the 3'-end region of msDNA-Ec67 may form a DNA (5). RNA hybrid not only with the 3' end of msdRNA (as EXAMPLE 4 shown in FIG. 2A) but also with the RNA transcript (lacz mRNA) from puC19-Ec67-20 as shown in FIG. Construction of Synthetic msDNA 9. Since the cells contain Ec67-RT, this enzyme may Two distinct synthetic msDNA molecules were con then start to synthesize cDNA by extending the 3' end structed. A 196-bp synthetic msDNA containing an of msDNA along the mRNA template. This would entire misr-msd region was synthesized from four dou 65 produce a single-stranded DNA of 152 bases which ble-stranded oligonucleotide units. The synthetic genes consists of the 67-base msDNA at the 5' end and the and their components are shown in FIG. 3. Eight sin 85-base cDNA (to the 5' end of the lac transcript; 80 gle-stranded oligonucleotides, forty-six to fifty-six bases bases from the lacz mRNA plus 5 bases from the linker) 5,405,775 11 12 at the 3' end. Identification of this cDNA was carried 10. Maxam, A. M. and Gilbert, W., Methods Enzymol. out by the polymerase chain reaction (PCR) (19) using 65,499 (1980). a 23-base oligonucleotide complementary to the 3' end 11. Ruskin, B. and Green, M., Science 229, 135 (1985). of the cDNA (P1; see FIG.9) and a 23-base oligonucle 12. Arenas, J. and Hurwitz, J., J. Biol. Chem. 262, 4274 otide identical to the 5' end of msDNA-Ec67 (P2) as (1987). primers. A DNA fraction containing cDNA was di 13. Yanisch-Perron, Y. et al., Gene 33, 103 (1985). gested with ribonuclease A and then used for the PCR. 14. Masui, Y. et al., "Experimental Manipulation of After the 25th cycle of the PCR, the DNA products Gene Expression' (ed. M. Inouye), pp. 15-32, Aca were fractionated on a 5% polyacrylamide gel and demic Press, New York (1983). detected by staining with ethidium bromide. A distinct 10 15. Hutchins, C. J. et al., Nucl. Acids Res. 14, 3627 band appeared at the position of approximately 150-base (1986). pairs. This band yielded two bands of approximately 80 16. Foster, A. C. and Symons, R. H., Cell 49,211 (1987). and 70-base pairs after Xbal, Pst and HindIIIdigestion. 17. Coleman, J. et al., Cell 37, 429 (1984). This is in good agreement with the predicted 152-bp 18. Southern, E., J. Mol. Biol. 98,503 (1975). cDNA depicted in FIG. 9, which is expected to yield 15 19. Saiki, R. K. et al., Science 230, 1350 (1985). two fragments of 80 and 72-bp upon Xbal digestion. We claim: The PCR did not yield any specific bands when puC19 1. A plasmid which comprises a retron which en without the 20-bp insert was used. codes a msDNA molecule selected from the group FIG. 13 shows nucleotide sequence of the region consisting of Ec67, Ec74, msl.00 and ms101, which has from the E. coli Cl-1 chromosome encompassing the 20 a DNA and an RNA portion, which retron comprises msDNA and the msdRNA coding regions and an ORF an msd gene, an imsr gene and an open reading frame downstream of the msdRNA region. The entire upper (ORF), the msd gene coding for the DNA strand of the strand beginning at the Ball site and ending just beyond msDNA molecule and the msr gene coding for the the ORF is shown. Only a part of the complementary RNA strand of the msDNA molecule, the msr gene lower strand is shown from base 241. to 420. The long 25 overlapping with and in opposite orientation with re boxed region of the upper strand (249-306) corresponds spect to the msd gene and the ORF coding for a reverse to the sequence of the branched RNA portion of the transcriptase (RT), said RT synthesizing cDNA from msDNA molecule. The boxed region of the lower an RNA transcript into themsDNA molecule. strand corresponds to the sequence of the DNA portion 30 2. The plasmid of claim 1 wherein themsDNA mole of msDNA. The starting site for DNA and RNA and cule is Ec67. the 5' to 3' orientation are indicated by large open ar 3. The plasmid of claim 1 wherein themsDNA mole rows. The msdRNA and msDNA regions overlap at cule is Ecla. their 3' ends by 7 bases. The circled G residues at posi 4. The plasmid of claim 1 wherein themsDNA mole tion 263 represents the branched rC of RNA linked to 35 cule is msl.00. the 5' end of the DNA strand in msDNA. Long solid 5. The plasmid of claim 1 wherein themsDNA mole arrows labeled a1 and a2 represent inverted repeat se cule is ms101. quences proposed to be important in the secondary 6. The plasmid of claim 1 wherein themsDNA mole structure of the primary RNA transcripts involved in cule contains a foreign DNA fragment positioned in an the synthesis of msDNA. Note that the nucleotide at 40 antisense orientation in the RNA or the DNA portion of position. 257 (U on the RNA transcript) and the nucleo the msDNA molecule. tide at position 373 (G on the RNA transcript) form an 7. The plasmid of claim 6 wherein themsDNA mole U-G pair in the stem between sequence a1 and a2. The cule is ms100. proposed promoter elements (-10 and -35 regions) 8. The plasmid of claim 6 wherein themsDNA mole for the primary RNA transcript are also boxed. The 45 cule is ms101. ORF coding for 586 amino acid residues begins with the 9. A prokaryote transformed with the plasmid of initiation codon ATG at base 418-420 to end with nu claim 1. cleotide 2175. Single letter designations are given for 10. The prokaryote of claim 9 which is a bacterium. amino acids. The YXDD amino acid sequence con 11. The bacterium of claim 10 which is E. coli served among known RT proteins is boxed. Numbers 50 12. An isolated retron which encodes a hybrid on the right hand column enumerate the nucleotide DNA/RNA msDNA molecule selected from the group bases and numbers with a enumerate amino acids. consisting of Ec67, Ec74, ms100 and ms101, which has Small vertical arrows labelled H and P locate the Hin a DNA and an RNA portion, which retron comprises dIII and Pst restriction cleavage sites, respectively. an msd gene, an imsr gene and an open reading frame The DNA sequence was determined by the chain termi 55 (ORF), the msd gene coding for the DNA strand of the nation method using synthetic oligonucleotides as prim msDNA molecule and the msr gene coding for the e.S. RNA strand of the msDNA molecule, the Insr gene overlapping with and in opposite orientation with re REFERENCES spect to the msd gene and the ORF coding for a reverse . Yee, T. et al., Cell 38,203 (1984). transcriptase (RT), said RT synthesizing cDNA from Dhundale, A. et al., Cell 51, 1105 (1987). an RNA transcript of the retron which encodes the Furuichi, T. et al., Cell 48, 47 (1987). msDNA molecule. Furuichi, T. et al., Cell 48, 55 (1987). 13. The retron of claim 12, wherein themsr and msd Dhundale, A. et al., J. Bacteriol. 164,914 (1985). genes are synthetic. Dhundale, A. et al., J. Biol. Chem. 263, 9055 (1988). 65 14. The retron of claim 13 which codes for thems100 Lira, D. and Maas, W., Cell 56, 891 (1989). msDNA molecule. Ternin, H. M., Nature 339,254 (1989). 15. The retron of claim 12 wherein the ORF is lo Lampson, B. C. et al., Science 243, 1033 (1989). cated upstream of the msd and downstream of the mSr. 5,405,775 13 14 16. The retron of claim 15, wherein the ORF is from antisense orientation in the RNA or the DNA portion of a source different than themsrand msd genes. themsDNA molecule. 17. The retron of claim 15 wherein the RT has 586 23. The retron of claim 22 wherein themsDNA mole amino acid residues. cule is ms100. 18. The retron of claim 12 which codes for the Ec67 24. The retron of claim 22 wherein themsDNA mole msDNA molecule. cule is msl.01. 19. The retron of claim 12, wherein a foreign nucleic 25. An isolated msDNA molecule selected from the acid sequence is positioned in the gene selected from the group consisting of Ec67, Ec74, ms100 and ms101. group consisting of themsr and msd genes. 26. The msDNA of claim 25 which is the synthetic 20. The retron of claim 12 which codes for the EcT4 10 ms100. msDNA molecule. 27. The msDNA of claim 25 which is Ec67. 21. The retron of claim 12 which codes for thems101 28. The msDNA of claim 25 which is EcT4. msDNA molecule. 29. The msDNA of claim 25 which is the synthetic 22. The retron of claim 12 wherein themsDNA mole ms 101. cule contains a foreign DNA fragment positioned in an 15 k k is

25

30

35

45

50

55

65 UNITED STATES PATENT AND TRADEMARK OFFICE CERTIFICATE OF CORRECTION

PATENT NO. : 5,405,775 Page 1 of 1 DATED : April 11, 1995 INVENTOR(S) : Inouye et al.

It is certified that error appears in the above-identified patent and that said Letters Patent is hereby corrected as shown below:

Column 1, Line 26, before “FIELD OF THE INVENTION” please insert the following paragraph: -- The U.S. Government has a paid-up license in this invention and the right in limited circumstances to require the patent owner to license otherS on reasonable terms as provided for by the terms of Grant No. GM44012 awarded by the NIH. --

Signed and Sealed this Ninth Day of September, 2003

JAMES E ROGAN Director of the United States Patent and Trademark Office