A genomic DNA segment from Petunia hybrida leads to increased transformation frequencies and simple integration patterns

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/19969043

A Genomic DNA Segment from Petunia Hybrida Leads to Increased Transformation Frequencies and Simple... Article in Proceedings of the National Academy of Sciences · December 1988 DOI: 10.1073/pnas.85.22.8568 · Source: PubMed

CITATIONS

READS

30

20

6 authors, including: Iris Heidmann Enza Zaden 29 PUBLICATIONS 1,321 CITATIONS SEE PROFILE

All content following this page was uploaded by Iris Heidmann on 12 January 2017.

The user has requested enhancement of the downloaded file.

Proc. Nati. Acad. Sci. USA

Vol. 85, pp. 8568-8572, November 1988 Genetics

A genomic DNA segment from Petunia hybrida leads to increased transformation frequencies and simple integration patterns PETER MEYER, SUSANNE KARTZKE, INGRID NIEDENHOF, IRIS HEIDMANN, KLAUS BUSSMANN, HEINZ SAEDLER

AND

Max-Planck-Institut ftr Zuchtungsforschung, D-5000 Cologne 30, Federal Republic of Germany

Communicated by P. Starlinger, August 4, 1988 (received for review April 20, 1988)

A 2-kilobase (kb) genomic fragment was ABSTRACT selected from Petunia hybida that increased transformation efficiencies by at least a factor of 20 after direct DNA transfer to petunia and tobacco protoplasts when supercoiled plasmid DNA was used. Because of this effect this fragment was named transformation booster sequence (TBS). Increased transformation frequencies were observed for plasmids that contained either the 2-kb fragment in dimeric or monomeric form or an internal 1.1-kb fragment of TBS. Analysis of ranformants revealed that preferentially one copy of foreign DNA is integrated. Thus, TBS improves the poor transformation frequencies of direct gene transfer using circular plasmids, while it conserves the simple integration pattern that is important for practical applications. Possible mechanisms of TBS action are discussed.

sequences has been reported for other transformations that also used circular plasmid DNA (10). Since synchronization procedures are not tolerated by all species and the technique is time consuming, alternatives for M-phase transformation were examined. Therefore, genomic DNA fragments were analyzed for a positive effect on transformation frequencies to select for autonomously replicating sequences, as has been performed in yeast (11) or for a transformation-enhancing fragment like the one described for Aspergillus nidulans (12). Here, we report on the isolation of a particular DNA fragment from Petunia hybrida, which was named transformation booster sequence (TBS), as it increased transformation frequencies without altering the simple integration pattern.

Although DNA transfer systems mediated by Agrobacterium tumefaciens are very efficient (for review, see ref. 1), direct DNA transfer into plant cells has gained some importance since its first application (2, 3). Several strategies to increase the relatively low transformation frequencies of direct DNA transfer have been tested. Considerable improvement was gained through transformation by electroporation (4) or using optimized Mg2e and PEG concentrations (5). However, both these methods include the use of linearized plasmid DNA or of an excess of carrier DNA. Although transformation frequencies are remarkably high, the DNA integration pattern is complex (4) and it remains to be shown whether, by using this technique, a sufficient number oftransformants can be generated that contain and express other genes of the introduced plasmid apart from the selectable marker. The present advantages of A. tumefaciens-mediated transformation are not only the high transformation frequencies obtained but, importantly, the simple integration pattern of the transferred DNA that guarantees practicability. Therefore, any techniques for direct DNA transfer should improve the quality of the transformation products as well as the frequency of transformation. One way to achieve a more simple integration pattern is the use of supercoiled plasmid DNA without the addition of carrier DNA, since it has been shown that in animal systems this leads to a greater proportion of undegraded and unrearranged integration events, while linear DNA often is subject to degradation (6). The use of circular DNA has the disadvantage of relatively low transformation frequencies (3). This could be overcome when protoplasts were transformed that had been synchronized into M phase (7). Relative transformation frequencies in the percent range were reached, while the simple integration pattern was conserved (8). The practical value of this technique has recently been shown as the coexpression of nonselectable sequences could be achieved (9). Such a coexpression reflecting the intact integration of neighboring

MATERIALS AND METHODS Protoplast Culture. Sterile shoot cultures ofP. hybrida var. Mitchell haploid (13) and Nicotiana tabacum Petit Havana var. SR1 (14) were grown on MS medium (15). Leaf material (2 g) was cut into small pieces and incubated for 16-18 hr in the dark in 50 ml of 11% mannitol containing 0.1% mazerozyme and 0.25% cellulase for petunia, and 0.2% mazerozyme and 0.5% cellulose for tobacco. The suspension was filtered through a 100-rum mesh and pelleted with an equal volume of seawater, supplied in 1986 by the Biologische Forschungsanstalt (Helgoland, F.R.G.) (petunia, 700 mosM; tobacco, 580 mosM). Pellets were resuspended in 10 ml of seawater, pelleted again, and floated in 0.6 M sucrose. After a final wash with seawater, protoplasts were dissolved in culture medium [petunia, 700 mosM V47 medium (16); tobacco, 500 mosM TM2 medium (17)]. Protoplasts were cultured under selection with kanamycin sulfate (100 mg/ liter) by the bead-type technique (18). Shoots were regenerated on MS medium containing kanamycin sulfate (100 mg/ liter) with indoleacetic acid (2 mg/liter) and benzylaminopurine (2 mg/liter) for petunia or zeatinriboside (1 mg/liter) for tobacco. Shoots were rooted on MS medium with kanamycin sulfate (100 mg/liter). Protoplast Transformation. Transformation was done in modification of a fusion technique (19) in a solution of 12.5% PEG 6000/50 mM Ca(NO3)2/225 mM mannitol/50 mM Hepes, pH 9, and a washing solution of 275 mM Ca(NO3)2 (pH 6). Supercoiled plasmid DNA (40 ,ug) and 106 protoplasts were used for each assay. Assay for Plasmid Stability and Replication. Mixtures of plasmids that were used for transformation contained a TBS plasmid and a reference vector that carried the lacZ gene. By transformation into Escherichia coli and plating on isopropyl

,B-D-thiogalactoside/5-bromo-4-chloro-3-indolyl j-D-galac-

toside medium, both plasmids could be distinguished as they gave rise to white vs. blue colonies. The proportion of the TBS plasmid within the mixture was determined before

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Abbreviation: TBS, transformation booster

8568

sequence.

Genetics: Meyer et al.

Proc. Natl. Acad. Sci. USA 85 (1988)

protoplast transformation and after reisolation of the plasmids from protoplasts at different times after transformation. Plasmids were isolated from transformed protoplasts as described by Werr and Lorz (20). No significant ratio variation was observed for mixtures of lacZ+ and lacZ- plasmids, provided they were of comparable size. Autonomous replication of plasmids was analyzed by E. coli transformation with plasmid DNA isolated from transformed protoplasts, which had been digested with Dpn I. Dpn I restriction requires methylation of the deoxyadenosyl residues at the recognition sequence GATC. Because plasmids were grown in dam' strains of E. coli, these residues were methylated (21). It has been shown that plasmids that replicate in mammalian cells lose this methylation and thus become resistant against Dpn I restriction (22). Plasmid Preparation. Plasmids were prepared as described

(23).

DNA Isolation and Transfer. Plant DNA was isolated as described by Bedbrook (24). DNA from transformed protoplasts was prepared according to Werr and Lorz (20). RadioEcoRV EcoRI EcoRV EcRY EcoRy

>. EcoRI

Kmr plants

8569

Table 1. Comparison of transformation frequencies of direct transfer into petunia (pl and p2) and tobacco protoplasts (tl-t7) TBS plasmids Control plasmids Size, Resistant Size, Resistant Plasmid kb calli Plasmid kb calli 140 5 pLGVllneo 6 pl pTBS4 10 p2 pTBS4 10 81 pLGVllneo 6 2 tl pTBS-4 10 78 pLGV1lneo 6 2 t2 pTBS-4 10 0 23 pLGV1lneo 6 t3 pTBS-4 10 33 pLGVllneo 6 2 t4 pTBS-2 8 35 pLGVllneo 6 1 t5 pTBS-1 8.8 41 7.2 2 pMP9 t6 pTBS-1 8.8 68 7.2 2 pMP9 t7 pTBS-1 8.8 26 0 pMP9 7.2 Numbers of resistant calli are given that derived from parallel transformation of split protoplast preparations with a TBS plasmid and the respective control vector. Plasmids carrying a TBS fragment show at least a 20-fold improved transformation frequency.

active probes for Southern blot analysis (25) were prepared according to Feinberg and Vogelstein (26, 27). Sequence Analysis. The sequence of fragment 4.1 was determined after subcloning into M13 and analysis by the dideoxy chain-termination method (28). Transformation of Bacteria. Bacteria were prepared for transformation and stored at -70°C as described by Hanahan

(29).

Recombinant DNA Techniques. All techniques not mentioned in detail were performed according to Maniatis et al.

(30). RESULTS Selection of a Genomic Sequence that Increases Transformation Frequencies. Randomly selected genomic fragments in the size range between 2 and 10 kilobases (kb) were isolated from N. tabacum var. SR1 and P. hybrida var. Mitchell haploid. Thirty different fragments were inserted plants

Kmr plants

Sal I

FIG. 1. Structure of the three TBS plasmids used for transformation experiments. The 2054-bp EcoRI TBS fragment was inserted as a dimer (pTBS4) or as a monomer (pTBS-2) into pLGVllneo. In pTBS-1, an internal EcoRV fragment located between positions 300 and 1412 (see Fig. 2) was cloned after attachment of Xba I linkers between two 255-bp Xba I/Hae II polylinker fragments derived from pUC19. The resulting Hae II fragment was blunt-ended by treatment with mung bean nuclease and was cloned into the EcoRI site of plasmid pMP9, which also had been treated with mung bean nuclease. Thus, the EcoRV fragment of TBS is flanked at each end by the recognition sites for Xba I, BamHI, Sma I, Kpn I, Sst I, and EcoRI, all of which can be used to cut out the intact TBS-containing sequence. The TBS fragment in pTBS-1 is in inverse orientation compared to the two other plasmids. Kmr, kanamycin resistance.

into the EcoRI site of the plant expression vector pLGVllneo (31) and transformation frequencies were compared in parallel protoplast transformation assays by using pLGVllneo and one of the plasmids that carried the genomic fragment. To ensure that differences in transformation frequencies were due to the inserted fragment, all other parameters that might influence transformation frequencies were carefully controlled. Parallel transformation assays were done with identical batches of protoplast preparations, which were split into equal parts. Equal amounts of plasmid DNA were used, resulting in a molar excess of the smaller control plasmid, and each plasmid preparation was analyzed by gel electrophoresis before transformation to make sure that the fast-migrating supercoiled conformation was the predominant component, usually only accompanied by a minor proportion of the retarded open circle form. One clone, which carried a 9.8-kb petunia insert, showed a roughly 4-fold increase in transformation frequencies when compared to the vector. A 2-kb EcoRI subclone of this petunia insert, TBS, was further analyzed by transformation of petunia and tobacco protoplasts. Three TBS constructs were made: (i) pTBS-4, which contains a dimer of the 2-kb fragment in tandem; (ii) pTBS-2, which contains the monomer; and (iii) pTBS-1, which carries an internal 1.1-kb EcoRV subfragment of the 2-kb sequence inserted in the opposite orientation (Fig. 1). In both tobacco and petunia TBS constructs showed on average at least a 20-fold increase in transformation frequencies compared to the vectors into which the TBS fragments were cloned (Table 1).

8570

Genetics: Meyer et al.

Proc. Natl. Acad. Sci. USA 85 (1988)

b

a

G T-A G-C

c

lo

TGCACTGCAAAWCCTTATATCTGGCCTTGTGArrCTCAAGTAGAGAAGCTTTCCACCAGTCTCTGATAGGATCCCACATCCT'CTAACITGCATCATC

20 1

AGrCTT7CCTAAcTGCTrGTCATATTCTACAGAGTGCAGCCMCTGI7('eTGTTCCrrGGrrACACTGGTc=AWCcTCCCAGACCcAACCCTGAT

30 1

ATCAGTT7CCAGTGCATACTTCCTCGrlrCAGTAGGAT1cCCCTTCTGATCTCAGCACCTCAATGCCTAAGAAGTArTTAGITTCCCAAATCrTTA

401

CTCTTGAAATGCTGATTCAGGzTTGCrrrG~rAGATCAAAAcATCTrrTGCcTGCTGTATTAACAGATCCATCCACATAAATCAGGATTATGACA

50 1

AGATCAGTCCCCTCTCTrlrIAAAMCAAGaGAATCATAAGrrTrrWATAAACAWCcTWATAAGGACAGTGGCTAAGCTsGTGATaCACTGCCT

60 1

TGATGCTTGCTmAAACCATAGAGAGACTrrCAACAWCCTGCACACCTrGGTCCCCCrrGGCTGTGAAAACCCTGAGGCAGAGACATATAAACrTCC

70 1

ATGAGGrCACCrrGTAGAAAAWATTTTMACATCCATCTGGAAAAGGAACCAWCCCTTGGAAWAWGATACAGATATGACCAGTrACAGTGCCA

8oi1

TTGGCCACTGGAGAAAAsTTTCATGGTAGTrcAGGCCTTcrrGrrGCAGTGTATCCCTTGGCCACTAGCCTGwCrTAMMCCTGTcAMcTTcACCAT

901l

TAGCrmGTArrAATTTrsTACAcCCArrrGGAcCCTATAGGTTGTTACCAGGGGGTAAAGGCAAsTCTCCAGTGTCTlCArrATCCTCAAGAGCCT

AA T-

C

T%

AA^

G A C * C-

C

A-T A

C T C

G-C

C-G G-C A-T C-G

T-A A

c A

ix

G

l oo1

GrATCTCAAGGGACATGGCCTCCATCCA7TTTTCArcTGAGCTGCrrCrTTGAAGGAMAGGTTCAGrATAAGTGGrAGMAAACACTCAAATAGCT

I 101

TGATAGTGAACAGGCCAAGrGATCATAGGAAACATAGTTAACTATAGGGTATGGG.ACATCCCCAGAGCCTrr=rrAGTGTACAAGTCCTTGAGCCAG

1201

ATGGGAGGACCTWcA-rrG

1 301

T-A G

rAGGTCTAcMcTWAAGrm-rACTGrAACTrC;AGAGGGATCATCTACA - ATGGTGCTCAAMT~CAGCCATTTOC

TAGCTrCAGACTCAWcTGACTCAACTGAn

GACCTGAAGG

_

ATCAGCTGM_ MTTGGZAGAATGAGTC

14 01

cCCAAAGTTGATATCCTCAT7GGCTCcTACArrsrCAGGAATAAGAACTGGAGTATCATCATCATCAGTATGAAAWCcTAGAAGGAAAATATCATTATA

1501

CACTAGCTWCAAGGCTGTATCTTCACEAGCTAAGTGAACCTGCTTGAGTGAACATATCAGGrrCATGGGAGATGGACTCrm-s AAAGGGA~cTGAAMcT

l bol

CTCTGAAAACTACATCCCTGCTCACTATGATCACCATT4ATCCAAG'rC A'rAC AAccTGTAAcCCCmGArCTCcAGAATAACCAATGAAGATGGTTT

I,0olCT7cGWTCTAGGTrWAAGTTGrCAcCCGGGCAGTGT7GCTGCAAAAGCAAGAC ACCC AAACAC'I'CTCAAATGATl Ho

A

IS*5 S

1

,

"')O I

A'FGAAC'AA'l C'C'('AA(.AGAG(.('C'AGC'AAC"l'C'Al'l'ACAAC'1'1'("AG'l'l'AA.-N(,~N '1

3'

6

TA-A G-C A-c G-C T-A

AGrACTTCATATAGACAT7CTT7ccr~AAGATTGGAGr~cAGC;A(-CTAI'IGA C AGCJ-'ATACA(;CA(,''C'I'I'GACACAG'TCTCCCCAAAACCTGGTAGGTA CACCACTCTGAAACTTAAGrGCCCTrTGCCArT7CAAAGAT(,-r( c IG-rGC lwl lC I("l(lc( A('AACAC ( A'1'1'1'1'(. ll caTGGTG(,cGAGGGACAWcTACrrrG

*T)

G-C

AGCTGGCATGTTCTGATAG

l1so1

-

A- * * G G G T C T A C A-T

G-C T

C T

T

cA AA

FIG. 2. (a) DNA sequence of the 2054-bp TBS fragment. The only three CG dinucleotides are boxed. At position 1374, there is the GTGGTATG consensus enhancer core segment, which is preceded by an imperfectly repeated motif, as indicated by underlying arrows. A repetition of another sequence motif GCAG(G) is boxed. (b) Theoretical secondary structure between nucleotides 1270 and 1385 of fragment 4.1 containing the transcription enhancer core sequence and the repetitive motif GCAG.

Characterization of TBS. The 2-kb EcoRI fragment was further characterized by Southern blot and sequence analysis. Hybridization of TBS to genomic DNA of P. hybrida and N. tabacum, cut with EcoRI, showed only moderate repetitiveness. Apart from hybridization to a 2-kb band, which represents the cloned TBS fragment, only a weak background hybridization was detected, which could be prevented by using more stringent washing conditions and which probably reflects some imperfect homology of TBS within the petunia and tobacco genome. Sequence analysis showed that the TBS fragment is 2054 base pairs (bp) long (Fig. 2a). It contains no continuous open reading frame larger than 550 bp. While the CNG content of the 108 CNG triplets in the TBS fragment corresponds to a random statistical distribution, the number of CG dinucleotides is remarkably low, as only three CG pairs are found while -100 would have been expected if distribution is random (calculation based on 23% C content and 21.1% G content). The only possible secondary structure of significant size is located between positions 1270 and 1385 (Fig. 2b). At one end (position 1374), an octamer GTGGTATG is located that corresponds to the consensus core sequence (32) found in animal and viral transcription enhancers. Within the 12701385 region, several sequence duplications occur and the motif GCAG is reiterated five times (Fig. 2b). To test the influence of an enhancer core region on the transformation frequency, the vector pLGVllneo was compared to pFL1, a construct in which the nopaline synthase promoter was replaced by the viral 35S promoter, including its enhancer region, which also carries an octamer sequence, GTGGAATG, homologous to the viral enhancer consensus core sequence. Transformation frequencies using pFL1 were at least 10-fold higher compared to pLGVllneo, which suggests that the consensus core sequence can be at least one of the factors contributing to the TBS effect. Transient expression studies revealed, however, that TBS did not increase the strength of the nos promoter nor was there any

difference in expression of the NPTII gene in stable pLGVllneo and pTBS transformants (data not shown). TBS did not increase transformation frequency through an increase in the stability of transforming DNA in protoplast or by initiating autonomous replication (for details, see Materials and Methods). Unexpectedly, we found that the proportion of smaller plasmids drastically increased with time whether or not TBS was present. For example, when two plasmids of 4.5 and 10 kb were used for transformation, the proportion of the 4.5-kb plasmid increased from 34% before protoplast transformation to 74% or 87% 1 or 6 days after

transformation, respectively. Analysis of Transformed Plants. Transgenic plants, which were generated by transformation with TBS plasmids or control vectors, were analyzed by Southern blot hybridization (Figs. 3 and 4). To determine copy number and integrity of the plasmids, the genomic DNA of seven vector transformants was cut with EcoRI, which cleaves only once within the plasmid, and was hybridized to the 2.3-kb EcoRI/Sal I fragment (probe A), which contains the selectable NPTII gene, and to the 750-bp EcoRI/Pst I fragment (probe B), which contains the 5' part of the ampicillin-resistance gene (Fig. 3). Hybridization shows that transformants contain one or two copies of the plasmid DNA, which are truncated to different extents. All transformants hybridize to probe A, which represents the selectable marker, while hybridization to probe B is only detectable for five transformants. DNA was cut with Xba I, which does not cut within the transferred plasmids, to determine the number of independent integration sites and the size of the integrated DNA. The plasmid copies are found at single integration loci. The assembly of concatamers greater than dimers could be excluded, as all hybridizing Xba fragments were