Tobacco plants transformed with cdc25, a mitotic inducer gene from fission yeast

  • Published on

  • View

  • Download

Embed Size (px)


<ul><li><p>Plant Molecular Biology 23:445-451, 1993. 1993 Kluwer Academic Publishers. Printed in Belgium. 445 </p><p>Tobacco plants transformed with cdc25, a mitotic inducer gene from fission yeast </p><p>Mark H. Bell 1, Nigel G. Halford 1., John C. Ormrod 2 and Dennis Francis 3 1Department of Agricultural Sciences, University of Bristol, AFRC Institute of Arable Crops Research, Long Ashton Research Station, Bristol BS18 9AF, England (* author for correspondence); 2 ZENECA Agrochemicals, Jealott's Hill Research Station, Bracknell, Berkshire RG12 6EY, England," 3School of Pure and Applied Biology, University of Wales, PO Box 915, Cardiff, CF1 3TL, Wales </p><p>Received 2 March 1993; accepted in revised form 29 June 1993 </p><p>Key words: cdc25, cell cycle, mitotic inducer, plant development, transgenic tobacco </p><p>Abstract </p><p>We investigated the effects of expressing a cDNA of cdc25, a mitotic inducer gene of Sch&amp;osaccharomyces pombe, on the development of transgenic tobacco plants (Nicotiana tabacum cv. Samsun). Nine inde- pendent primary transformants were regenerated containing the cdc25 sequence under the control of a cauliflower mosaic virus 35S gene promoter. Eight of the nine plants showed altered leaf morphology, the lamina being lengthened and twisted and the interveinal regions being pocketed. One of these was sacrificed for analysis of the root meristem, where the cells were found to be significantly smaller than in the wild type. The other seven were grown on and showed precocious flowering, flowers being pro- duced earlier and in significantly greater numbers than in the wild type. They also developed abnormal flowers on short stalks developing in a position normally occupied by the most proximal axillary bud of otherwise normal flower pedicels. The presence or absence of these phenotypes in the primary transformants and in the T2 generation was associated with the presence or absence of detectable lev- els of cdc25 transcripts. </p><p>Introduction </p><p>The initiation and completion of mitosis in eu- caryotic cells is under strict regulatory control. A central role in this regulatory process is played by maturation-promoting factor (MPF) which is re- quired for G2 to M-phase transition [5, 19, 30]. MPF consists of two subunits, a 34 kDa protein serine/threonine kinase encoded by the gene cdc2, and a 56 kDa 'cyclin' protein, encoded by the gene cdcl3, which is required for activation of the kinase. Homologues ofcdc2 and cyclin genes have been found in a wide range of eucaryotes, includ- ing plants [8, 9, 12, 13, 14, 16]. </p><p>In fission yeast (Saccharomyces pombe), the products of a number of genes regulate p34 cd2 activity, including those of cdc25, mikl and wee1 [7, 20, 26, 33]. wee1 and mikl encode related cdc2-specific protein kinases which phosphory- late p34 cdc2 at a tyrosine residue at position 15 [20], thus preventing the binding of ATP and inactivating the enzyme [10]. cdc25 encodes a protein phosphatase which removes the phos- phate group on tyrosine 15, activating the enzyme [ 18]. The effect ofcdc25 is to some extent dosage- dependent and increases in cdc25 levels cause a decrease in cell size at mitosis [27]. </p><p>Homologues of cdc25 have been characterised </p></li><li><p>446 </p><p>in a wide range of eucaryotes, including Droso- phila [6], budding yeast (Saccharomyces cerevi- siae) [32] and man [34]. So far a plant homo- logue has not been identified, but the high degree of conservation of other parts of the regulatory mechanism of cell division suggests that such ho- mologues probably exist. </p><p>The aim of this work was to determine whether and how plant development could be affected by over-expression of cell cycle control genes, cdc25 was chosen because of the dosage-dependent na- ture of its effects in yeast while tobacco was cho- sen for the introduction of the gene because of its ease of transformation. A chimaeric construct was made of the cdc25 coding sequence and a cauliflower mosaic virus (CaMV) 35S gene pro- moter and nopaline synthase gene terminator and introduced into the tobacco genome by Agrobac- terium-mediated transformation of leaf discs. This had dramatic effects on leaf development and flowering patterns in the transgenic plants. </p><p>transformed using Agrobacterium essentially as described by Horsch et al. [ 15]. Transformed cells were selected on shoot-inducing medium containing 100/~g/ml kanamycin and 200 #g/ml carbenicillin. Shoots were selected twice on root- inducing medium containing 100/~g/ml kanamy- cin. Rooted plants were potted in compost and grown in a greenhouse as described by Marris et al. [24]. Plants were measured and nodes la- belled using the nomenclature of McDaniel [25]. </p><p>Southern blot analys&amp; </p><p>Confirmation of transformation was determined by Southern blot analysis [38]. Genomic DNA was isolated using the method described by Kreis et al. [17] and restricted with enzymes Eco RI and SalI. The DNA was electrophoresed, Southern-blotted and probed with the cdc25 cDNA sequence. </p><p>Materials and methods </p><p>Chimaeric constructs </p><p>A cdc25 cDNA clone, pCDC25-S9, was kindly provided by Professor Paul Nurse (Imperial Can- cer Research Fund, Oxford). The chimaeric con- struct used to transform tobacco was made by inserting the coding region of cdc25 between 800 bp of the CaMV 35S gene promoter [31 ] and a 200 bp terminator sequence from the nopaline synthase gene in pUC19 [39]. The chimaeric gene was then inserted into the binary vector pBIN 19 [ 3 ] and electroporated into Agrobacterium tume- faciens strain LBA4404 using the method de- scribed by Shen and Forde [35]. </p><p>Agrobacterium-mediated transformation </p><p>Plants of Nicotiana tabacum cv. Samsun were grown in plastic boxes containing the revised me- dium of Murashige and Skoog [28], with an 18 h photoperiod at 2 C. Leaf discs were taken and </p><p>Detection of cdc25 transcripts </p><p>The cdc25 message was amplified using the tech- nique described by Gurr et al. [ 11]. Total RNA was extracted from leaves and treated with DNAse. First-strand cDNA was synthesised from 1/~g of total RNA using 4 gg of oligo-dT primer and 400 units of moloney murine leu- kaemia virus (MMLV) reverse transcriptase. A 400 bp section of the cdc25 sequence was then specifically amplified using primers 5'-GCCC- TGCCTTACCGACTCC and 5'-GGCGCGT- AGCTTGTCCACC in a PCR reaction of 25 cycles at 94 C for 1 min, 55 C for 2 min and 72 C for 1 min in a Perkin Elmer Cetus thermal cycler. The PCR products were electrophoresed, Southern-blotted and probed with the cdc25 cDNA sequence. </p><p>Cell size measurements </p><p>Secondary root tips from one of the transformed plants and a wild-type plant were fixed in 3:1 (v/v) </p></li><li><p>ethanol/acetic acid and stored at 5 C for at least 24 h. The root tips were Feulgen-stained at 25 C (optimum hydrolysis time 30 rain) and the root cap removed under a dissecting microscope. The terminal 300 #m of stained meristem was dis- sected in a droplet of 45 ~ (v/v) acetic acid and the cells were tapped out to form a monolayer, without squashing [1]. The coverslip was re- moved by the dry ice method [4] and the slides were dehydrated, counter-stained with 1~o (w/v) Fast Green in ethanol and made permanent with DPX mounting medium. </p><p>Cell areas were measured from the images por- trayed on a TV monitor interfaced to a Vickers M85A scanning microdensitometer. Tracings of cell perimeters were computed into area measure- ments with a Grafpad graphics tablet [36]. Thirty cells per slide and four slides per treatment were measured. Measurements of nuclear DNA con- tents were also recorded to determine the cell cycle limits of the cell distributions, and in all cases ranged from 2C(G1) to 4C(G2). No values of &lt; 1.8C (indicative of nuclear breakdown) or &gt; 4.4C (indicative of polyploidy) were recorded. </p><p>Results </p><p>The coding region of the yeast mitotic inducer gene cdc25 was inserted between a CaMV 35S gene promoter and a nopaline synthase termina- tor to make the chimaeric gene CaMV-cdc25- nos. The construct was inserted into the binary vector pBin19 and stably integrated into the to- bacco genome by Agrobacterium-mediated trans- formation of leaf discs. Nine independent pri- mary transformants were regenerated and screened for the presence of the construct by Southern blot analysis. Genomic DNA was iso- lated and cut with the restriction enzymes, Eco RI and Sal I, which cut at either end of the chimaeric gene to generate a fragment of 2.2 kb. The DNA was electrophoresed, blotted and probed with the yeast cdc25 sequence. This failed to hybridise to DNA from untransformed tobacco (data not shown) but clearly hybridised to DNA from the nine transformants (Fig. 1). </p><p>447 </p><p>Fig. 1. Southern blot of genomic DNA from the nine inde- pendent primary transformants, probed with the fission yeast cdc25 sequence. </p><p>Of the nine transformants, one showed no change from the wild-type phenotype shown by controls which had been taken through tissue cul- ture in an identical fashion but which had not been transformed or subjected to kanamycin se- lection. However, eight showed dramatic changes in leaf morphology. One of these was sacrificed before it reached maturity to provide material for analysis of the root meristem (see below). The other seven were allowed to flower and showed significant changes in flowering patterns compared with the wild type. These pheno- types were inherited by the T2 generation and were associated with the presence of detectable levels of cdc25 transcripts in the T1 and T2 gen- erations. </p><p>Leaf morphology </p><p>A typical leaf from a transformed plant is shown in Fig. 2. The lamina has lengthened and twisted and is 'pocketed' in places. All the leaves showed this phenotype, although the number and position of the pockets in the lamina was not constant. As the leaves aged, the phenotype became more ac- centuated and at the same time the leaves became a paler green. </p><p>Flowering </p><p>Apart from the one plant which showed no change in phenotype from the wild type, the transformed plants all flowered between 54 and 68 days after potting out, compared to 80-90 days for the con- </p></li><li><p>448 </p><p>Fig. 2. Abaxial view of a typical leaf lamina from a trans- formed plant. </p><p>trois. This precocious flowering resulted from the plants flowering at an earlier stage of develop- ment, rather than from an increase in the rate of development, the mean number of leaves pro- duced before flowering being 10.4, compared with 25.8 in the controls. There was also an increase in the total number of flowers produced by the transformed plants, with a mean of 49.6 com- pared to 26.2 for the controls (SED 9.29, 8 d.f., P &lt; 0.05). After self-pollination these flowers pro- duced pods of a normal size and seeds with nor- mal germination rates. </p><p>The transformants were also characterised by the presence of up to 40 small abnormal flowers per plant (Fig. 3). These developed on short stalks branching off from otherwise normal flower pedicels in a position normally occupied by the most proximal axillary bud. The petals were miss- ing entirely in these flowers and the other organs did not develop properly. These 'petal-less flowers' were infertile but did not affect the de- velopment of the normal flower on the same pedicel. </p><p>Fig. 3. A pedicel from a transformed plant. The bud at the apex grew on to form a normal fertile flower. An abnormal flower can be seen forming at the end of a stalk growing from a position normally occupied by the most proximal axillary bud. This flower contained sepals but no petals and only vestigial internal organs. </p><p>Detection of cdc25 transcripts in the transformed plants </p><p>cdc25 transcripts could not be detected by clas- sical northern blot techniques, indicating that ex- pression levels were low. However, they could be detected after reverse transcription with MMLV reverse transcriptase followed by amplification by the polymerase chain reaction using primers spe- cific to the cdc25 sequence. The PCR products synthesised in this way were Southern-blotted and probed with the yeast cdc25 sequence. This method detected cdc25 transcripts in each of the seven transformants which showed the pheno- types detailed above (the plant which was sacri- ficed for cell size measurements was not tested) (Fig. 4), but not in the transformant which showed a normal wild-type phenotype (Lane 8 in Fig. 4). This is a clear demonstration of a relationship </p></li><li><p>449 </p><p>Fig. 4. Southern blot of PCR products obtained from ampli- fication of a 400 bp fragment of the cdc25 transcript in eight of the primary transformants and a control plant (C), probed with the fission yeast cdc25 sequence. </p><p>between the presence of detectable levels of cdc25 expression and the changes in leaf morphology and flowering patterns. </p><p>Analyses of T2 transgenic plants </p><p>One of the transgenic plants was selfed and the resulting seeds germinated on water agar to pro- duce a T2 generation and 25 T2 plants were grown to maturity. Of these, 23 were subsequently shown to be kanamycin-resistant whereas 2 had lost resistance. All of the 23 plants which were kanamycin-resistant displayed the same leaf and flowering characteristics as the T1 plants and all were shown to contain cdc25 transcripts. Con- versely, the two plants which had lost kanamycin resistance showed a wild-type phenotype and did not contain detectable levels of cdc25 transcripts. This clearly demonstrates that the phenotypes segregate together, and that they are associated with expression of the cdc25 gene. </p><p>Cell size in the root meristem </p><p>As stated above, one of the transgenic plants was taken for analysis of the root meristem before it reached maturity. The distributions of cell areas for secondary root tips in the transformant and a control plant are shown in Fig. 5. The data were analysed with the Kolmogrov-Smirnoff two- sample non-parametric test, which examines </p><p>&gt;, o - </p><p>"=1 o- </p><p>18 20 2.2 2.4 26 28 </p><p>Log area midpoint (#m 2 ) </p><p>16 18 20 22 24 26 28 </p><p>Log area midpo int ( 2 ) </p><p>Fig. 5. Distribution of cell areas in secondary root tips in a transformed plant (A) and a wild-type plant (B) (n = 120). </p><p>whether two independent samples are part of the same population [37]. The test indicated a sig- nificant difference between the distributions (P</p></li><li><p>450 </p><p>maclonal variation could be the cause. Further- more, we have demonstrated that the effects are associated with the presence of detectable levels of cdc25 transcripts both in the primary transfor- mants and in the T2 generation. We conclude that expression of the yeast cdc25 gene is the cause of the observed phenotypes. </p><p>cdc25 encodes a protein phosphatase which forms part of a complicated network of interact- ing proteins regulating the cell cycle in yeast. Its only known substrate is p34 cdc2. We have no di- rect evidence that it is affecting cell cycle control when constitutively expressed in the transgenic plants. However, one possible explanation of the reduction in cell size in the root meristem of one of the transgenic plants is that the cells entered mitosis prematurely. (Notably, fission yeast mu- tants which over-expr...</p></li></ul>


View more >