3.2 ETAPA 2: Aplicación de NS a materiales de cemento
3.2.7 Ensayos y técnicas de monitoreo de los especímenes de mortero
Total RNA was isolated from flower buds and fmit using the Trizol method. This was
followed by poly (A) mRNA purification using the mRNA MessageMaker Reagent
Assembly (Gibco BRL). cDNA was synthesized from the m RNA using S uperScript I I
reverse transcriptase system (Gibco B RL) and used as templates to amplify short DNA fragments o f 1 4 5 bp from MADS-box region. Degenerate primers DEG- l and DEG-2,
corresponding to conserved residues M GRGKV Il, LCDAEV in the MADS-box
respectively were used in PCR. Because several bands from the PCR were detected on agarose gel, the DNA in a band of the expected size was excised fro m the gel and purified using Highpure PCR purification kit (Roche). The DNA fragment was cloned
into the pGEM-T vector, transformed into Escherichia coli and selected on plates
containing ampici l lin. 60 transformed colonies were picked at random i nto culture and
grown overnight; plasmid DNA purified and sequenced .
MADS-box sequences representing indivi dual genes in tomato were i dentified from the sequencing of the 1 45 -bp fragments. Sequences were al igned using the P i l eup program (GCG software) and the alignment submitted to the BOXS HADE server to highlight similarities and d i fferences (Appendix E). Differences in nucleotide seq uence in the region outside of the primer binding s i tes were used to determine whether clones represent similar or different genes.
The anal yses of 52 short seq uences i denti fied 1 8 different clones repre senting M !\ DS box genes in tomato. Each of the 1 8 di fferent sequences was analysed for similarity to
known M A DS-box sequences Llsing the B LAST program (Atsehul et a l . , 1 997). These
anal yses identi fied two fragments, Ti\;[ J 0 (Tomato /vf;l DS-box J 0) and T/vfl 8 as
representing tomato MADS-box genes previoLlsly une haracterised .
The longer fragment of TMi 0 was subsequently ampl ified with overlapping gene specific primers, TM I O-P I and TM I O-P2 and a 3' anchor primer using the rapid amplificatio n of c DNA ends (3' Ri\CE) techn ique ( Ohara et a!., 1 98 9 ) . The sequence o f the 3 ' anchor primer was part of the Not! primer-adapter used for cDNA synthesis. TM I O- P l and T M 1 0-P2 were used i n primary and secondary reactions respectively, in combin atio n with the anchor primer. The ampl ification conditions were as fol l ows: initial d enaturation at 94°C for 2 minutes, 3 0 cycles of 94°C for 30 seconds, SO°C for 3 0 seconds, 7 2°C for 1 minute and a fi nal elongation step at 7 2°C for 5 m i nutes. A I : 1 00 fold dilution of the primary P CR product was used as template i n the secondary PCR. These reactions produced a 0 .9-kb DNA fragment o f TMi O.
4.2.2 TM1 0 characterisation
The l onger DNA fragment of TMi O was c loned into the pGEM-T vector uti lising the single deoxyadenosine CA) added to PCR fragments by the Taq DNA polymerase. The DNA fragment \vas sequenced in both directions by M 1 3 forward and reverse primers using the A B I prism sequencer (Waikato DNA Sequencing Facility , Hami lton, New Zeal and ) . The length of the TM1 0 cDNA fragment was 8 64 nucleotides starting from
the
A TG
tran s l ation start s ite to the polyadenyl ation region. TAn 0 has an open reading frame o f 603 bp encoding 20 1 amino acids and a 3' untransl ated region of 2 6 1 nucleotides ( F i g ure 4 . 1 ) . I t carries sequences correspond i ng t o the 4 regions (MADS box, I -reg i o n . K-box and C-tcrminal) typical of p l ant M A D S-box genes (Krizck and Meyerowi tz, 1 996) . The derived protei n has a computed m o l e c u l ar weight of 23 k Daand i soelectric p H o f 6 . 5 , which is comparable to those o f o ther M ADS-box proteins (Mandel et aI . , 1 99 8 ) .
TCAAA TGAAGAGGATAGAGAA T CCAGT T CATCGACAAG T CACT M G R G K V Q M K R I E N P V H R Q v T
F C K R R A G L L K K A K E L S V L e D
A E I G L F I F S A H G K L Y E L A T K GGAAG CATGCAAGGGCTGATTGAGAGG TACATCAAGTCAACCAAGGGAG T T GAGG T GGCT G S M Q G L I E R Y I K S T K G V E V A GAGGAAGCCpAAGATACACAACCTCTGGACCCAAAP.GAGGAGATCAP.CATG C T GAGGAAT E E A K D T Q P L D P K E E l N M L R N GAGATT GACGTACTCCAGAAAGGCTTAAGC TACATGTATGGGGGAGGCGCAGGAACAATG E I D V L Q K G L S Y M Y G G G A G T M ACACTAGATGAACT T CATT CACTTGAAAAGTACCTTGAAATT TGGATGTATCATATT CGT T L D E L H S L E K Y L E I W M Y H I R T CAGCAAAGATGGATATCATGTTTCAAGAGATCCAACT G T TGAAGAATAAGGFAGGGATA S A K M D I M F Q E I Q L L K N K E G I C TGGAAGCTGCAAACAAATATT TACAGGATAAGATAGATGAGCAATACACT G TGACTAAC L E A A N K Y L Q D K I D E Q Y T V T N ATGACCCAGAATT T GACTGACTTTCAATGCCCACTAACTGTACAAAATGAGATATT T CAG M T Q N L T D F Q C P L T V Q N E l F Q T T TTAACATATGCTCACTA TGTAAGTTA T TCTTGTTGTGAAGCATCTA TGTAATTTGGTA F *
AGGAGATGTAATAATGATGATTGAGTPATTTCACTTTGGAGAGATGAACATATAAGTATG T TA T TA TGTTCAA TTTAGGTAATATGTTTAGTGTGTGAGCCTTTTTAGTGTATCTTCTCT AGTA TGGTGCTACTTATTATATATGTCA TCTTA TAATTTCTGAGTCAACTTCTTGTTTTG
T T A T TCAAAAAAAAAAAAAAAAAA
Figure 4. 1 . Sequence map of TMI O cDNA and derived amino acid residues. The nucleotide sequence of the positive strand of TM 1 0 fragment. The 3 ' untranslated region has been italicised. The boxed sequences indicate degenerate primer regions. The * marks the translational stop signaL
4.2.3 TMIO shows homology to AGL1 2
F o r comparison with other MADS-box sequences, the ami no aci d composition o f T M 1 0 was calcul ated and compared with related MADS-box p roteins; it was found to be closest to that of Arahidopsis AGL 1 2 (Table 4 . 1 ; Roun s l ey et aI., 19 95 ). F urther, the amino acid s equence of TM I O aligned with other MADS-box protein sequences showed high similarity to AGL l 2, across the entire sequence (Figure 4.2). The ami no aci d sequence o f T M l 0 has an overall i dentity of 64% t o A GL 1 2 protein o f Arabidopsis thaliana. Within the conserved MADS-box (56 amino acids), TM I O showed 84%
i de ntity to AGL 1 2 and 73% to Tomato MADS-box 4 (TM4 ) . To further exp lore TM I O relationship with other MADS-box sequences, an unrooted phylogenetic tree was constructed using amino acid sequence from the conserve d M ADS-box, the i ntervening region and the K -box, avoiding the variable C-term inal region. TM 1 0 and 1 9 other MADS-box protein sequences, obtained from the publ i c database, were analysed using PI LEUP (GCG software) with Kimura's distance correction m ethod. Sequences were clustered using the neighbor-j oining analyses of Saitou and Nei ( 1 987) and p lotted using Treeview soft\vare ( Page, 1 996). Phylogenetic analyses assigned TM 1 0 to the o rphaned group of A rah idops is A G L l 2 ( F igure 4 . 3 ) .
F o r fllrther characterisation o f the TMI 0 sequence, t h e secondary structure o f its conceptual protein was predicted using the PSIPRED model (Jones, 1 999). The pred i cted structure revealed a pattern of (X-hel ices, f)-strands and coi l s typical of p lant MADS-box p roteins and most similar to the AGL 1 2 protein ( F igure 4.4). Overal l , the sequence analyses suggested TM I 0 is the tomato homologue o f AGL 1 2 .