PARALLEL SESSION C – GENOMICS AND PROTEOMICS · PARALLEL SESSION C – GENOMICS AND PROTEOMICS ... Laboratoire de Biologie Végétale et Microbiologie, ... (AFLP-TP) analisys. Roots

PARALL EL SESSI ON C – GENOM ICS AND PROT EOMI CS

Proceedings of the XLVIII Italian Society of Agricultural Genetics – SIFV-SIGA Joint MeetingLecce, Italy – 15/18 September, 2004ISBN 88-900622-5-8

Poster Abstract - C.01

TRANSPOSON MUTAGENESIS IN PETUNIA HYBRIDA EXPANSIN GENEFAMILY

S. ZENONI, S. DAL SANTO, S. SORDO, G.B. TORNIELLI, M. PEZZOTTI

Scientific and Technological Department, University of Verona, Strada Le Grazie 15, 37134 Verona,(Italy)

Petunia hybrida, expansin, wall extension, gene function

In plants, organ shape is determined by organised and regulated control of cell expansion together withcell division. Emerging data support the view that the direction and magnitude of the enlargement ofthe primary cell wall largely determine the expansion pattern and thereby the final shape and size ofthe cells.

Cell wall proteins are believed to play important roles in regulating cell wall extensibility which is akey parameter determining cell expansion. Among cell wall proteins studied to date, expansins areunique in the ability to induce immediate cell wall extension in vitro and cell expansion in vivo.Expansin genes have been identified in many plant species and make up a large superfamily that isdivided into two major families α- and β-expansins, on the basis of the sequence divergence andbiochemical activity. Studies of expansin gene expression indicate that the different expansin genes areexpressed in different organs, tissues and cell type, and they respond distinctively to such treatments ashormones, light, pollination.

The reverse genetics approach based on transposon mutagenesis was used to study the function ofPetunia hybrida expansin genes. Four expansin genes (PhEXP2, PhEXP3, PhEXP4, PhEXP5) and thecorresponding insertion mutants were isolated by the screening of large populations of W138 petuniaplants, characterised by high copy number and high transposition activity of dTph1 element.Expression analysis of these genes in different plant organs and phenotypical analysis of the expansininsertion mutants were performed.



GENE EXPRESSION ANALISYS DURING GRAPE LEAF DEVELOPMENTAND SENESCENCE BY HIGH DENSITY FILTERS AND RT-PCR METHODS

M. PINDO*,****, E. BLANZIERI**, M. BERTAMINI*, C. SEGALA*, P. FONTANA*, H.A.BECKER***, R. VELASCO*, C. MOSER*

*) Area Biologia Avanzata; Istituto Agrario S. Michele a/Adige, S. Michele a/Adige - Trento**) Dip. di Informatica e Telecomunicazioni, Università di Trento***) Max Plank Institut für Züchtungsforschung, Koeln; Germany****) Dip. Scienze e Biotecnologie Agrarie, Università di Modena e Reggio Emilia

grape, gene expression, ESTs, arrays, leaf development

Leaf development and senescence are biological processes under tight genetic control. It has beenobserved that during leaf expansion photosynthetic rates typically increase and vice-versa during leafsenescence they decline. A photosynthetically efficient leaf is important during ripening of grapeberries in order to achieve high sugar and secondary metabolites concentrations. Gene expressionanalysis on a genome-wide scale has opened new possibilities to answer complex biological questions.The microarray technology and the high-throughput sequencing are the main technical bases of such arevolution in the experimental scale.

In order to obtain a collection of grape expressed sequences we constructed several standard cDNAlibraries from different tissues. A subset of these cDNA clones (from leaf, bud, sprout andinflorescence tissues) were sequenced and amplified by PCR and then orderly arrayed on nylonmembranes for gene expression studies. The high-density nylon filters contain 4010 amplified cDNAsdouble spotted corresponding to about 2300 clusters (unigenes). Radioactively-labeled RNAs from'Pinot noir' leaves at 5 different development stages were used to hybridize the high-density filters toprofile gene-expression during development and senescence of grape leaves. Analysis of geneexpression data has been carried out by looking at genes involved in specific metabolic functions andby clustering methods. Through RT-PCR method preliminary results has been validated and confirmedof several genes action during leaf development and senescence.



PROFILING GENE EXPRESSION OF STILBENES AND FLAVONOIDSPATHWAYS OF GRAPEVINE

P. GATTO*,**, A. ZAMBONI*,**, J. MUTH***, D. PRUEFER***, M. S. GRANDO*, C. MOSER*,F. MATTIVI*, R. VELASCO*

*) Istituto Agrario San Michele all’Adige (ISMAA), 38010 San Michele all’Adige, Trento**) Dipartimento Scientifico e Tecnologico, Università degli Studi di Verona***) Fraunhofer Institut (IME) fuer Molekularbiologie und angewandte Oekologie, 52074 Aachen,Germany

grapevine, phenolic compounds, cell cultures, cDNA arrays, functional genomics

Stilbenes and flavonoids are phenolic compounds present throughout the plant kingdom, which playkey roles in a variety of developmental and biochemical processes as well as environmental responses.Phenolics from grapevine have been the subject of numerous studies, because of their implications inphytopathology, human health and winemaking process. The pathway leading to the production ofpolyphenols has been extensively studied but the knowledge of transcriptional regulation ofbiosynthesis is still very little. To provide new information about the biosynthetic pathway of grapephenolic compounds at the molecular level, we have been initiated a functional genome project basedon the macroarrays technology.

The project was started with the construction of three differential cDNA libraries obtained bysubtractive hybridization among cDNA pools of three different grape tissues (V. vinifera): leaf, root(cv. Pinot noir) and berry (a mix of four cultivars characterized by different resveratrol content: cv.Pinot noir, Teroldego, Moscato bianco, the rootstock Freiburg 933/60). Around 4000 amplified clonesof each library were spotted on nylon membranes to generate three tissue-specific arrays. These high-density filters are being used to compare transcript profiles in grape cultivars with high and lowphenolic compounds content for three different developmental stages (veraison, maturation and post-maturation). Differentially expressed genes will be then correlated with specific HPLC profiles ofselected cultivars. Preliminary analysis of gene expression data will be presented.

Within the functional genomics project an important and complementary role is played by theestablishment of in vitro grape cell culture for the production of phenolic compounds under differentelicitors. Up to now we obtained cell cultures from shoot of different grape species: V. vinifera (cv.Pinot noir, cv. Merzling), V. amurensis and Vitis riparia x Vitis berlandieri. HPLC profiles of treatedgrape cell cultures were carried out in order to estimate the induction of stilbenes biosynthesis. A highproduction of resveratrol - a hydroxylated stilben - was observed. In parallel, gene expression studiesof grape cell culture showing high stilbenes content will be performed using both the cDNA arraysapproach and other strategies like cDNA subtractive hybridization. The genomic analysis coupled withthe biochemical analysis will increase the knowledge of stilbenes and flavonoids pathway, which willgenerate new commercial opportunities for wine, food and pharmaceutical industry.



A FUNCTIONAL GENOMICS APPROACH APPLIED TO DROUGHTRELATED GENES IN DURUM WHEAT

P. GALEFFI*, A. LATINI*, M. SPERANDEI*, G. CAVICCHIONI*, E. PALMIERI*, C.CANTALE*, M. DETTORI***, W. PFEIFFER****, K. AMMAR****, A. PELLEGRINESCHI****,M. IANNETTA**

*) ENEA BIOTEC/GEN, CR Casaccia, Via Anguillarese 301, 00060 Roma, [email protected]**) ENEA BIOTEC/DES, CR Casaccia, Via Anguillarese 301, 00060 Roma, Italy***) CRAS, Sardinia, Italy****) CIMMYT, Mexico

abiotic stress, drought related genes, durum wheat, stress tolerance, molecular assisted breeding

Cold and drought have afflicted agriculture over the ages. Today, the importance of crop resistance towater stress, extremes of salinity, and harsh temperature is further increasing, in connection with bothexpanding cultived areas and increasing of extreme weather conditions due to global climate change.Many stress-inducible genes have been identified and insights into their functional roles in stresstolerance is gaining. This makes it feasible to improve crop stress tolerance by targeting stress-relatedgenes either for genetic manipulation or for assisted breeding. The expression of the DREB2 gene afterinduction of drought-stress in a controlled greenhouse has been analysed in several italian varieties likeCreso, Ciccio, Simeto, Gianni, Cannizzo, Colosseo and Yellow-green, a laboratory mutant fromCappelli. Further 17 varieties, some coming from “field environment”, are going to be analysed. Workis in progress with specific oligos inferred from data bank genes, RNA extracts and RT-PCR andmultiple-PCR analyses and real-time-PCR. Interesting preliminary results of PCR analyses have beenobtained from these time-course experiments of drought-stress.

A functional genomics approach has been used to select and characterize the different varieties. Thefinal goal is to individuate those genotypes better responding to drought-stress in view of obtainingnew varieties by molecular assisted breeding.



ISOLATION AND CARACTERIZATION OF GENES MODULATED BYNITRIC OXIDE IN MEDICAGO TRUNCATULA

M. DE STEFANO*, P. MORETTO*, A. FERRARINI*, D. PADRINI PERONI*, E. BAUDOUIN**,E. PUPPO**, M. DELLEDONNE*

*) Università degli Studi di Verona, Dipartimento Scientifico e Tecnologico, Strada Le Grazie 15,37134 Verona**) Laboratoire de Biologie Végétale et Microbiologie, Université de Nice-Sophia Antipolis, CNRS,FRE 2294, Parc Valrose, 06108 Nice

nitric oxide, differentially expressed genes, plant defense, symbiontic association

The physiological importance of nitric oxide (NO) in plant life is now widely appreciated. Recentstudies highlights its involvement during leaf expansion, root growth, senescence, iron homeostasisand abiotic stress. Moreover, NO plays a prominent role in plant defense during plant-pathogeninteraction. Although its role in symbiosis has not been unraveled, a few data suggest that NO couldplay a role in the establishement of symbiotic associations. The symbiotic interaction between legumesand rhizobia leads to the formation of roots nodules, which are the sites of atmospheric nitrogenfixation and trasformation into ammonium. The infection and the establishment of symbiosis arecontrolled through a complex signalling network that involve reactive oxygen species (ROS) and NO.These reactive molecules are also involved in gene regulation during the hypersensitive response (HR)to pathogens. Therefore, plant pathogenesis and symbiosis may be considered as variations of acommon theme.

To unreveal the mechanisms regulated by NO in pathogenesis and symbiosis, we started atrascriptional analysis of genes modulated by this molecule in the model plant Medicago truncatula.

Differentially expressed genes were identified by cDNA-amplification fragment length polymorphismtranscript profiling (AFLP-TP) analisys. Roots from 4 week-old aeroponic-grown plants were treatedwith the NO donors sodium nitroprussiate (SNP) and nitrosoglutatione (GSNO). Using 32 primercombinations we isolated 1800 cDNA fragments which present an expression pattern altered betweenthe control and at least one of the treatments. Differentially expressed cDNA fragments were elutedfrom gel, reamplified and cloned in pGEM-T vector. The validation of the expression profiles of asubset of fragments was assessed using microarray analysis.

Sequencing of the entire set of cDNA fragments modulated by NO is now in progress.



GENE FAMILIES AS A TOOL FOR UNDERSTANDING EVOLUTION: THECOMMON BEAN CASE

A. CARBONI, F. DEL BIANCO, P. RANALLI

Istituto Sperimentale per le Colture Industriali, Bologna (Italy)[email protected]

common bean, NBS LRR, Resistance Gene Homologues, evolution, orthology, paralogy

It is widely accepted that two major and diverged geographical gene pools exist in common bean(Phaseolus vulgaris L.), one Mesoamerican and one Andean and that these gene pools reflect multipleand separate domestication events. Genetic isolation of wild and cultivated common bean, betweenand within the Middle and South America, has resulted in morphological and molecular differentiationof this species: habitus, shapes and storage seed proteins (phaseolin), isozymes, nuclear, mitochondrialand chloroplast DNA polymorphisms, etc.

Meanwhile conserved genes have found their way into the mainstream of molecular systematics andevolution, gene families have been regarded with suspect for evolutionary inference and conflictingresults often accompanying phylogenetic analysis derived from their RNA, DNA or protein sequences;causes contributing to these conflicts relate to ambiguities in identifying the disjoint subtypes ofhomology: orthology and paralogy. Concepts of orthology and paralogy are become increasinglyimportant as whole-genome comparison allows their identification in complete genomes: functionalspecifity of genes is assumed to be conserved among orthologs and is different among paralogs.Understanding how and how much the evolutionary pressures due to biotic stresses are responsible ofthe mechanism of differentiation among genotypes and trying to associate the emerging of newspecifities and functions of paralogous genes with these evolutionary forces has become the goal ofour research.

We used an innovative approach, i.e. tracking the dynamics of a fast evolving gene family, the NBS-LRR, to investigate the possibility that Phaseolus vulgaris L. may be in the process of an incipientspeciation into a Mesoamerican and an Andean species. To assess the state of this undergoingallopatric process Resistance Gene Homologues (RGHs) were amplified with degenerate primers in 2geographical divergent genotypes (one Mesoamerican and one Andean) and subsequently sequenced.More than 300 sequences of RGHs, derived from the Nucleotide Binding Site (NBS) domain of thenon TIR-NBS-Leucine-Rich Repeat (LRR) resistance gene sub-family, were used to follow theevolution of this sub-family on a relatively short time scale (7.000 yrs ago ca. since its domesticationprocess began in the different areas).

To measure the sequence structure and divergence, the best-fit substitution model, as well as to followand to describe correctly this dynamics we used specific softwares and particular and dedicatedanalysis; four different methods were used to reconstruct the phylogeny of the divergent RGHs:Minimum Evolution (ME), Weighted Maximum Parsimony (MP), Maximum Likelihood (ML) andBayesian analysis. The rate of this process was estimated and preliminary results deriving from theexamination of singleton, paralog or ortholog nucleotide sequences showed the presence of specific

genotypes clades and that half of the sequence lineages have undergone observable paralogousduplications, confirming our initial hypothesis.



DYNAMICS OF 5S RDNA IN MEDICAGO TRUNCATULA: CHROMOSOMEDISTRIBUTION, SEQUENCE ANALYSIS AND MOLECULAR PHYLOGENY

E. FALISTOCCO, V. PASSERI, G. MARCONI

Department of Botany and Agroenvironmental Biotechnologies, University of Perugia, Borgo XXGiugno 74, 06121 Perugia (Italy)[email protected]

5S rDNA, Medicago truncatula, Medicago species, in situ hybridization, evolution

In eukaryotes, the 5S rRNA genes are arranged in tandem arrays: each repeated unit is composed ofconserved coding sequences and of a more variable intergenic spacer region separating the codingregion from the next repeat unit. The chromosome organization of 5S loci is an usefull way to give aninsight into the intraspecific variability and the genome plasticity. Moreover the sequence analysis ofthe intergenic spacer may provide enough information for a molecular phylogenetic search amongrelated species. In this study the ribosomal 5S gene of Medicago truncatula has been isolated, clonedand sequenced. The two selected clones contain inserts of different length: one includes the codingsequence and the intergenic spacer while the other the complete gene sequence, the flanking spacersand the terminal and initial part of the gene. The two clones showed a complete identity in theircoding regions and a high level of identity between the spacer of the short insert and one of the spacersof the longhest insert. Genome organization of 5S genes has been investigated in 20 naturalpopulations by employing fluorescent in situ hybridization by using the cloned sequences ashomologous probe. 5S rRNA genes were isolated and cloned also in the perennial species Medicagoarborea, Medicago falcata and Medicago sativa. Multiple nucleotide alignment revealed a high degreeof similarity between the spacers of M. arborea, M. falcata and one of the spacers of M. truncatula. Inthe 5S rDNA repeat of alfalfa the spacer region differs from that of the other species for an additionalsequence of 33 nucleotide at 5’ end. These results induce to consider the spacer regions an importantsource of information to explore the molecular evolution in genome of these species.



CHROMOSOMAL ORGANIZATION OF TANDEM REPEATED DNASEQUENCES IN PICEA ABIES

S. MINELLI*, F. PANARA*, F. GELLI*, A. ZUCCOLO**, I. JURMAN**, M. MORGANTE**,P.G. CIONINI *

*) Department of Cellular and Molecular Biology, Section of Cytology and Genetics, University ofPerugia, Via A. Pascoli, 06123 Perugia (Italy)[email protected]**) Department of Agricultural and Environmental Sciences, University of Udine, Via delle Scienze208, 33100 Udine (Italy)

Picea abies, tandem repeated DNA sequences, in situ hybridisation

In order to study the sequence composition and structure of the Norway spruce genome, differentgenomic libraries were made, sequenced and used for several different analyses. Four tandemlyrepeated sequences were selected from a randomly sheared total genomic DNA library (PAG004P22F,PAG003M02C, PAG004E03C, PAG006O16F) and used in FISH experiments together with an alreadycharacterised centromeric tandem repeat, PATR140. In situ hybridisation experiments were carried outin order to study the chromosomal organization of these highly repeated DNA sequences and tocharacterize the P. abies chromosome complement. Squashes made with colchicine-treated seminalroots were incubated, after DNA denaturation in 70% formamide, with the heat-denatured probeswhich were labelled with digoxigenin-11dUTP or biotin-16dUTP by polymerase chain reaction. Thedigoxigenin or biotin at the hybridisation sites was detected by using sheep antidigoxigenin-fluoresceinor streptavidin-Cy3, respectively.

PAG004P22F repeats found nucleotide sequence homology at the centromeric region of nine out oftwelve chromosome pairs and at the nucleolus organizing region of four pairs. PAG006O16F-relatedsequences were found at the centromeres of four pairs and the secondary constrictions of five out ofsix satellited pairs. Sequences related to PAG004E03C, PAG003M02C, probes showed centromericlocalization in three chromosome pairs each that coincided with the three known locations for thepatr140 probe. Each centromeric region could contain either different tandem repeats or only one kindof repeat. One chromosome pair did not show centromeric labelling after hybridisation with any probe.Two findings seem to be worth noting. In five out of six satellited chromosome pairs, the same tandemrepeats are found both at centromeres and secondary constrictions. Surprisingly, and unlike what isusually observed, no tandem repeat is localized at the chromosome ends. Most chromosome pairs ofthe complement showed distinctive patterns of labelling after hybridisation with the probes. Bycombining the differences in labelling pattern and the morphology of the chromosomes, all pairs canbe distinguished from each other.



ISSR ANALYSIS OF GENETIC POLYMORPHISM AMONG CULTIVATEDAND WILD NICOTIANA SPECIES

L. DEL PIANO, L. BARBATO, M. ABET, C. SORRENTINO, E. COZZOLINO, A. CUCINIELLO,M. SICIGNANO

Istituto Sperimentale per il Tabacco, Via P. Vitiello n.108, 84018 Scafati (SA), [email protected]

ISSR, Nicotiana tabacum, Nicotiana species genetic diversity

The Inter Simple Sequence Repeat (ISSR) is a PCR based technique in which DNA fragments locatedbetween adiacent, oppositely oriented microsatellites (Simple Sequence Repeats, SSR), are amplifiedby Polymerase Chain Reaction (PCR) using primers that are anchored at 5’ or 3’ end of a repeat regionand extend into the flanking region.

The aim of this paper was to value the potential of ISSR technique to determine the degree of intra-and inter-specific genetic variation in the genus Nicotiana.

Genomic DNA of twenty-four tobacco lines (Nicotiana tabacum L.) and thirty-six Nicotiana species,belonging to almost all the sections of the genus, was extracted and amplified utilizing 10 differentprimers (14-22 bp). PCR products were resolved on agarose gel, stained with ethidium bromide andvisualized by UV transilluminator. The electrophoretic banding patterns were recordered and analzedby utilizing an image analysis system. The number of markers generated per primer ranged from 15 to30, and the dimensions from 200 and 2500 bp.

The amount of genetic polymorphism present among N. tabacum lines was limited as evidenced by thehigh degree of similarity in the ISSR profiles of different tobacco examined. A greater amount ofpolymorphism was revealed among wild Nicotiana species, as generally different band patterns wereobserved.

These results are consistent with our previous findings on genetic variability in Nicotiana tabacum andNicotiana species by RAPD analysis confirming the high genetic similarity among cultivated tobaccovarieties.



NUCLEOTIDE DIVERSITY AND LINKAGE DISEQUILIBRIUM IN POPULUSNIGRA

G. ZAINA, M. MORGANTE

Dipartimento di Scienze Agrarie e Ambientali, University of Udine, Via delle Scienze 208, 33100Udine, Italy

Populus nigra, association mapping, single nucleotide polymorphism (SNP), linkage disequilibrium(LD)

Association analysis is promising to be a very effective mapping approach in outcrossing long-livedplants, such as forest trees, where the long generation times and the difficulties in obtainingsegregating populations make other mapping strategies difficult and time consuming. Thanks toimprovements in sequencing technology, the direct detection of nucleotide polymorphisms (SNPs) isbecoming the primary genetic marker system to identify the associations between the allelic forms of agene and phenotypes in natural populations. Linkage disequilibrium (LD) plays a central role inassociation analysis, determining the number and density of markers required and the experimentaldesign to choose.

Data on nucleotide polymorphisms and linkage disequilibrium analyses are not available in Populusnigra. Before embarking into an association study to find the genes involved in the determination ofthe bud set and bud burst time in this species, we surveyed sequence diversity at 31 loci from 12different genotypes of P. nigra, representing a total of 24 independent chromosomes sampled. Elevenloci represent gene segments obtained from candidate genes involved in flowering time determinationin Arabidopsis; the other 20 loci represent single-copy genomic regions obtained by sequencing theinsert ends of BAC clones from a poplar genomic library. The frequency and the nature ofpolymorphisms were examined in detail; structure of SNP haplotypes and linkage disequilibriumwithin and between loci were also analysed. A total of 108 single nucleotide changes were found over13177 bp sequenced (1 change per 122 bp). SNPs resulted more frequent in non-coding regions (1 per70 bp) than in coding sequences (1 per 195 bp). On the basis of SNPs distribution, the overallnucleotide diversity (p) is 0.0024. Insertions and deletions were also found, but they were quite rare (1per 1098 bp). Twenty-one loci out of 31 present more than one SNP; in those loci sequence diversity isorganised into a number of haplotypes ranging from 3 to 11 (on average 4 haplotypes per locus).

In order to evaluate linkage disequilibrium within the sequenced regions, LD was estimated using theR2 measure for SNP pairs within each end of the BAC inserts and each gene. LD was found to be low(R2= 0.25 at 5 kb) but significant compared to the levels observed between physically unlinked loci.No significant LD was found across distances of 100 kb. Linkage disequilibrium was also compared inthe two pools of genomic and coding sequences. LD in coding sequences is higher (R2= 0.3 at 200 bp)than in the other sequences (R2= 0.25 at 200 bp).

On the basis of our data, in P. nigra we found a low nucleotide variability, if compared to other foresttrees, and a low but significant LD extending over a few kb distances.



CHARACTERIZATION OF TY1/COPIA-LIKE RETROTRANSPOSONSEQUENCES IN SPECIES OF HELIANTHUS

T.GIORDANI*, L. NATALI*, V. FERRILLO*, M. VUKICH*, S. MINELLI**, P.G. CIONINI**, A.CAVALLINI*

*) Dipartimento di Biologia delle Piante Agrarie, Sezione di Genetica, Via Matteotti 1/B, 56124 [email protected]**) Dipartimento di Biologia Cellulare e Molecolare della Università, Sezione di Citologia e Genetica,Via A. Pascoli, 06123 Perugia, Italy

molecular markers, Helianthus annuus, sunflower, retrotransposons

Retrotransposons are mobile genetic elements, of putative retroviral origin, medium-highly repeated,interspersed within the genome and ubiquitous in eukaryotes, both in animals and in plants. Duringspecies evolution, retrotransposons have occupied many different sites in the genome, through theproduction of an intermediate RNA which is retrotranscribed to extrachromosomal DNA by someenzymes (autonomously produced by the retrotransposon) and then inserted in a new site. Such areplicative transposition mechanism has contributed to genome structure leading to a rapid increase ofretrotransposon copy number with important consequences on gene activity regulation.

Among retrotransposons, elements with long-terminal-repeats (LTRs) and without LTRs can bedistinguished. LTRs are direct repeats situated at the extremities of the retrotransposon. LTR-retrotransposons are distinguished into two major groups, Ty1/copia and Ty3/gypsy, according tosequence similarity and to gene order.

A repeated sequence, pHaS211, 752 bp, was isolated from a partial genomic library of H. annuus. Itrevealed high similarity to the RNAse-H gene of Ty1/copia retroelements. Slot blot hybridizationsrevealed that pHaS211-related sequences are medium repeated (3x104 copies per haploid genome) inthe sunflower. Chromosome walking techniques allowed to isolate a number of putative LTRs, ofpHaS211-related copia-like retroelements, indicating the occurrence of different subfamilies. RT-PCRperformed on RNAs isolated from plantlets of sunflower exposed to abiotic stresses (cold or drought)showed that pHaS211-related elements are transcribed.

We have also analysed the occurrence of pHaS211 in the genome of Asteraceae belonging to differenttribes: our sequence appears to belong to a specific copia-subfamily, since it hybridises only to theDNA of certain species and is highly redundant only in Helianthus. In this genus, pHaS211-relatedsequences are found in both annual and perennial species and their frequency shows an eight-foldvariation (9.9 x 102 to 8.1 x 103 copies). FISH analysis of pHaS211 probe to the metaphasechromosomes of Helianthus species showed scattered labelling over all chromosomes, indicating largedispersal of Ty1/copia-like sequences. However, Ty1/copia-like sequences are less represented atcentromeric chromosome regions and have preferential telomeric localisation. Southern blothybridisation patterns of pHaS211 sequences to DNA digests of Helianthus species was used to builda phylogenetic tree, splitting the genomes studied into annuals and perennials.



WHOLE GENOME ANALYSIS OF RECOMBINATIONAL EVENTSINVOLVING LTR RETROTRANSPOSONS IN RICE (ORYZA SATIVA L.)

A. ZUCCOLO, M. MORGANTE

Department of Agricultural and Enviromental Science, University of Udine, Via delle Scienze 208,33100 Udine (Italy)[email protected]

LTR retrotransposon, recombination, Oryza sativa, genome structure

344 PAC and BAC clones sequenced by the Rice Genome Program were considered to accuratelyidentify a curated set of retrotransposons in Oryza sativa L.. 307 complete elements were identifiedand classified as members of the Ty-3-gypsy family (225 elements) or Ty-1-copia (71 elements)family. Elements were further classified into subfamilies through similarity searches carried out onLTRs. The 307 complete elements were then used to analyse the distribution of partial and completeLTR retroelements in the rice genome using similarity searches on the almost complete Oryza sativa(var. Japonica) genome sequence constituted of 12 chromosome pseudomolecules representing about358 Mbp: 17.16 % of the genome sequence considered corresponds to LTR retrotransposons. LTR-retrotransposons seem to be evenly dispersed across the rice genome. 3810 complete elements wereidentified; gypsy-like elements are 3.7 times more abundant than the copia ones. Sequence divergencebetween the LTRs from each element of a sample of 702 LTR retrotransposons was used to estimatethe time of the insertion into the genome: most elements (more than 85%) seem to have inserted lessthan 5 Myrs ago whereas only a small portion (less than 0.2%) is more than 20 million years old. Thesequences of the most conserved retroelements domains (in this case reverse transcriptase andintegrase) were used to build phylogenetic trees. The results were then used in comparison to maize:many retroelements subfamilies seem to predate the divergence between the two species andorthologus subfamilies show different relative abundance in the copia but not in the gypsy group. Riceseems to be characterized by an assortment of retrotransposons and a lack of overrepresented families:this feature is in striking contrast with that identified in maize genome. Evolution of most familiesseems to be driven by purifying selection acting on the protein domains to retain active transposition.When the chromosome pseudomolecules were searched for solo-LTRs, 7476 occurrences were found.The ratio between solo-LTRs and complete retroelements (1.92) as well as the incidence of nestedorganization of the retroelements (28.4% of the total) suggest that the mechanisms aimed at thecounterbalancing of retrotransposition effects could be active in rice genome. The ratio greatly varies(0.1-23.2) according to the different subfamily considered. The analysis of its variation along thechromosomes shows that recombinational events involving retrotransposons are less frequent aroundthe centromeric regions. A significant positive correlation has been found between the ratio of soloLTRs to complete elements and the average insertion age of each family. In comparison toArabidopsis, homologous recombination based mechanisms seem to be playing a greater role inreducing genome size in rice than illegitimate recombination based ones. Nevertheless, legacies of thelatter kind of recombinational mechanism can still be identified in the rice genome in the form ofincomplete elements: the amount of rice genomic sequence that was not accounted for by eithercomplete elements or by solo-LTRs was 9.8 Mbp out of a total of 56 Mbp matching retrotransposonsequences. The number of incomplete retrotransposon elements was 612 compared to a total number

of 3247 of complete elements. The ratio of incomplete to complete elements was higher for copia(0.21) than gypsy (0.18) elements. Solo-LTRs were proportionally more abundant in comparison toentire elements in the gypsy family than in the copia one even though the two families of elementsshow similar distributions of insertion times (slightly shifted towards younger elements in the gypsyfamily) as well as along the chromosomes. A different factor from either time or physical locationmust therefore account for this significant difference.



GENERATING A POLLEN FUNCTIONAL MAP USING OAT-MAIZEADDITION LINES

I. SALOMONI*, M.E. PÈ*, M. SARI-GORLA*, R. OKAGAKI**, R. PHILLIPS**, L.GIANFRANCESCHI*

*) Department of Biomolecular Sciences and Biotechnology, University of Milano, Via Celoria 26,20133 Milano (Italy)[email protected]**) Department of Agronomy and Plant Genetics, University of Minnesota, 1991 Upr Buford Circle,St. Paul, MN, USA

EST, functional map, pollen, addition lines, Zea mays L.

Even though pollen is a rather simple two- or three-celled organism, the construction of a functionalmale gametophyte requires the coordinated activity of a large pool of genes. Although gene sequencesand expression data are rapidly being accumulated, constituting a valuable source of information thatshould be exploited to integrate genetic maps with functional information, only about 2900 maize ESTcontigs (TC) out of a total of approximately 29400 have been mapped so far onto existing linkagemaps. The recent development of oat-maize addition lines has supplied a shortcut to the creation ofmaize functional maps, avoiding the identification of sequence polymorphisms and the use of largesegregating populations. In the present paper we report the use of oat-maize addition lines for mapping1000 maize pollen EST contigs. The strategy used is based on PCR amplifications performed on DNAderived from oat-maize addition lines corresponding to the 10 maize chromosomes. The analyzed ESTcontigs, retrieved from the TIGR database, have been selected according to the presence of at least oneEST derived from a pollen cDNA library. Map position was unavailable for all ESTs but one. Usingthe same standardized PCR conditions for all primer pairs, it was possible to assign about 57% of theTCs to specific maize chromosomes. Important information concerning gene duplication andchromosomal distribution were obtained. Those results represent a good starting point in theconstruction of a pollen functional map and they will certainly be very useful for the identification ofcandidate genes underlying mutants affecting viability and functions of the male gametophyte.Furthermore, those results could contribute to the assembly of sequence data, produced in the maizesequencing project, by identifying overlapping BAC clones and resolving alignment conflicts anddoubts. Finally, it is currently under development the analysis of radiation hybrid lines obtained fromthe single oat-maize addition lines, to obtain finer chromosomal localization of the EST contigs.



ASSOCIATION STUDY FOR TIMING OF BUD SET IN PICEA ABIES

E. DE PAOLI, M. MORGANTE

Dipartimento di Scienze Agrarie e Ambientali, University of Udine, Via delle Scienze 208, 33100Udine, Italy

Picea abies, association mapping, single nucleotide polymorphism (SNP), linkage disequilibrium (LD)

Adaptation to the local climatic conditions is one of the key features of forest trees. The steep cline inmany quantitative traits as time of bud set, time of bud flush, frost tolerance as well as droughtresistance in these species provide evidence of both natural selection and large populationdifferentiation with high Qst values compared to the Fst values assessed for neutral markers. Time ofbud set as a hallmark of annual growth cessation is related to frost resistance in autumn in Norwayspruce, shows a strong clinal variation across latitude and a high level of hereditability. In order tounveil the genetic factors underlying this trait in Norway spruce, we started an association study basedon the selection of more than ten candidate genes reported to be involved in the determination offlowering time in Arabidopsis, the list including genes coding for phytochromes, circadian clockfactors, vernalization–related factors and central integrators. New data on nucleotide variation at theputative orthologs of the selected candidate loci as well as at ten control neutral loci is being collected.This data set will provide a baseline on the distribution of single nucleotide polymorphisms (SNPs)and the extent of short range linkage disequilibrium among diverse Norway spruce Europeanpopulations distributed over the latitudinal cline observed for bud set. The comparison of nucleotidediversity data with phenotypic data will be used to detect significant associations and the role of theloci analyzed and estimate the size of effect of various alleles at those loci.

As a preliminary analysis for SNP discovery, three phytochrome genes (O, N and P) branching offfrom the same lineages resulting into present A- and B-type genes in angiosperms have been partiallyresequenced in haploid seed material from two populations selected at the ends of the cline (one fromNorth Italy and one from Sweden). The overall level of nucleotide diversity among the regionsanalyzed (accounting for more than 3200 bp sequenced in several individuals) resulted to be lower (p =0,0012; 5 segregating sites every 1000 bp) than the average value assessed in a previous sourvey on 26random EST-derived genomic loci (p = 0,0043; 13.31 segregating sites every 1000 bp). In order toevaluate linkage disequilibrium within the regions sequenced, R2 measures were calculated. Despitethe complete lack of SNP association in Phytochrome N and P genes, high LD was detected in aregulatory region of Phytochrome O gene with a R2 value still significant (R2 =0.5, P<0.05 Fischer’sexact test) over 900 bp compared to the low level of LD measured in the above mentioned study (R2

half length: 200 bp). To date no significant divergence of Tajima’s D values from neutrality wasfound, in agreement with the hypothesis of mutation-drift equilibrium.



CLASS A KUNITZ-TYPE PROTEINASE INHIBITOR GENEPOLYMORPHISM IN SOLANUM GENUS

A.S. SPERANSKY*, A.A. KRINITZINA*, P. FASANO**, P. POLTRONIERI**, T.A. VALUEVA*,A.B. SHEVELEV*, A. SANTINO**

*) INBI, A.N.Bakh Institute of Biochemistry, Russian Academy of Sciences, Moscow, Russia** Institute of Sciences of Food Productions, ISPA-CNR, section in Lecce, Via Monteroni km 7,73100 Lecce

Solanum tuberosum, proteinase inhibitors, resistance gene, evolutionary conservation

Kunitz-type proteinase inhibitors (PKPI) are widely spread in the Plant kingdom. Recent studies ofpotato cultivars have shown that these proteins are encoded by a highly polymorphic PKPI-genefamily and proposed to group these genes in three subfamilies: A, B and C, on the basis of theirstructure. In two potato cultivars Provita and Saturna, it was found by cytological chromosomalmapping the presence of three different loci in the potato genome (one for each of these three groups)so that a unique inhibitor- allele profile could be found in each cultivar (1). In the present work, wecollected a mini-bank of 50 independent genomic PCR clones from cv. Istrinskii, a Russian cultivar,encompassing PKPI group A loci. The obtained sequences were recognized as amplified copies of fiveindependent genes. Four of them showed 100% identity with published PKPI sequences from differentCultivars; the fifth sequence was totally similar to a PKPI gene from Solanum nigrum: this finding isnoteworthy since S. nigrum is not an ancestor species of cultivated varieties of potatoes. Together withthe genes from Solanum tuberosum cv. Istrinskii, we cloned three PKPI-genes from wild speciesSolanum andigenum and one sequence from Solanum stoloniferum. All four new genes exhibited anextremely high similarity with previously characterized PKPI-genes from cultured Solanum tuberosumvarieties. The obtained data suggest these conclusions: 1) there are at least two independent loci inpotato genome, since five different PKPI-A genes were found in the tetraploid potato genome. 2)-PKPI-A inhibitors are encoded by a highly conserved gene group, based on the finding that all allelecopies of the genes from cv. Istrinsky are 100% identical to PKPI from Solanum nigrum and to PKPIsfrom other different potato cultivars. Furthermore, PKPI-A genes from S. andigenum andS. stoloniferum exhibit a high similarity (96-99%) with corresponding sequences from other Solanumspecies. Sequence classification could show a common evolutionary fate of the multiple PKPI genesincluding the ones located in non-allelic genomic loci.

Supported by: NATO-Russia JSTC.RCLG.980102 project.1) Heibges A, Glaczinski H, Ballvora A, Salamini F. Structural diversity and organisation of threegene families for Kunitz-type enzyme inhibitors from potato tubers (Solanum tuberosum L.). Mol GenGenomics 2003, 269: 526-34.



ESTree DB AND E STuber DB : A FULLY AUTOMATE DPROCEDURE FOR EST SEQUENCE ANALYSI S AND DATABASEMANAGE MENT

B. LAZZARI*, A. CAPRERA***, L. MILANESI**, A. STELLA*, F. BIANCHI****, A.VECCHIETTI*, C. COSENTINO*****, A. VIOTTI*****, C. POZZI*

*) Parco Tecnologico Padano, Via Haussmann 7, [email protected]**) Istituto Tecnologie Biomediche, CNR, Via Fratelli Cervi 93, 20090 Segrate (MI) (Italy)***) CISI, Via Fratelli Cervi 93, 20090 Segrate (MI) (Italy)****) Università degli Studi di Milano, Di.Pro.Ve., Via Celoria 2, 20133 Milano (Italy)*****) Istituto Biologia e Biotecnologia Agraria, CNR, Via Bassini 15, 20133 Milano (Italy)

ESTree DB, ESTuber DB, Peach, Truffle, Bioinformatics

A fully automated pipeline has been developed to process EST sequences using public softwareintegrated by in-house developed Perl scripts and data deriving from peach and truffle EST sequenceanalyses have been collected into two MySQL databases, called ESTree DB and ESTuber DB. A php-based web interface has been prepared to surf and query the databases, that are currently available atthese URLs: http://www.itb.cnr.it/ESTree/ and http://www.itb.cnr.it/ESTuber/. A standardizedprocedure has been set up in order to process sequence data and generate a fully implemented databasefor each run of the pipeline. The databases contain about 7000 peach ESTs and 3000 truffle ESTs. Thelatter were produced from a cDNA library derived from Tuber borchii mycelium grown on solidmedium. Peach data have been produced in the frame of the activities of the National Consortium forPeach Genomics (ESTree). The program phred has been used for base calling, producing three files foreach sequence (electropherogram, text file and quality file) that are stored in the database. Sequencefiles and quality files have been processed with the program Lucy in order to identify and removevector contamination and low quality regions using default parameters. Vector-free high qualitysequences have been submitted to the program CAP3 to perform contig assembly. Stringencyparameters have been modified (-p 95, -d 60) to prevent over assembly and help identify potentialparalogs. All the EST sequences and all the contig consensus sequences have been submitted to theBLASTx program for annotation. BLASTx is run locally against the Genbank nr protein database.Blast output has been parsed with the MuSeqBox BLAST parser program and MuSeqBox output isstored in a MySQL database. A number of accessory Perl programs has been integrated into thepipeline to allow data flow among the main public programs and to recover further information fromintermediate elaboration steps and store it into the database. A Unigene data set has been defined ineach DB marking as Unigene all the singleton sequences and the longest sequence of each contig. Inthe DB sites multiple sequence and contig tables are presented that contain links to single sequence orcontig pages and to every sequence-related information. A text search utility is available and queriescan be performed either on the sequence and contig report tables or on the BLAST report fields,exploring every database field. BLAST E-value intervals can also be chosen by users to select high-homology annotated sequences. Selected output sequences can also be retrieved in EXCEL format.Local BLASTn, BLASTp, BLASTx and tBLASTx programs are available to perform BLAST and

batch BLAST searches on the ESTree and ESTuber databases. The resulting databases provides acollection of data for improving knowledge on peach and truffle functional genomics.



SET UP OF A MUTANT COLLECTION OF MEDICAGO TRUNCATULA FORFORWARD AND REVERSE GENETICS

L. MOLINARI*, L. SCARAMELLI**, G. BRUSCHI**, L. LANFALONI*, A. PORCEDDU*, C.SCOTTI**, M. ODOARDI**, E. PIANO**, S. ARCIONI*

*) Plant Genetics Institute, National Research Council (CNR), section of Perugia**) Istituto Sperimentale Colture Foraggere, Lodi

Medicago truncatula,,activation tagging, tilling, tannini condensati, saponine

We are using Medicago truncatula to gain knowledge on traits which are relevant for alfalfacultivation. Our approach toward the improvement of forage quality is the selection of mutants whithan increased accumulation of anti-bloating compound in leaves and with altered accumulation ofsaponins. Condensed tannins (CT), the most important anti-bloating factors of forages, are expressedin Medicago spp only in seeds. By an activation tagging approach we aim to select Medicagotruncatula mutants which accumulate CT in leaves. At the moment, two strategies which differ for theinsertional elements are being evaluated. The first employs a T-DNA construct with 4 copies of a 35Senhancer while the second is based on the tobacco Tnt1 and Tto retrotransposons. Leaves of trangenicplants will be analyzed by a CT specific colorometric test based on the DMACA dye . Plants positiveto the DMACA test will be further analyzed by HPLC. The DNA sequence flanking the insertionelement will be cloned and re-introduced in alfalfa for a functional analysis.

Saponins are antinutrional coumpouns which play an important role in plant –pathogen interaction.The screening of the mutant collection will be performed by an emolytic test. Plants showing analtered level of saponin accumulation will be used to identify the molecular determinants of the alteredaccumulation level. In addition morphological traits of plant architecture and male and female fertilityare considered.



A GENE CO-EXPRESSION NETWORK FOR GLOBAL ANALYSIS OF GENEEXPRESSION IN BARLEY

P. FACCIOLI*, V. TERZI*, C. MORCIA*, P. PROVERO**

*) Istituto Sperimentale per la Cerealicoltura, Via S.Protaso 302, 29017 Fiorenzuola d’Arda (Pc)**) Fondazione per le Biotecnologie, Villa Gualino, Viale Settimio Severo 63, 10133 Torino

functional genomics, gene expression, bioinformatics, co-expression network, cereals

The recent advent of high-throughput technology has shifted the focus of plant molecular biologistsfrom single genes to whole genomes and genomics methods examining mRNA and protein populationare now available that can offer insights into gene expression characterization.

Gene expression data sets can be conveniently visualized as gene networks where genes arerepresented by nodes and the relationships between them are represented by connections. In this studywe developed a multistep approach to gene expression analysis and functional prediction based on theconstruction of a barley gene co-expression network resulting from in-silico study of several publiclyavailable cDNA libraries. The network was divided into communities and the functionalcharacterization of them was analysed. Annotation confidence values were used as selection criteriafor choosing those communities most likely to be biologically relevant. As a result of this analyticalprocess we identified potential new members of existing functional categories and assigned probablecellular function to several poorly characterized genes.

Experimental validation has been carried out on several predictions to check the relationship betweennetwork and biological properties.



SNP DEVELOPMENT STRATEGIES IN OLIVE (OLEA EUROPAEA L.)

S. REALE*,**, S. DOVERI*, A. ANGIOLILLO**, F. PILLA**, P. DONINI*, D. LEE*

*) Molecular Research Group, NIAB, Huntingdon Road, Cambridge CB3 0LE (UK)[email protected]**) Department of S.A.V.A., University of Molise, Via de Sanctis, 86100 Campobasso (Italy)

Olea europaea, genes, SNP, SRAP, SCAR

The expansion of trade in olive oil creates increasing problems with authenticity and provenanceissues. With the vast numbers of trees grown world wide there remain many unresolved problems withnomenclature: synonymy, omonimy and identity. These issues can be resolved unambiguously withthe use of DNA markers (microsatellites or SSRs). However these types of markers, based on repeatlength variants are electrophoresis-based and are not suited for analyses using high throughput orquantitative formats such as LightCycler® or TaqMan® assays. For this reason we are developingsingle nucleotide polymorphism (SNP) and sequence characterized amplified region (SCAR) markersfor genes that can be used for variety identification or quantitation of alleles within genotype mixtures.The latter would help in determining composition of olive oil blends.

SNP identification is made by comparing the sequences of fragments obtained by PCR from tendiverse varieties. DNA fragments were generated from primers designed from olive gene sequencesavailable in the database and using the fingerprinting technique sequence related amplificationpolymorphism (SRAP), that targets expressed regions. Three cDNA sequences in the NCBI database,Cu/Zn superoxide dismutase, cycloartenol synthase and lupeol synthase show a high abundance ofSNPs. From a total of 3.5 kb of sequence, 22 SNPs (1/150 bp in introns and 1/180 bp in exons) and 2indels were identified.

This high level of polymorphisms, which manifests itself as a high level of heterozygosity observed inindividuals of this allogamous species, suggests that SNP screening can be performed with one or twovarieties and effective SNPs can be identified by ‘heterozygote screening’. The increase efficiencyoffered by this method will increase the numbers of SNPs identified for genotype analysis.

Markers developed in this study will be useful for traceability and provenance issues, as well asproviding loci for map construction, association studies and marker assisted selection breeding for theolive industry.



FUNCTIONAL GENOMICS IN PEACH (PRUNUS PERSICA): AN ESTAPPROACH

A.VECCHIETTI*, B. LAZZARI*, F. BIANCHI**, F. BARALE, F. SALAMINI*,**, C. POZZI*

*) Parco Tecnologico Padano – CERSA, Via Haussmann 7, 26900 Lodi (Italy)[email protected]**) Università degli Studi di Milano, Di.Pro.Ve., Via Celoria 2, 20133 Milano (Italy)

peach, genomics, EST, microarray, physical mapping

Italy is the second peach producer worldwide with approx. 2.7x106 q/year. Some countries (i.e. Chile,USA and France) are conducting an efficient policy of quality and quantity improvement by means ofaggressive breeding programs and genomics-based approaches. Italian scientists are taking part to theinternational effort to build a genomics system in Rosaceae and specifically in peach through anational consortium (ESTree: http://www.itb.cnr.it/ESTree). The current members of ESTree are:Fondazione Parco Tecnologico Padano, FPTP, Lodi; University of Padova, Department of Biology andDepartment of Environmental Agronomy and Crop Science (DAAPV); Istituto Sperimentale per lafrutticoltura, Roma; University of Milano, DiProVe; Istituto di Biologia e Biotecnologie Agrarie- Sez.Roma, CNR.In the frame of this consortium, the activity of our research group has been devoted to thepreparation of four cDNA libraries derived from peach mesocarps at different maturation stages (fromS1 to S4). A total of approx. 50.000 clones have been arrayed on macroarrays. Using a semi-automatedprotocol, 8000 plasmid DNAs were purified and sequenced. Approximately 7000 ESTs were analysedon our ESTree database (ESTreeDB). Current statistics indicate 2582 singletons and 779 contigs. Theaverage redundancy rate was estimated at 15% across the libraries. Gene ontology was conductedbased on BLAST results and on studies conducted on the KEGG database. Gene expression studieswere undertaken using 4800 oligonucleotides (70mers) that are microarrayed (in collaboration withCISI-Excellence Center, Milan) and screened with RNAs derived from fruits at contrasting stages ofmaturation. SNP-mapping of ESTs is also underway: 40 ESTs involved in secondary metabolism havebeen selected and are being mapped on the TxE and IxF genetic maps (in collaboration with ISF,Rome). ESTs physical mapping is also planned in collaboration with Clemson University, SC, USA.



MACROARRAY EXPRESSION ANALYSIS OF TRANSGENIC AND WILDTYPE DURUM WHEAT PLANTS

R.A. CIFARELLI, F. CARRIERO, O. D’ONOFRIO, G. LAURIA, A. PETROZZA, F. CELLINI

Metapontum Agrobios, S.S. Jonica 106, km 448.2, 75010 Metaponto (MT) Italy

macroarray, durum wheat, expression analysis

In order to investigate the eventual presence of differences in gene expression between geneticallymodified plants and their untransformed counterpart, we have utilized the DNA macroarraytechnology that allows a rapid and parallel analysis of many thousands of genes.

2200 individual sequenced cDNA clones of a durum wheat expression library were amplified and thePCR fragments were spotted on nylon filters by using the robotic workstation G3 (Genomic Solutions,PE).

The arrayed DNA fragments were then hybridised with 33P labelled cDNA prepared by retro-transcribing total RNA extracted from both durum wheat plants transgenic for the bialaphos resistance(BAR) gene and untransformed plants.

The analysis, performed with total rna prepared from three different developmental stages of grainrevealed that there are no qualitative differences in gene expression between transgenic and wild typedurum wheat plants.

Current investigations are being carried out to measure any quantitative differences in gene expressionbetween the analysed samples.



MASS SPECTROMETRIC APPROACHES TO THE CHARACTERIZATIONOF LOW-MOLECULAR WEIGHT GLUTENIN SUBUNITS

V. CUNSOLO*, S. FOTI*, V. MUCCILLI*, R. SALETTI*, D. LAFIANDRA**, S. MASCI**

*) Dipartimento di Scienze Chimiche, Università degli Studi di Catania, Catania, Italy**) Dipartimento di Agrobiologia e Agrochimica, Università della Tuscia, Viterbo, Italy

low molecular weight glutenin subunits, mass spectrometry, cysteine residues, glutenin polymers

The gluten macropolymer, which plays a major role in determining the viscoelastic properties of wheatflour, is constituted by high molecular weight (HMW-GS) and low molecular weight (LMW-GS)glutenin subunits linked together by intermolecular disulphide bridges. Based on their mobilities onSDS-PAGE, HMW-GS are divided into x- and y-type subunits, whereas three groups of LMW-GS canbe identified: B, C and D.

Glutenin subunits form both intra- and inter-chain disulfide bonds and, according to this property, theycan be classified into three groups: i) “chain branchers” subunits in which at least three cysteins areavailable for intermolecular bonds and therefore have a branching effect on the gluten macropolymer(likely the HMW-GS); ii) “chain extenders”, in which an even number of cysteine residues are present(i.e. B-LMW-GS); iii) “chain terminators”, which contain an extra cysteine residue (i.e. C and D-LMW-GS).

The LMW-GS present an extraordinary heterogeneity, because they are coded by a great number ofgenes, coding for different polypeptides. In this investigation, enriched B and C LMW-GS fractions,obtained by fractionated precipitation from hydro-alcoholic extracts of the bread wheat cultivarChinese Spring, have been characterized by RP-HPLC coupled with ESI and MALDI massspectrometry. Both procedures are very sensitive, allowing determination of more than 70 componentsstarting from 50 µg of extract. Moreover these techniques provide valuable insight into the structure ofthe components of the C and B LMW-GS fractions examined, including the determination of thenumber of cysteine residues present in some subunits, which is a major point in the analysis of theseproteins.



VALIDATION OF A RECOMBINANT ANTIBODIES REPERTOIRE TOSELECT INTRACELLULAR REAGENTS FOR FUNCTIONAL PROTEOMICAPPLICATIONS

M.E. VILLANI, M. DI CARLI, C. LICO, E. BENVENUTO, A. DESIDERIO

ENEA, BIOTEC-GEN, CR Casaccia, Via Anguillarese 301, 00060 Roma

intrabodies, proteomics, phage display, molecular repertoires

Antibodies can mediate phenotypic knock-out of physiological and pathological functions. One of themajor drawbacks in the use of antibodies as vehicles for in vivo interference is represented by thedifficulty to functionally express in the cell cytoplasm these proteins that are naturally secreted. In fact,in this reducing compartment disulphide bonds cannot be formed and folding enzymes are absent,often resulting in insoluble protein aggregates that are rapidly degraded. Since disulphide bonds appearto contribute significantly to antibody stability, folding of most antibodies under reducing conditionsdoes not normally lead to functional proteins. This is a serious restraint for applications that require aspecific interference to cytoplasm-resident proteins.

A repertoire of intrinsically stable antibodies has been constructed in our lab, using the scaffold of ascFv antibody (scFvF8) with remarkable molecular properties. This molecule, derived from a mousemonoclonal antibody, was functionally expressed in the cytoplasm of bacteria, yeasts and plant cellsand showed a long in vivo half-life. The thermodynamic characterisation of scFvF8 also demonstrateda very high intrinsic stability of this molecule. The derived repertoire ('F8 library') was generatedaccording to a modelling-assisted design, resulting in a collection of 5 x 107 independent clones.Recombinant antibodies with new specificities and biochemical characteristics similar to the cognateantibody (high stability and solubility in the cytoplasm) have been selected from the ‘F8 library’.

The aim of the present work was to assess if all single chain antibodies isolated from this library couldreally perform as “intrabodies”, being able to recognize the antigen when expressed in the cytoplasm.The data reported highlight the peculiarity of this molecular repertoire from which it is always possibleto isolate effective intrabodies, to be used in most functional proteomics applications.



SEQUENCE CHARACTERISATION AND COMPARISON OF PICEA ABIESAND GINGKO BILOBA GENOMES

A. ZUCCOLO, F. CATTONARO, I. JURMAN, M. VIZZI, M. MORGANTE

Department of Agricultural and Enviromental Science, University of Udine, Via delle Scienze 208,33100 Udine (Italy)[email protected]

Picea abies, Gingko biloba, repetitive DNA, amplification, retrotransposon

None of the Gymnosperm species has been well characterized at the genome level so far. These plants,however, share several interesting features under an ecological perspective as well as considering theirgenome structures: many of them have very large genomes whose physical size can be greater than1*1010 bp per haploid genome. We took a wide and comprehensive approach to study the structure ofGymnosperm genomes and considered two very distant members (they separated more than 200million years ago) of this group: Norway spruce (Picea abies (L.) Karst.) and Ginkgo (Ginkgo bilobaL.). Seven different genomic libraries were constructed from Norway spruce and used as source ofclones to be sequenced. Two of these libraries were made using randomly sheared (nebulised) genomicDNA to provide a random sample of the genome in one case and a random sample of itshypomethylated fraction in the other. Over 4000 sequences corresponding to more than 2 Mbp havebeen obtained from these last two libraries. A similar sequencing approach was undertaken in Ginkgobiloba using a randomly sheared library; 1025 sequences were obtained for a total of more 606 Kbp.A clustering analysis was carried out in order to assess the amount of repetitive sequences in bothspecies; 18% of ginkgo sequences (189/1025) were arranged in 67 contigs using the software cap3; inNorway spruce, we created 10 subsets each containing 1025 random chosen sequences out the 2494obtained from a random sheared genomic DNA library. The mean amount of repetitive sequences wasin this case 24.9% of the total (S.D.=1.13%) arranged in a mean number of 94.9 contigs A similarresampling approach was applied in maize using 3414 sequences: 43.5% of sequences were repetitive(S.D.=2.06%) and arranged in a mean number of 87 contigs. Sequence analyses on all libraries inNorway spruce reveal that: a) a relevant fraction of the genome is made of repetitive sequences; b)repetitive DNA is hypermethylated in Norway spruce; c) LTR retrotransposons are the majorcomponent of the repetitive fraction while other classes, although all present, are not abundant (LINEs,SINEs, tandem repeats); d) no single family of retrotransposons makes up a large fraction of thegenome, while many families seem to have amplified to high levels. Ginkgo, whose genome size isroughly half that of Norway spruce, contains a lower fraction of repetitive sequences; the LTRretroelement related sequences represent more than half (96 out of 189 sequences: 50.79%) of therepetitive fraction; this proportion is higher than that found in Norway spruce (412 out of 1050repetitive sequences identified in the total set of 2494: 39.23%). In the fraction of Ginkgo biloba DNAconsidered no LINEs, SINEs or DNA transposons were identified. Of the Ginkgo sequences analyzed,three, beside the conserved parts of LTR retroelements (120 sequences) or rDNAs (2 sequences), seemto be shared with Norway spruce: only one of them shows significant similarity over a stretch longerthan 100 bp. Four of the 120 Ginkgo retroelement related sequences exhibit a similarity greater than85% over more than 100 bp with the Norway spruce retroelement related ones. In consideration of the

very distant evolutionary relationship between the two species it is surprising to find this degree ofsequence conservation in putatively non functional repetitive DNA.

Neighbour joining trees were built using the conserved reverse transcriptase domain fromretroelements of Ginkgo and Norway spruce, in order to better investigate the phylogeneticrelationships between the two species. In both species no single family seems to make up a relevantfraction of the genome. These results provide a different model for genome expansion inGymnosperms from the one that seems to apply to cereals where single LTR families often make up10% or more of their large genomes. Results will be discussed in light of the evolutionary history ofGymnosperms



COMPGRASS: A SYNTENIC APPROACH TO CANDIDATE GENESIDENTIFICATION FOR BARLEY DEVELOPMENTAL TRAITS

L. ROSSINI**, L. NICOLOSO**, S. FRANZAGO**, A. VECCHIETTI*, D. WARE***, F.SALAMINI*,**, C. POZZI*

*) Parco Tecnologico Padano – CERSA, Via Haussmann 7, 26900 Lodi (Italy)[email protected]**) Università degli Studi di Milano, Di.Pro.Ve., Via Celoria 2, 20133 Milano (Italy)***) Cold Spring Harbor Laboratory, NY, USA

barley, synteny, mutants, transcription factor, candidate gene

Barley (Hordeum vulgare) is characterized by a vast collection of mutants, affecting as diverse traits asplant height, ear fertility and flower morphology. These characters are controlled by a set of genes forwhich monogenic mutants exist. Detailed genetic analyses in our labs have been conducted on theseand other mutants altering foliar organs and plant architecture, providing the starting point for theunderstanding of barley developmental processes (Castiglioni et al. 1998, Pozzi et al. 2000). Linkageanalyses have positioned 35 developmental mutant loci in a barley genetic map built with molecularmarkers (Castiglioni et al. 1998, Pozzi et al. 2003).Here we adopted an approach based on syntenybetween rice and barley to individuate candidate genes for the barley developmental loci previouslymapped by our group. Map positions reported in Pozzi et al., 2003 were used to select sequencedRFLPs linked to the mutant loci. BLAST searches conducted on the rice genome at www.gramene.orgallowed the identification of syntenic chromosomal regions in rice. This protocol was applied to 21mutant loci and for the majority of the hosting genomic regions synteny between barley and rice wasconfirmed. Annotation of the rice genomic sequence by multiple means (mostly athttp://ricegaas.dna.affrc.go.jp/) allowed the identification of putative candidate genes in rice. As a firststep towards the isolation of the barley ortholog, CLUSTALW was used to align homologoussequences derived from other grasses. When possible, degenerate primers were designed to PCR-amplify the sequences of interest from barley and SNP mapping of them is in progress, in order toprovide additional circumstantial evidence of the correlation between genes and mutant phenotypes.Specifically, five genes were further investigated: one belonging to the GRAS family, one AP2-like,one Polycomb-like, one myb family and a gene from the SBP family. The barley orthologues wereisolated and are being SNP-mapped on the high-density molecular map developed in our lab. Also,part of the analysis has been automated by means of two scripts we developed in collaboration with dr.D. Horner (UNIMI) and the COMPgrASS database and facility will be available online at the Grameneweb site.

CASTIGLIONI et al. (1998) Genetics 149:2039-2056POZZI et al. (2000) Genetics 154: 1335-1346POZZI et al. (2003) Heredity 90: 390-96



PHASEOLUS VULGARIS BAC LIBRARY: IDENTIFICATION ANDCHARACTERIZATION OF BAC CLONES CONTAINING THEINSECTICIDIAL LECTIN LOCUS

I. GALASSO*,**, L. LIOI**, C. LANAVE***, B. CAMPION****, R. BOLLINI*, F. SPARVOLI*

*) Istituto di Biologia e Biotecnologia Agraria, CNR, Milan, Italy**) Istituto di Genetica Vegetale, CNR, Bari, Italy***) Istituto Tecnologie Biomediche, CNR, Bari, Italy****) Istituto Sperimentale per l’Orticoltura, MiPAF, Monatanaso Lombardo, Lodi, Italy

APA locus, BAC library, common bean, gene family

Most of the species belonging to the genus Phaseolus contain in their seeds a family of homologouslectin and lectin-related proteins (Arcelin/Phytohemagglutinin/alpha-Amylase Inhibitor), which areencoded by a single locus, the APA locus. In common bean (P. vulgaris L.) this locus shows a highvariability. The three major components are present together only in some wild accessions, while mostgenotypes contain only phytohemagglutinin (PHA) and alpha-amylase inhibitor (alpha-AI) and somewild Mesoamerican accessions contain only PHA and Arcelin (ARC). This variability indicates acomplex evolution of the APA locus, which might be better understood by the isolation andcomparison of the entire locus from genotypes with different sets of APA components.

Today, the most commonly used system for constructing large insert libraries is the bacterial artificialchromosome (BAC) system.

The BAC library was construct using P. vulgaris accession G12949, CIAT, which contains the entiremultigene lectin family Arc/PHA/alpha-AI and shows high resistance against seed eating insects.High-molecular-weight DNA was partially digested with HindIII and cloned in the HindIII site ofpIndigoBAC5 vector. The library consists of 30,720 clones with an average insert of 135 kb, providinga coverage of 6 genome equivalents. The library is maintained and grown on eighty 384-wellmicrotiter plates.

To identify BAC clones containing APA genes, the entire library, was gridded onto two 22.5 x 22.5cm high-density filters double-spotted in a 4 x 4 array and probed with PHA, Arc4-II and alpha-AIclones isolated from the same genotype (1). After filter hybridization about 50 BAC clones wereidentified as positives. Among them, only 39 were confirmed as positives after a second screeningusing the two-lectin specific PCR primers P1 and P2 (2). In addition, to produce a contiguouschromosomal region covering the APA locus, the resulting 39 positive BAC clones were fingerprintedfor contig assembly. Our preliminary results confirm that the APA locus is very long, probably morethan 250Kbp, in fact to reconstruct the entire locus at least 4 overlapping clones appear necessary.However, to obtain a more detailed figure on the APA length and organization the sequencing of a firstBAC clone containing a large part of the APA locus is in progress.

(1) Lioi L, Sparvoli F, Galasso I, Lanave C, Bollini R (2003) Theor Appl Genet 107: 814–822(2) Mirkov TE, Wahlstrom JM, Hagiwara K, Finardi-Filho F, Kjemtrup S, Chrispeels MJ. (1994) PlantMol Biol 26: 1103-1113



PTI1-LIKE AND PTK-LIKE PROTEIN KINASES

A. CARPI*,**, S. GIRARDI*, A. DI MARTINO*, L. CECCATO*, M. VEDOVATO*, G. DIMAIRA*,**, V. LORUSSO*, M. SANAVIO*, V. ROSSI*, M. FRASSON**, A.M. BRUNATI**, A.DONELLA-DEANA**, E. BARBARO***, C. MOSER***, R. VELASCO***, F. LO SCHIAVO*,M. TERZI*, F. FILIPPINI*

*) Department of Biology, University of Padua, V.le G. Colombo 3, 35131 Padova (Italy)**) Department of Biological Chemistry, University of Padua, V.le G. Colombo 3, 35131 Padova(Italy)***) IASMA Agricultural Institute, Via E. Mach 1, 38010 San Michele all’Adige (Italy)

protein kinase, Arabidopsis thaliana, Vitis vinifera, tyrosine phosphorylation, two-hybrid

Comparative analysis of whole sequenced genomes from model eukaryotes shows that roughly 3% ofgenes encode protein kinases, while functional evidence on phosphoproteomes demonstrates that>30% of cellular proteins is phosphorylated by kinases. So far, most of the >900 putative proteinkinases detected in silico in the proteome of Arabidopsis thaliana have been tentatively attributed tothe Ser/Thr class; in addition, a few histidine kinases have been identified and recent bioinformaticevidence suggests the presence in Arabidopsis of a complement of putative protein tyrosine kinases(PTKs).

However, in all organisms, complete biochemical characterization concerns only a limited part of thekinase complement, as well as evidence on the involvement of kinases in cell pathways is oftenincomplete and/or controversial. For this reason, an approach to as yet uncharacterized kinases maycontribute to shed light on (or even to unravel) multiple pathways, eliciting further investigations.

We have started cloning of roughly 40 genes from Arabidopsis encoding putative kinases, a number ofwhich – based on bioinformatic markers – is expected to show either tyrosine or dual (Ser/Thr andTyr) specificity. More recently, we have started also a comparative analysis aimed at identifying,cloning and characterizing homologous kinase genes from Vitis.

So far, cDNAs encoding four such Arabidopsis kinases have been cloned in binary pYES vectors forthe expression in S. cerevisiae and subsequent purification as 6xHis-tagged chimerae; cloning offurther kinase cDNAs from Arabidopsis and Vitis is in progress.

Preliminary evidence is reported here on (i) the purification and biochemical characterization of aPTI1-like kinase, showing peculiar biochemical properties; (ii) two-hybrid screens and analysis ofkinase-interacting proteins, suggesting an involvement and a possible regulatory role inpathogen/stress response pathways and (iii) in silico evidence on receptor-tyrosine kinase (RTK)-likesequences, suggesting that RTKS might play a pivotal role in plant development, in agreement withour previous and unpublished (Floriduz M. et al., submitted) biochemical and cellular data.

References:

- Barizza E, Lo Schiavo F, Terzi M and Filippini F (1999) Evidence suggesting protein tyrosinephosphorylation in plants depends on the developmental conditions. FEBS Lett. 447: 191-194- Carpi A, Di Maira G, Vedovato M, Rossi V, Naccari T, Floriduz M, Terzi M and Filippini F (2002)Comparative proteome bioinformatics: Identification of a whole complement of putative proteintyrosine kinases in the model flowering plant Arabidopsis thaliana. Proteomics 2(11): 1494-1503- "The Arabidopsis Genome Initiative" (2000) Analysis of the genome sequence of the flowering plantArabidopsis thaliana. Nature 408: 796-815



FUNCTIONAL CHARACTERIZATION OF A MAIZE RPD3-TYPE HISTONEDEACETYLASE

S. LOCATELLI*, S. VAROTTO**, S. CANOVA**, M. MOTTO*, V. ROSSI*

*) Istituto Sperimentale per la Cerealicoltura, Via Stezzano 24, 24126 Bergamo, [email protected]**) Dipartimento di Agronomia Ambientale e Produzioni Vegetali, Università di Padova, Vialedell’Università 16, 35020 Legnaro (PD), Italy

histone deacetylase, histone acetylation, chromatin, gene activity, Zea mays

Post-translational modifications of histones, including acetylation, methylation, phosphorylation, andubiquitination, establish an epigenetic histone code that is recognized and interpreted by transcriptionalregulators and chromatin remodeling factors. Histone acetylation is one the best-characterized type ofhistone modifications. The enzymes responsible for maintaining the steady-state balance of histoneacetylation are the histone acetyltransferases (HATs) and histone deacetylases (HDACs). Mechanismsand factors controlling gene activity by affecting chromatin structure are largely conserved ineukaryotes, including plants. However, the sessile nature of plants, which makes them more sensitiveto environmental signals, and the relative plasticity of their cell fate suggest that specific features ofthe chromatin-mediated control of gene transcription exist in plants. Interestingly marked differences,reflected by specific patterns of histone acetylation and by novel classes of HDACs not identified inother experimental systems, have been reported (Loidl, 2004 Trends Plant Sci 9: 84-90). DifferentHDAC genes have been identified in plants and their classification into three distinct gene families hasbeen proposed by The Plant Chromatin Initiative (http://www.chromdb.org). The first family, namedHDA gene family, contains members related to the yeast sequences Rpd3 and Hda1.

In this study we functionally characterize a maize Rpd3-type HDAC by producing plants with over-expression and down-regulation (using antisense approach) of the ZmRpd3/101 gene. Molecularcharacterization of over-expressing (OE) and antisense (AS) plants reveals changes in the histoneacetylation and methylation pattern. Phenotypic analysis indicate various level of defects in plantdevelopment and architecture suggesting a role of this enzyme in regulating different metabolicpathway. Particularly, delaying in flowering and altered development of male and femaleinflorescences were observed in both OE and AS plants. Cytological observations on mutant plantsand in situ hybridization experiments with markers for meristems and inflorescences are in progress.Results obtained using chromatin immunoprecipitation (ChIP) assay indicate that ZmRpd3/101 doesnot affect histone acetylation of repetitive regions of the maize genome. The hybridization of DNAmicroarrays with probes obtained from OE and AS seedlings and endosperms is in progress in ourlaboratory to identify the ZmRpd3/101 target genes.



ALFALFA MOB1-LIKE PROTEINS ARE INVOLVED IN CELLPROLIFERATION AND LOCALIZE TO THE DIVISION PLANE DURINGCYTOKINESIS

S. CITTERIO*, S. VAROTTO**, E. ALBERTINI***, M. LUCCHIN**, P. PARRINI**, G.BARCACCIA**

*) Dipartimento di Scienze dell’Ambiente e del Territorio, Università di Milano – Bicocca, Piazzadella Scienza 1, 20126 Milano, Italy**) Dipartimento di Agronomia Ambientale e Produzioni Vegetali, Università di Padova, Agripolis,Via dell’Università 16, 35020 Legnaro, Padova, Italy***) Dipartimento di Biologia Vegetale e Biotecnologie Agroambientali, Università di Perugia, BorgoXX Giugno 74, 06121 Perugia, Italy

Medicago sativa, Mps-one-binder (Mob), cell cycle control, cytokinesis

Progression through the cell cycle is central to the proliferation of cells and fundamental to the growthand development of all organisms, including higher plants. The spatial control of cell division islargely dependent on plant-specific cytoskeletal structures, such as the pre-prophase band, whichmarks the division plane before mitosis and the phragmoplast, which is required to lay down the newcell wall during cytokinesis. Mob1 proteins, a novel group of eukaryotic proteins, seem to play animportant role in the cell cycle control, being involved in chromosome separation and completion ofcytokinesis. An EST with significant similarity to the yeast Mob1 genes was isolated by mRNAprofiling in a reproductive mutant of alfalfa and its full-length was obtained by RACE. Sequence-tagged site amplification and Southern blot hybridization revealed a multiple gene family in alfalfawith at least three separate members. Sequencing analysis of genomic DNA and cDNA clones enabledto find out two introns in all Mob1-like members. The organization of the genomic clone withinformation on CDS and indication of position and composition of regulating motifs and splicing sitesis available in the EMBL database as accession number AJ635582. For mRNA expression analysis, a400 bp specific cDNA fragment of the Mob1-like gene was transcribed in vitro to obtain DIG-UTPlabelled RNA sense and antisense probes. Moreover, to produce an antibody for the detection of theMob1-like protein, a peptide of 15 amino acids from the N-terminal region of the protein was selectedon the basis of its predicted structure. The quantitative comparison of both transcripts and proteinsdetected in dividing and non-dividing root and leaf cells, as assessed by flow cytometry analysis,proved that Mob1-like is a gene mainly involved in cell proliferation. Nevertheless, the presence ofMob1-like transcripts and proteins in quiescent cells suggests an added function for the membersbelonging to the Mob1 family. Analysis of gene expression was performed in root tips by means of insitu hybridization in order to localize Mob1-like transcripts in alfalfa apical meristems Thehybridization signal was detectable in the cells of the root apex where periclinal cell divisions mainlyoccur and the protoderm differentiates. As expected, the signal was not detectable in the root cap.Western blotting of alfalfa proteins isolated from plantlet roots and cotyledons revealed a doublet ofproteins of about 28 kDa and 47 kDa, providing that Mob may be a component of multi-domainproteins. Immulocalization in synchronized cells with the anti-Mob1 polyclonal antibody allowed todetermine that Mob1-like proteins are expressed in a cell cycle-dependent manner. Along with acortical localization late in interphase, the Mob1-like proteins were visualized during mitosis in the

cytoplasm around the daughter nuclei as grain clusters from which fibrillar structures radiate in alldirections, preferentially toward the cell midplane. At the end of mitosis, Mob1-like proteins localizedin the cell plate during cytokinesis by marking the progressive formation of the phragmoplast from theinside to the cell periphery. Grains likely correspond to sites in which microtubules are reorganizedduring cell cycle progression, the yeast SPBs and animal MTOCs.



TRANSCRIPTIONAL AND PROTEOMIC ANALYSIS UPON UPR(UNFOLDED PROTEIN RESPONSE) INDUCTION IN DEVELOPING LIMABEAN COTYLEDONS

E. PONZONI*, U. RESTUCCIA*, I. GALASSO*,**, R. BOLLINI*, F. SPARVOLI*

*) Istituto di Biologia e Biotecnologia Agraria, CNR, Milan, Italy**) Istituto di Genetica Vegetale, CNR, Bari, Italy

Because of their abundance, storage proteins represent a good model system to study secretory proteinsynthesis and accumulation in plant cells. We are currently investigating the role of different kind ofstresses (inhibition of glycosylation, reducing agents, heat treatment, calcium ionophores) on storageprotein folding in the endoplasmic reticulum (ER) of developing lima bean (Phaseolus lunatus)cotyledons. This subcellular compartment is extremely important in developing bean cotyledons,whose major function is the accumulation and compartmentalisation of secretory storage proteins.Correct folding of newly synthesised proteins in the lumen of the ER is a fundamental prerequisite fortheir transport to other cellular compartments. Misfolded proteins, if not refolded by residentchaperones, are destined to degradation or might accumulate as aggregates in the ER (1). In both casesthese events detrimentally affect the function, localisation and, in the case of storage proteins,eventually the amount of proteins that are accumulated in the seed. The way cells monitor thephysiological load placed on their ER and respond to perturbations in ER functions is known asunfolded protein response (UPR) (2). The UPR has been extensively studied and characterised inmammalian cells and in model organisms, such as yeast and the nematode Caenorabditis elegans (2),very few data are available for plant cells (3).

We are interested to identify the function of genes/proteins involved in the mechanisms that lima beancotyledonary cells activate when the UPR is triggered. The UPR is induced by treating developingcotyledons with tunicamycin, an inhibitor of protein N-glycosylation that causes accumulation ofmalfolded proteins in the ER (1). In order to identify genes which expression is induced by thistreatment, we used the suppression subtractive hybridisation (SSH) method to generate a subtractedcDNA library. About 30.000 recombinant clones have been obtained and about 1500 have been doublespotted onto a 3x3 array. Four replicas of each array were prepared and hybridised both withsubtracted and unsubtracted control and treated mRNA probes in order to identify also poorlyexpressed genes. The hybridisation results of the arrays are under analyses and will be presented.

To carry out the analysis of proteome variation after the treatment with tunicamycin, we are focusingon the components of the ER of treated and control cotyledons. The 2DPAGE analysis is beingperformed on different fractions of purified ER membranes as in (4).

(1) Sparvoli F., Faoro F, Daminati M.G., Ceriotti A and Bollini R. (2000) Plant J. 24:825-836.(2) Harding H.P., Calfomn M., Urano F., Novoa I. and Ron D. (2002) Ann. Rev. Cell. Dev. Biol.18:575-99.(3) Martinez I.M. and Chrispeels M.J. (2003) The Plant Cell 15:561-576.

(4) Maltman D.J ., Simon W.J., Wjeeler C.H., Dunn M.J., Wait R. and Slabas A.R. (2002)Electrophoresis 23 :626-639.



IN-VITRO EXPRESSION AND CHARACTERIZATION OF A RAPESEEDPROTEASE INHIBITOR ACTIVE AGAINST A MICROBIAL GLUTAMYLPROTEASE

M. VOLPICELLA*, M. SCIANCALEPORE*, F. DE LEO**, B. GATTULLI**, R. GALLERANI*,L.R. CECI**

*) Department of Biochemistry and Molecular Biology, University of Bari, Via Amendola 165/A,70126 Bari (Italy)**) Institute of Biomembranes and Bioenergetic, CNR, Trani Unit, Via Corato 17, 70059 Trani (Italy)

rapeseed, protease inhibitor, Streptomyces griseus, glutamyl protease

Protease inhibitors (PIs) in plants have been extensively investigated. Their physiological role can beeither the regulation of endogenous proteolysis and/or the control of exogenous proteases, typicallythose present in insect guts or in other pathogens. Under this aspect PIs can be regarded as one of thedefence strategies of plants against pest and pathogens. Plant PIs are usually classified according to themechanistic class of proteases they inhibit: serine-, cysteine-, aspartic- and metallo- proteases.Inhibitors of serine proteases are the most numerous and the best characterized. Among them, severaldifferent structural sub-families have been catalogued.

A novel family of plant inhibitors of serine proteases has been recently hypothesised in Cruciferae,after the description of two trypsin inhibitors: MTI-2 (Mustard Trypsin Inhibitor-2) and RTI-III(Rapeseed Trypsin Inhibitor-III), which could not be catalogued in any of the already known familiesof plant PIs. Currently the new PI family (MSI, Mustard Inhibitors family) also includes six potentialinhibitors from Arabidopsis, whose genes were detected in the course of whole genome-sequencingprograms.

In order to establish the content of genes coding for inhibitors of the MSI family in rapeseed, thescreening of a genomic library was carried out. Three genes were identified, one of which codes for theRTI-III inhibitor already purified from rapeseed seeds. Of the other two genes, one codes for aninhibitor with the same reactive site described for MTI-2 and RTI-III, the latter one (named RTI-2)contains an unusual Glu residue in the P1 position of its reactive site, which would make it an inhibitorof glutamyl-proteases (specific endopeptidases that hydrolyse peptide bonds on the carboxyl side ofGlu residues). This is a result of particular interest since up to now serine proteases with suchspecificity have been described only in bacteria, but not in plants or insects. We have therefore startedthe characterization of this novel protease inhibitor.

Recombinant RTI-2 was obtained by in-vitro expression from the yeast Pichia pastoris. The initialdifficulties to characterize its inhibitory properties, due to the lack of a proper substrate, wereovercome by the expression in Bacillus subtilis of a specific glutamyl protease from Streptomycesgriseus. Analysis of the biochemical interaction between RTI-2 and the S. griseus glutamyl proteasegave useful information about the activity of the inhibitor, opening the possibility to understand itsphysiological role.



FUNCTIONAL CHARACTERIZATION OF MBD PROTEINS INARABIDOPSIS

F. SCEBBA*, M. DE BASTIANI**, A. ANDREUCCI***, L. PITTO*, G. BERNACCHIA**

*) Istituto di Fisiologia Clinica, Laboratorio di Terapia Genica e Molecolare, Area della ricerca CNR,Via Moruzzi 1, 56100 [email protected]**) Dip. Biologia, UniFE, Via Borsari 46, 44100 [email protected]***) Dip. Sc. Botaniche, UniPI, Via L. Ghini 5, 56126 Pisa

epigenetic control, plant chromatin, Arabidopsis, transient transfection

Cytosine DNA methylation and posttranscriptional modifications of the aminotermini of core histonesprovide epigenetic codes for genome regulation. In vertebrates, a family of methyl-CpG bindingproteins (MBD proteins) mediates the biological effects of methylation. Its best-characterized familymember is MeCp2, a transcriptional repressor that recruits histone deacetylases and interacts withhistone methyltransferases. More recently in silico analysis in Arabidopsis and other plant species,showed the existence of a protein family containing domains homologous to the mammalian MBDdomain. Previous work has shown that all the members of the Arabidopsis MBD protein family arenuclear proteins and that at least 3 of them (namely MBD5, MBD6 and MBD7) are able to specificallybind methylated DNA in vitro. In accordance with this observation, MBD5 and MBD6 showheterochromatic localization.

To investigate the functional role of these plant MBD proteins in Arabidopsis, different experimentalapproaches have been performed. To verify if AtMBDs can act as transcriptional repressors (like theanimal counterparts), we fused the coding sequences (or portions of them lacking the MBD domain) toa heterologous GAL4 DNA binding domain. Transient transfections of these constructs intoArabidopsis protoplasts showed transcriptional repression of reporter genes that contained GAL4binding elements. Interestingly, this down-regulation appeared associated to a more general toxiceffect of the MBD protein overexpression. A similar effect was obtained by overexpressing twoArabidopsis histone deacetylases AtRPD3 and AtHD2, while the GAL4 activation domain alone wasable to strongly increase the activity of the reporter gene. In parallel, we attempted to identifymolecular partners of AtMBD proteins. To this end, we tested whether the AtMBDs, as themammalian homologues, were able to interact with histone deacetylase by GST pull downexperiments. Preliminary results show that AtMBD7 is capable of binding AtRPD3, at least in vitro.Furthermore, a two-hybrid screening was performed to reveal other proteins able to interact withmethyl-DNA binding proteins. The results of these screening procedures and the characteristics ofputative interacting partners will be presented.



PROTEOME ANALYSIS OF TRANSGENIC WHEATS OVER-EXPRESSINGPOLYPEPTIDES RELATED TO TECHNOLOGICAL PROPERTIES ANDPLANT DEFENCE

F. SCOSSA*, S. ROBERTI*, M. JANNI*, D. FERRI*, S. MASCI*, R. D'OVIDIO*, D.LAFIANDRA*, O. D. ANDERSON**, A.E. BLECHL**, D.D. KASARDA**, W.H. VENSEL**

*) Dipartimento di Agrobiologia e Agrochimica, Università della Tuscia, Viterbo, Italy**) U.S. Department of Agriculture, A.R.S., W.R.R.C., Albany, CA, USA

transgenic wheat, low molecular weight glutenin subunits (LMW-GS), polygalacturonase inhibitingprotein (PGIP), proteomics

In addition to conferring the desired characteristic/phenotype, transgene expression may affect othercomponents of plant metabolism. The effects of transgene expression, whether direct or indirect, arerelevant to basic research questions, industrial applications, and the issue of 'Substantial Equivalence'of GM plants.

Wheat endosperm proteins are of major importance in regard to the nutritional and technologicalproperties of wheat flours. Consequently, the potential effects of transgene introduction on theseproteins deserve consideration.

Here we present preliminary proteomic data concerning the effects of a transgenic modification onwheat endosperm proteins. We focused our studies on transgenic lines over-expressing a gene for alow-molecular-weight glutenin subunit (LMW-GS) correlated with dough quality, or over-expressing agene for a bean polygalacturonase-inhibiting protein (PGIP) that appears to protect against phytogenicfungi. The LMW-GS transgene is controlled by an endosperm-specific promoter, whereas PGIPtransgenes are controlled by either an endosperm-specific promoter or by the maize Ubiquitin1promoter, the latter resulting in constitutive expression throughout the plant.

We are carrying out a proteomic comparison between the wild type cultivar, the 'null' genotype(derived from the transgenic line, but has lost the transgene by segregation), and the transformed lineexpressing the transgenic protein, by using different two-dimensional electrophoretic techniques,followed by N-terminal amino acid sequencing of protein spots that represent differential expression.

Preliminary results show that the line with over-expression of the transgenic LMW-GS showssilencing or deletions of genes for several alpha-gliadin proteins, presumably part, or all, of a block ofgene copies, which are expressed in the untransformed cultivar. Moreover, in the same transgenic line,a higher protease activity has been observed, that is likely a mechanism in response of transgeneoverexpression.



A PROTEOMIC APPROACH TO VERIFY IN VITRO AND IN VIVOEXPRESSION OF WHEAT GLUTEN PROTEINS

P. FERRANTE*, S. MASCI*, R. D'OVIDIO*, D. LAFIANDRA*, G. RAZZINO*, B. MATTEI**

*) Dipartimento di Agrobiologia e Agrochimica, Università della Tuscia, Viterbo, Italy**) Dipartimento di Biologia Vegetale, Università La Sapienza, Roma, Italy

heterologous expression, gluten, gliadin, glutenin, proteomics

Wheat storage proteins (gluten proteins) are mainly composed by gliadins and glutenins. Bothcomponents are characterised by extended repeated regions rich in glutamine and proline. Gliadins aremonomeric proteins, whereas glutenins are polymeric proteins stabilised by disulphide bonds.However, it is possible that gliadins become part of the glutenin fraction when a different distributionof cysteine residues occurs.

Gluten proteins are responsible of qualitative properties of wheat flours and wheat-based foods. Doughviscoelastic properties depend mainly on gliadins and glutenins. Gliadins are considered the mainfactor triggering the most common human gluten intolerance, the celiac disease, although gluteninsubunits seem also involved.

In order to study structure-function relationships of gluten components, and to understand better thegenome organisation and the proteome composition of wheat endosperm, we are currently expressingheterologously (in E. coli) gliadin and glutenin genes, to obtain sufficient amounts of singlepolypeptides to be used in rheological and allergenicity tests. Since the great majority of the genes hereused derive from genomic clones, we are verifying if they are actually expressed in planta. To thisaim, we have used two-dimensional electrophoresis of heterologously expressed proteins incomparison to gluten patterns of the wheat cultivars used to construct the genomic library, and haveperformed N-terminal amino acid sequencing and mass spectrometry peptide mass mapping ofputative corresponding spots. Moreover, because bacterially expressed heterologous polypeptides mayundergo deletions within the proline-glutamine-rich repetitive region, we are using mass spectrometryto verify their correspondence with the expected polypeptides.



PROTEOMIC ANALISYS OF S-NITROSYLATED PROTEINS INARABIDOPSIS DURING THE DISEASE RESISTANCE RESPONSE

M.C. ROMERO-PUERTAS*, N. CAMPOSTRINI*, M. PERAZZOLLI*,**, F. ZANINOTTO*, L.ZOLLA***, C. CIAMBELLA***, P.G. RIGHETTI*, M. DELLEDONNE*

*) Università degli Studi di Verona, Dipartimento Scientifico e Tecnologico, Strada Le Grazie 15,37134 Verona**) Università Cattolica S.C., Via Emilia Parmense 84, 29100 Piacenza***) Dipartimento Scienze Ambientali, Piazzale Università, Università Tuscia, Viterbo

nitric oxide, proteomics, S-nitrosylation, plant disease

The free radical gas nitric oxide (NO), first characterised as an endothelium-derived relaxation factor,is a pleiotropic signaling molecule in animal cells, involved in a myriad of cellular processes includingneuronal signaling, blood pressure homeostasis and immune response. In plants, NO is an emergingregulator implicated in several processes and acts as a key signal in plant resistance to incompatiblepathogens by triggering resistance-associated cell death. Recent evidences indicate that in animaltissues NO regulates these diverse biologic processes by directly modifying proteins. NO and relatedspecies can oxidize, nitrate or nitrosylate proteins (Stamler et al., 1998). Nitrosylation refers to thebinding of a NO group to a transition metal or cysteine residue, is a reversible modification and plays acentral role in NO-mediated signalling (Stamler et al., 2001). Accumulating data suggest that manyproteins are nitrosylated by NO (for over 100 representative examples are available on line athttp://www.cell.com/cgi/content/full/106/6/675/DC1) and this suggests nitrosylation as an ubiquitouspost-translational modification regulating protein function. Indeed, nitrosylation shares many featuresin common with phosphorylation, the prototypic posttranslational modification involved in signaltransduction regulation. Both modifications are reversible and specific, allowing cells to flexibly andprecisely modify protein function in response to environmental signals (Mannick and Schonhoff,2002).

Proteomic studies represent a powerful complement to transcriptomic studies because they allowevaluation of the expressed proteins and potential post-transcriptional modifications as S-nitrosylation.To detect and characterize genes encoding proteins which are S-nitrosylated in Arabidopsis thalianaundergoing hypersensitive disease resistance response, a proteomic analysis has been undertaken.Proteins were extracted from Arabidopsis leaves 4h and 8h after challenge with an avirulent pathogen(Pseudomonas syringae pv. tomato carrying the avirulence gene avrB). Unchallenged leaves were usedas control. The so-called “biotin switch” method (Jaffrey et al., 2001) has been used to detect andpurify the pool of proteins S-nitrosylated in response to pathogen infection. After 2-Dimensionalelectrophoresis, standard maps of infected and non-infected leaves at two different time points havebeen generated and analysed with the PDQuest software. S-nitrosylated proteins differentiallyexpressed were extracted from the gels, and identification by mass spectrometry is in progress.Proteins will then be identified by searching the protein database as the National Center forBiotechnology Information (NCBI). We expect to characterize the function of the corresponding genesby comparing post-translational modification profiles and gene expression data deriving from genome-wide microarray analysis.



mtDNA VARIATION IN POTATO (SOLANUM TUBEROSUM) AND ITS WILDRELATIVES

N. SCOTTI*, S. COZZOLINO**, T. CARDI*

*) CNR-IGV, Institute of Plant Genetics Res. Div. Portici, Via Università 133, 80055 Portici (NA),[email protected]**) Dept. of Plant Biology, University of Naples “Federico II”, Via Foria 223, 80139 Naples, Italy

potato, Solanum tuberosum, mitochondrial DNA

Previous analyses in our laboratory showed a polymorphism in the rpl5-rps14-cob mitochondrialregion of Solanum tuberosum (cultivated potato) and the wild species S. commersonii and S.etuberosum (Bastia et al 2001 TAG 102: 1265-1272). In S. commersonii we detected the lack of therpl5-rps14-cob linkage and the presence of a novel rpl5-rps14-rps10-cox1 arrangement (Scotti et al2003 TAG 108: 87-94). The re-organization of the ribosomal protein genes during evolution is wellknown and is frequently due to the rearrangement and recombination of the mitochondrial genome ofhigher plants (Scotti et al 2004 Curr Genet 45 (6): 378-382).

In order to investigate the variability and the evolution of the mtDNA in Solanum spp., we analyzedthe organization of the rpl5-rps14-cob and rpl5-rps14-rps10-cox1 arrangements in a number of wildspecies of different origin, ploidy and EBN. PCR analyses were carried out on Solanum tuberosum andits wild relatives using specific primers for the coding regions of the above reported genes. Out of thesix mtDNA primer pairs tested, four detected interspecific polymorphisms in the linkage between rpl5-rps14, rps14-rps10, rps14-cob and ψ cob-rps10 genes. In addition, primers for rpl5-rps14 and ψ cob-rps10 genes showed also intraspecific variability in S. trifidum and S. chacoense species, respectively.

Cluster analysis allowed to arrange the species analyzed in 14 groups. Each group contained speciescharacterized by the same organization in the mitochondrial regions tested. Although the clusteranalysis did not show a strong correlation between species present in each group and theirmorphological taxonomy, our results showed that all species with EBN=2 were characterized by therps14-cob linkage; whereas those with EBN=1, except S. cardiophyllum, did not show it. In S.etuberosum (EBN=1) the primer pair for the rps14-cob linkage detected a shorter amplificationproduct indicating the presence of a shorter rps14-cob intergenic region. The rps14-cob linkage, absentin most of the primitive species (EBN=1), starts to be present in the species belonging to the seriesYungasensa that is involved in the evolution of the wild and cultivated Tuberosa species (Hawkes J.G.1990 The Potato: Evolution, Biodiversity and Genetic Resources. Belhaven Press, Oxford).



THE PRESENCE OF PHOSPHOLIPASE A2 ACTIVITY IN PLANTMITOCHONDRIA AS DEMONSTRATED BY A NEWSPECTROPHOTOMETRIC ASSAY

D. TRONO*, **, M. SOCCIO**, N. DI FONZO**, D. PASTORE*

*) Department of Animal, Plant and Environmental Sciences, Agricultural Faculty, University ofMolise, Via De Sanctis, 86100 Campobasso, [email protected]**) Experimental Institute for Cereal Research, SS 16 Km 675, 71100 Foggia, Italy

plant mitochondria, phospholipase A2, durum wheat, enzymatic assay

Phospholipase A2 (PLA2) is a family of enzymes that specifically hydrolyze glycerophospholipids atthe sn-2 position to yield free fatty acids and lysophospholipids. The existence of plant PLA2s has beenpresumed for some time and several studies have been reported concerning their implication inimportant cellular processes, such as lipid metabolism, auxin-stimulated cell growth and plant defenceresponse to wound stress and pathogen attack. Only recently PLA2s have been purified from differentplant sources and identified as “secretory” PLA2 (sPLA2) and “intracellular” PLA2 (iPLA2), the latterbeing located in the cytosol or associated to plasmatic membrane, whereas no so called “cytosolic”PLA2 (cPLA2) have been so far demonstrated in plants. Moreover, to date, no information is availableabout the existence of any PLA2 activity in plant mitochondria. In this work we have investigatedabout the existence of PLA2 activity in mitochondria from durum wheat, maize, barley, spelt, tomatoand lentil seedlings as well as from potato and topinambur tubers by using a new properly developedcontinuous spectrophotometric method based on the PLA2/lipoxygenase (LOX) coupled reactionsusing 1-palmitoyl-2-linoleoyl-sn-glycero-3-phosphocholine as substrate. Measurements have beencarried out at 234 nm by monitoring free linoleic acid generated by PLA2 by means of its conversioninto conjugated diene hydroperoxide due to LOX reaction. The new method has been validated on thebasis of an investigation carried out on a purified sPLA2 from bee venom and appears to be sensitiveand more simple, accurate and reproducible than other methods already described in literature. Wehave evaluated the existence of a PLA2 activity in plant mitochondria on the basis of the assays, bothin the homogenate and in the fraction of purified mitochondria, of phosphoenolpyruvate carboxylaseand cythocrome c oxidase, marker enzymes of the cytosol and mitochondria, respectively.Mitochondrial PLA2 existence has been demonstrated in durum wheat, maize, barley, spelt and tomatoseedlings, whereas it was absent in lentil seedlings and potato and topinambur tubers, thus showingdifferences among species/plant materials. Moreover, assays carried out using intact and osmoticallyruptured mitochondria suggest that at least 60% of PLA2 activity is present either in the matrix spaceor at the matrix face of inner membrane. A first characterization of the mitochondrial PLA2 in plantshas been carried out by using durum wheat, which displays a high mitochondrial/cellular ratio of PLA2

activity. We have observed the typical features of an enzymatic activity: heat inactivation, substratespecificity and dependence on protein content, pH and substrate concentration (with Km = 90 ± 8(S.D.) mM and Vmax = 3.0 ± 0.08 x 10-2 E.U. x mg-1 of protein). The enzymatic properties resemblethat of other PLA2, with pH optimum at 8.9, activation by Ca++ and inhibition by the Ca++ complexantEGTA. Moreover, we have observed competitive inhibition (Ki ≅ 14 mM) of mitochondrial activity by

palmitoyl trifluoromethyl ketone (PACOCF3), a classical PLA2 competitive inhibitor. On the otherhand, mitochondrial activity was unaffected by bromoenol lactone, a typical inhibitor of the Ca++-independent PLA2, thus suggesting the absence of a Ca++-independent PLA2 in DWM.

Our data show that a mitochondrial PLA2 exists in plants. Further investigations are required to shedsome light on the physiological role of the enzyme.



ARRAY ANALYSIS THE GENE EXPRESSION ASSOCIATED WITHCHLOROPLAST DEVELOPMENT IN BARLEY ALBINA MUTANTS

C. CROSATTI*, J. T. SVENSSON**, S. CAFFARRI***, C. MARÈ*, R. BASSI***, A. M.STANCA*, T. J. CLOSE**, L. CATTIVELLI*

*) Istituto Sperimentale per la Cerealicoltura, Via S. Protaso 402, 29017 Fiorenzuola D'Arda (PC),Italy**) Department of Botany and Plant Sciences University of California, Riverside, CA, 92521-0124 -USA***) Université Aix-Marseille II, Laboratoire de Génétique et Biophysique des Plantes Départementde Biologie 163, Avenue de Luminy, 13288 Marseille Cedex 09, France

oligo-array, barley, chloroplast development

We investigated four albina and xantha barley mutants representing successive steps in chloroplastbiogenesis and the corresponding wild type (WT) with the Affymetrix Barley1 GeneChip® (about22.000 probe sets) to assess the variation of gene expression associated with chloroplast developmentChloroplast defective mutants display a large number of genes with altered steady state levels withrespect to the corresponding WT. When mRNA isolated from leaves of mutant plants grown at 20°Cwere compared with mRNA extracted from green leaves of WT plants grown under the sameconditions, a number of probe sets were found more than 3 fold up- or down-regulated. Since themutants analyzed represent successive steps in the chloroplast biogenesis, the number of probe sets up-or down-regulated decreased according to progress in chloroplast development. Moving from alb-e16

(the most extreme mutant) to xan-b12 (the genotype closest to WT) the number of genes up- or down-regulated during growth at 20°C dropped from 1482 to 410 (from 12.6 to 4.7% of the probe setsdetected as present in each comparison) suggesting a progressive normalization of the transcriptome aschloroplast development proceed. Genes caractherised by induction orn repression in the mutantsrespect to WT, were found to encode for protein localized in the chloroplast as well as for nonchloroplast localized polypeptides, demonstrating the effetc of the chloroplast on the whole cellmetabolism. The comparative analysis of gene expression in the four mutants allowed theidentification of class of genes as well as of metabolic process whose normal expression is dependentfrom single steps of the chloroplast development. Beside the genes coding for photosynthetic relatedprotein (down regulated in the mutants) our analysis has found a tight control of the chloroplast on theexpression of the genes coding for the component of the protein synthesis machinery (ribosomalprotein, tRNA ligase, elongation/initiation factors).



ARRAY ANALYSIS OF COLD ACCLIMATION IN BARLEY ALBINAMUTANTS REVEALS THE KEY ROLE OF THE CHLOROPLAST DURINGPLANT ADAPTATION TO LOW TEMPERATURE

J. T. SVENSSON*, C. CROSATTI**, C. CAMPOLI**, R. BASSI***, A. M. STANCA**, T. J.CLOSE*, L. CATTIVELLI**

*) Department of Botany and Plant Sciences, University of California, Riverside, CA, 92521-0124 -USA**) Istituto Sperimentale per la Cerealicoltura, Via S. Protaso 402, 29017 Fiorenzuola D'Arda (PC),Italy***) Université Aix-Marseille II, Laboratoire de Génétique et Biophysique des Plantes Départementde Biologie 163, Avenue de Luminy, 13288 Marseille Cedex 09, France

cold resistance, oligo array, barley, chloroplast mutants

A number of works have suggested the involvement of the chloroplast and chloroplast activity in coldacclimation. Previously, we have shown that barley plants carrying a mutation preventing chloroplastdevelopment, beside the expected albino phenotype, are completely frost susceptible as well asimpaired in the expression of several cor (cold-regulated) genes. We investigated four albina andxantha barley mutants representing successive steps in chloroplast biogenesis and the correspondingwild type (WT) with the Affymetrix Barley1 GeneChip® (about 22.000 probe sets) to assess the effectof the chloroplast on the expression of cold-regulated genes. First, by comparing control WT vs coldhardened WT plants a set of 1570 probe sets up- or down-regulated more than 3 fold were identified.Expression of the WT cor genes were analyzed in control and cold hardened mutants allowing us toseparate the WT cor genes into three main classes: (1) chloroplast independent (cold regulated to asimilar extent in all genotypes); (2) chloroplast dependent (cold regulated only in WT and notregulated by cold in any mutants); and (3) chloroplast development dependent ( cold regulated in WTand in the mutants closest to WT). Notably, about 70% of the WT cor genes showed a tight chloroplastcontrol among these are over a hundred genes involved in the protein synthesis machinery, but,surprisingly, none of previously known cor genes. Although the mutants failed to cold regulate most ofthe genes associated with cold acclimation in the WT, they still retained the ability to induce the cbffamily of transcription factors. This suggests that additional factors deriving from the chloroplast inaddition to cbf’s are required to promote the molecular changes associated with cold acclimation.



PROGRAMMED CELL DEATH PATTERNS IN ALFALFA (MEDICAGOSATIVA L.) REPRODUCTIVE ORGANS

S. VAROTTO*, S. CITTERIO**, E. ALBERTINI***, G. BARCACCIA*

*) Dipartimento di Agronomia Ambientale e Produzioni Vegetali, Università di Padova, Agripolis, Viadell’Università 16, 35020 Legnaro, Padova, Italy**) Dipartimento di Scienze dell’Ambiente e del Territorio, Università di Milano – Bicocca, Piazzadella Scienza 1, 20126 Milano, Italy***) Dipartimento di Biologia Vegetale e Biotecnologie Agroambientali, Università di Perugia, BorgoXX Giugno 74, 06121 Perugia, Italy

plant reproduction, PCD, TUNEL assay, MOB domain

Many cells may die during development and differentiation in plants. In angiosperms, late in thereproductive phase, programmed cell death (PCD) occurs in both the anther and the pistil. Productionof functional male gametes relies on deterioration and death of the anther tapetum, a tissue whose mainfunction appears to nurture microspores with critical surface molecules and to allow pollen dispersal atmaturity. The pathway of female gametogenesis frequently begins with the death of all but one of thereduced megaspores, with surrounding nucellar cells of the endothelium that degenerate in concertwith embryo sac expansion. In alfalfa, in the ovule usually a primary sporogeneous cell develops into amegaspore mother cell which directly undergoes meiosis, resulting in a tetrad of linearly arrangedhaploid megaspores. Only the chalazally oriented megaspore undergoes three rounds of mitosis anddevelops into an eight-nucleated embryo sac, while the other three megaspores are subjected toprogrammed death. Usually, female and male meiosis are not synchronous in alfalfa: the microsporemother cells enter into meiosis earlier than the megaspore mother cells. As a consequence, when themegagametogenesis occurs in ovules, uni-nucleated microspores are already released from the tetradsand microgametogenesis has begun. At this stage of microspore development, shortly before mitosis,in alfalfa the nourishing tissue of the anther has begun its degeneration. It is well known that tapetumafter having secreted beta-1,3 glucanase for callose dissolution undergoes cell death, depositing its cellcontent within the exine cavities thus forming the pollen coat. The mechanisms that control PCD inplants are largely unknown, even though the cellular deterioration patterns that have been described forplant reproductive tissues are similar to those observed in other plant and animal developmentaltissues. Relatively unusual in plant reproductive cell death is that it often involves cell rupture,releasing cellular contents that are incorporated into functional components of other differentiating celltypes. Tissues and cells that most prominently undergo PCD during reproductive development inalfalfa were analyzed by means of DNA fragmentation, a phenomenon accompanying PCD in avariety of cell systems, using the TUNEL assay. TUNEL staining was detectable in the degeneratedmeiotic megaspores and nucellar cells of the endothelium surrounding a viable embryo sac as well asin the degenerated tapetum cells of anthers containing mature pollen grains. Preliminary resultscollected in an alfalfa reproductive mutant have shown that a member of the Mob1-like gene familycan be associated to PCD in ovules and anthers. Localization signals of both transcripts and proteinswithin female and male reproductive tissues overlapped spatially and temporarily the signals related tothe occurrence of DNA fragmentation. Overall results were consistent with the association of Mob1-like gene products with PCD in alfalfa reproductive organs. Interestingly, protein database searches

revealed that in Arabidopsis and rice the MOB domain (KOG0440) can be combined with elements ofthe NB-ARC domain, a signaling motif shared by animal cell death gene regulators as well as withLeucine-Rich Repeats, short sequence motifs involved in protein-protein interactions.



IDENTIFICATION AND PRELIMINARY CHARACTERIZATION OF miRNAsIN MAIZE

E. MICA, L. GIANFRANCESCHI, F. BRIANI, M.E. PÈ

Department of Biomolecular Sciences and Biotechnology, University of Milano, Via Celoria 26,20133 Milano (Italy)[email protected]

microRNA, gene expression, Zea mays L.

MicroRNAs (miRNAs) are short RNAs, 21-24 nt in length, which play an important role in post-transcriptional gene regulation. The mechanism of action is similar, but not identical, to RNA-silencing; miRNAs are part of a ribonucleic complex that degrades the mRNA of a gene, or regulatesits translation level upon base paring between the sequence of the small RNA and target mRNA. Asingle miRNA is produced from one to several longer primary transcripts (pre-miRNA), that assume astem-loop structure, originating from MIR genes. Plant miRNAs are very similar to those characterizedin animals, with some peculiarities. Plant miRNAs are homologous to the coding sequence of targetmRNAs, but the region of homology can tolerate up to 2-3 mismatches.

To-date MIR genes and miRNAs have been identified in Arabidopsis thaliana and Oryza sativa. Bycomparison analysis between these 2 species, it’s clear that the 21-24 nt sequence and stem-loopstructure are highly conserved (Bartel and Bartel, 2003). Eight miRNAs out of the 19 sequences foundin A. thaliana are also present in O. sativa. Target genes of these miRNA seem to be transcriptionfactors.

Recently is has been shown in maize the involvement of a miRNA in the regulation of leafdevelopment (Juarez et al., 2004).

Here we present our results in identifying miRNAs sequences in Zea mays L and their correspondingMIR genes, evolutionary related to those of rice. The maize EST data base was screened for maizeexpressed sequences containing complementary regions to rice miRNAs. BLAST algorithm, with amaximum of 3 mismatches allowed, was used to investigate for potential miRNA target in maize.Seven of the 8 rice miRNAs tested produced at least one significant blast hit in maize. These maizeESTs have putative functions often similar to those supposed for target genes in rice.

Starting from rice pre-miRNAs sequences we designed primers to amplify correspondent maizegenomic regions. As expected every pair of primers gave rise to several PCR products. These productswere cloned and sequenced. Based on sequence homology and on the determination of their RNA moststable secondary structure, among the 8 miRNAs tested we have identified 5 potential miRNAs inmaize. These 5 sequences have one, or more, pre-miRNA sequences with a stem-loop folding structurequite linear and with few bulges, compatible with the expected structure of pre-miRNAs. Thesesequences have, with few exceptions, no similarity with any annotated EST. The sequence of all ofthem was found within the hypomethylated maize genomic sequences annotated in the TIGR database.Expression pattern was determined by Northern blot analysis of low-weight RNAs purified from

different maize tissues in three different genotypes. This analysis showed tissue and genotipic variationrelative to expression level for all the identified miRNAs.

ReferencesBartel and Bartel (2003). Plant Physiol. 132: 709-719Juarez et al. (2004). Nature 428: 84-88



INVESTIGATION ON P. HYBRIDA MEI2 AND MIP1 GENES

M. VIVIANI, M. PURELLI, S. LONGHI, G.B. TORNIELLI, M. PEZZOTTI

Department Scientific and Technological, University of Verona, Strada Le Grazie 15, 37134 Verona(Italy)

Petunia hybrida, mei2, mip1, meiosis, gene function

In the eukaryotic kingdom the meiotic process plays an important role in providing genetic variabilityand, not surprisingly, this biological event is widely conserved among different organisms. In S.pombe, an eukaryote considered as a model for studying meiosis, the proteins Mei2 and Mip1 arefundamental actors of this process: Mei2, an RNA binding protein, in the cytoplasm, is responsible forthe synthesis of pre-meiotic DNA, while, in the nucleus, promotes the first meiotic division. Mip1 is awd-repeat protein that interacts weakly with Mei2 in the cytoplasm, probably helping Mei2 folding,but its function may be more complex.

The aim of this work is to elucidate the role of mei2 and mip1 genes in Petunia hybrida, an organismconsidered as model in plant studies. We reported the isolation of the two genes (PhMEI2, PhMIP1),of their regulative regions, and the isolation of their insertional mutants by a reverse genetics approachto investigate their functions.



GAMETOPHYTIC FACTOR 1, A GENE INVOLVED IN POLLEN-PISTILINTERACTION

M. VIANELLO*, M. E. PÈ*, M. PEZZOTTI**, M. SARI GORLA*

*) Department of Biomolecular Sciences and Biotechnology, University of Milano, Via Celoria 26,20133 Milano (Italy)[email protected]**) Dipartimento Scientifico e Tecnologico, University of Verona, Ca' Vignal 1 - Strada le Grazie 15,Verona (Italy)[email protected]

pollen function, pollen-pistil interaction, maize, AFLP-TP

The male gametophyte has a very short life-span in the life cycle of higher plants and the pollen grainis a rather simple two- or three-celled organism. However, it is able to exist as a free organism, and,once on a female receptive surface, to germinate and produce a pollen tube that, through stylar tissues,reaches the ovary and achieves fertilization. This process represents a peculiar and striking example ofcell migration in plant development, largely based on cell surface interactions between the malegametophyte and the female sporophytic tissues.

Maize pollen is an interesting model for studying gene expression during tube growth and pollen-styleinteraction, since in maize pollen tube growth is extremely large (measured in cm) as compared topollen grain diameter (in mm). In order to identify and isolate genes involved in pollen-pistilinteraction, we are focussing on the maize mutant Gametophytic factor1 (Ga1), one of the severalgenes shown to be responsible of the phenomenon of “cross sterility”.in maize and other species. Infact, Gametophytic factors strongly affect pollen tube growth: pollen bearing recessive ga alleles aremuch less competitive as compared to pollen carrying dominant Ga alleles in stylar tissues carryingthe same Ga dominant allele. In maize, several Gametophytic factors have been described, which showdifferent degrees of competition between the two types of pollen. For the Ga1 gene it has beenidentified a strong allele, named Ga1-s, whose presence in stylar tissue completely prevents ga1pollen to accomplish fertilization. Onto female tissues from either homozygous Ga-1s/Ga1-s orheterozygous Ga1-s/ga1 plants, ga1 pollen grains germinate and pollen tubes grow, although neverreaching the ovules. The feature of the mutant pollen tubes and the mechanism of cross-incompatibility have not been described yet. However this mutant is a powerful tool to shed light onthe genetic control of pollen function and on the molecular basis of pollen-pistil interaction.

To address these phenomena, AFLP-TP technique was performed in order to identify differentiallyexpressed transcripts in two maize isogenic lines characterized by the presence of the dominant (Ga1-s) or recessive (ga1) allele. Non pollinated Ga1-s/Ga1-s and ga1/ga1 silks (maize modified styles),and Ga1-s/Ga1-s and ga1/ga1 silks pollinated with either Ga1-s or ga1 pollen grains were compared.Here we present our preliminary data on non-pollinated silks and silks 4 hours after pollination, using32 primer combinations for selective amplification. The analysis of our amplification profiles allowedthe identification of 170 putative genes differentially expressed, grouped in several classes accordingto the various degrees of modulation observed between the two genotypes and/or different type of

pollination. Most of the AFLP fragments were re-amplified by PCR, cloned in pCR4-TOPO vector andtheir sequence is currently being analysed.



FUNCTIONAL CHARACTERIZATION OF AN ACTIVATED TAGGEDARABIDOPSIS MUTANT DEFECTIVE IN PLANT MEIOSIS

F. CONSIGLIO*, R. MAISTO**, A. ERRICO**, R.A. BRESSAN***, C. CONICELLA*

*) CNR-IGV, Institute of Plant Genetics, Research Division Portici, Via Università 133, 80055 Portici,Italy**) DISSPA Dept. Soil, Plant and Environmental Sciences, University of Naples "Federico II", ViaUniversità 100, 80055 Portici, Italy***) Dept. Horticulture & Landscape Architecture, Purdue University, West Lafayette, Indiana (USA)

Arabidopsis, gain of function, meiosis gene isolation

The mutagenesis approach in Arabidopsis is crucial in allowing the genetic dissection of meiosis.Besides the loss-of-function mutations which are not able to identify essential genes or functionallyredundant ones, the activation tagging mutagenesis can widen the possibility to identify meiotic genes.

A gain of function approach was employed in Arabidopsis thaliana (Ath) using T-DNA tagging with avector containing multimerized transcriptional enhancers from CaMV35S (Weigel et al. 2000. PlantPhysiol 122: 1003-1013).

Secondary screening of Ath lines generated by Bressan lab resulted in the identification of a mutantline b16 exhibiting siliques with few seeds. We report the molecular, genetic, and cyto-istologicalanalyses of this line.

We have identified by TAIL-PCR the position of the T-DNA insertion in b16 chromosome III. RT-PCR was performed to analyse the expression patterns of the genes flanking the T-DNA insertion anda chromatin remodelling factor was identified as the activated gene in b16 mutant. A functionalcomplementation analysis is being used to confirm the role of the over-expressed gene in plant meiosisprocess. Homozygous b16 was crossed to wild type and in F1 progeny all plants exhibit the mutantphenotype coupled to selective marker. The F2 generation showed an aberrant segregation ratio.However a previous self-cross segregation analysis of hemizygous b16 evidenced that the mutation issporophytically expressed and monogenic. Histological investigations showed that 50% of the ovulesarrested the embryo sac development. As concerning the pollen, there is a dropping in stainabilityvalues. Univalents have been identified in micro- and macrosporogenesis. Moreover, duringmicrosporogenesis segregation irregularities have been observed in meiosis I and meiosis II. Furtherinvestigation is required to assess chromosome behaviour in macrosporogenesis.



CONSTITUTIVE COSUPPRESSION OF THE GA 20-OXIDASE 1 GENE INTOMATO LEADS TO SEVERE DEFECTS IN VEGETATIVE ANDREPRODUCTIVE DEVELOPMENT

I. OLIMPIERI, R. CACCIA, M.E. PICARELLA, G.P. SORESSI, A. MAZZUCATO

Department of Agrobiology and Agrochemistry, University of Tuscia, Via S. Camillo De Lellis snc,01100 Viterbo (Italy)[email protected]

cosuppression, GA 20-oxidases, Lycopersicon esculentum, tomato

Knowledge on the role of hormones in plant growth and development has resulted from the studies ofexogenous application and endogenous measurements of hormone levels in various wild-type andmutant genotypes. The recent identification of several genes encoding hormone biosynthesis andresponse-related enzymes has greatly facilitated our understanding of the function and regulation ofplant growth regulators. For instances, along the gibberellin (GA) biosynthesis pathway, thetranscription of genes involved in the final part of the metabolic chain has been reported as beingstrictly controlled by both endogenous and exogenous cues. In particular, the transcription of the GA20-oxidase (GA20ox) gene family is recognized as a key regulation step for the control of active GAs.In tomato, GA20ox genes form a small multigenic family composed of three members, which aredifferentially transcribed during development. Previous results obtained in our laboratory have shownthat LeGA20ox1 is a tightly regulated gene, which responds to the stimulus of pollination andfertilization to drive the growth of the ovary into a fruit.

To study in more depth the role of LeGA20ox1 in the vegetative and reproductive development oftomato plants, a cosupression approach was undertaken in order to silence the expression of the gene.A 638-bp truncated 5’ sequence of the gene was cloned in sense orientation downstream to the CaMV35S promoter in the pBI121 vector. Cotyledon explants from cv Chico III tomato plants weretransformed with the described construct via Agrobacterium-based standard protocols. Seven primarytransformants that grew extremely slowly were rescued, controlled with PCR amplification and grownin tunnel; all of them showed a more or less severe brachitic habit. Since such plants were not fertilewhen open pollinated, one plant showing a severe dwarf phenotype was selected and in vitromicropropagated in order to have a clone of 25 individuals. Untransformed control plants wereobtained by in vitro propagation of a wild-type near-isogenic plant. Transgenic plants weredramatically affected in the vegetative development; internodes were much shorter than in the controldetermining the severely dwarf phenotype. Moreover, transformed plants showed leaflets that weresmaller and narrower than the control.

Although no difference was noticed in the inflorescence phenotype, flowers of the cosuppressed plantswere somehow defective in comparison to the control: the insertion of floral organs was not regularlyalternated, the corolla was paler and the anther cone was more elongated and slightly distorted. Malesporogenesis was severely affected in transformed plants and the number of viable pollen grains wasvery low. In the pistil, ovule development and female meiosis appeared unaffected and ovaries were

regularly stimulated to grow when pollinated with normal pollen. On going pollination experimentswill reveal if and at which extent the inhibition of ga biosynthesis is important in the completion ofseed and fruit development. If these plants will prove to be female fertile, they could be easily crossedwith mutants involved in hormone synthesis or response or with other genetic stocks suitable to studythe interaction of gas with other classes of growth regulators.



CHARACTERIZATION OF TWO NEW VIVIPAROUS MUTANTS IN MAIZE

A. GIULINI*, D. DURANTINI*, N. RASCIO**, N. LA ROCCA**, R. TUBEROSA***, S.STEFANELLI***, C. SANGUINETI***, G. CONSONNI*, G. GAVAZZI*

*) Department of Crop Production, Via Celoria 2, 20133 Milano, Italy**) Department of Biology, Via Colombo 3, 35121 Padova, Italy***) Department of Agroenvironmental Sciences and Technology, Via Farin 44, 40127 Bologna, Italy

abscisic acid, viviparous, transpiration,, gene expression

The viviparous mutants of maize can be classified into two classes according to their phenotypes andmetabolic effects; class 1: those impaired in ABA synthesis. If blocked in early steps of biosynthesisthey have reduced or suppressed carotenogenesis (pale green or white seedling), whereas those with ablock downstream of the first committed step in ABA biosynthesis have normal carotenogenesis(green seedling); class 2: those concerned with the perception or the signal transduction of thehormone, referred to as response mutants (green seedling). All known viviparous mutants are recessiveand their common characteristic is that seed maturation fails to go to completion leading to prematuregermination. We recently identified two new viviparous mutants: vp374* and vp105*. They havegreen seedlings and are thus not blocked in carotenogenesis. They presumably represent mutationallesions in steps of the hormon synthesis downstream of xanthophylls production or they are responsemutants impaired in the ABA receptor(s) or in the signal transduction.

vp374* complements vp1 or vp8 (1L) and is located on chromosome 1L, as determined by crosses withthe B-A translocation stocks. vp105*, not allelic to vp1 , vp8 and vp10 exhibits a characteristicphenotype consisting of green shoot apex with occasional vivipary. Their allelism test with vp14 (1L)is underway.

Viability of homozygous mutants is different between the two genotype: vp105* is lethal at theseedling stage while about one–half of the vp374* mutants (13 out of 42) can be rescued. To figure outwhy vp105* dies after the differentiation of the 4th leaves we analyzed the stomatal conductance(mmol m-2 s-1) in the second leaf of seedlings grown for 14 days in light saturation conditions, atdifferent relative humidity. The results indicate that vp105* is likely impaired in stomatal regulation.Mutant seedlings in fact show an increased transpiration loss in normal conditions if compared tonormal siblings. The concentration of ABA in the leaves of the two mutants does not differ from thatof wild-type seedlings whereas the embryonic ABA content is significantly lower.

Mutants deficient in ABA biosynthesis or in response to ABA signals show a difference in theirsensitivity to exogenous hormone. ABA deficient mutants grown in presence of ABA, are highlyinhibited in their growth, those insensitive are inhibited to a much less extent. We tested the twomutants and they both appear less inhibited, even though to a variable extent, than their wild-typecounterparts.

To test how vp374* and vp105* respond to ABA in terms of ABA responding genes, we performedan RT-PCR analysis. To this aim 30 days old embryos of the mutant were collected, incubated on

solidified MS medium without or with ABA (10mM) for 48h and then assayed for presence of threeABA inducible genes Glb1, Rab17 and Lea. The results obtained indicate that the mutants maintaintheir capacity to respond to ABA by inducing the three genes. This observation coupled with theembryo culture analysis and the measurements of ABA concentration in the seed suggest that theviviparous phenotype of vp374* and vp105* is due not to a lack of response to the hormone but ratherto a lack of ABA in the mutant seed.



PIN-1 EXPRESSION IN MAIZE SHOOT APICAL MERISTEM DURINGDEVELOPMENT

N. CARRARO*, J. TRAAS**, S.VAROTTO*

*) Dipartimento di Agronomia Ambientale e Produzioni Vegetali, Università di Padova, Vialedell’Università 16, 35020 Legnaro (PD), [email protected]**) Laboratoire de Biologie Cellulaire, INRA, Route de Saint Cyr, 78026 Versailles cedex, France

Arabidopsis, auxin, PIN-1, Shoot Apical Meristem (SAM), maize

The shoot apical meristem (SAM) of the Angiosperms is established during embryogenesis andinitiates post-embryonic development of the aerial parts of the plant. This complex structures grow ina stereotypic fashion determined by the genetic program of the plant itself and respond to the externalenvironmental cues (Traas & Vernoux 2002 Phil. Trans. R. Soc. Lond. B. 357:737-347). This is theway that the phyllotaxis is established and, even though we are far away from understanding all thesteps of this complex process, some of the implied genes have been identified. Among those weconsider the PIN-FORMED. This multigenic family includes eight members in Arabidopis thalianaand the most extensively studied up to date is PIN1. It controls lateral organ initiation by regulatingcell differentiation at the periphery of the meristem. In fact it is supposed to encode a membraneprotein that acts as efflux carrier for auxin. So far it is known that auxin can be transported through thestem by diffusion or by active transport. This second flux is supposed to be the major one and isestablished thanks to the influx membrane carriers (proteins encoded by the AtAUX gene family) andthe efflux membrane carriers cited above. These last ones can be re-oriented during the formation ofthe lateral structures of the stem, in a way that the plant can accumulate auxin at the place of the futureoutgrowing primordial (Friml et al. 2003 Nature 426:147-153). In other words you have auxin maximaat the periphery of the SAM where the new forming primordia will grow. The auxin transport is still acomplex phenomenon that implies several other genes: in particular the analysis of the pin1 and pidmutants has provided a great help to comprehend this phenomenon. These phenotypes are similar andshow the so-called "naked stem", which has no lateral structures but still maintain a growing SAM.These genes have been studied in other model species, for example Brassica juncea and show similarsequences and protein structures, so that in the future they can be cloned and analysed in species ofagronomic interest. In this frame we are using an antibody raised against AtPIN1 inimmunolocalization experiments in sections of maize SAM, during post-embryonic development.Moreover two maize mutants Barren stalk1 (Ritter et al. 2002 American Journal of Botany 89(2): 203-210 defective in axillary meristem development and barren inflorescence 2 (McSteen & Hake 2001Development 128:2881-2891) which regulates axillary meristem development in maize inflorescencewill be analysed with the same antibody. Molecular experiments are in progress in order to identifyPIN sequences in maize genome.



THE GLOSSY1 GENE OF MAIZE IS INVOLVED IN DIFFERENT ASPECTSOF EPIDERMIS DEVELOPMENT

M. STURARO*, M. MOTTO*, H. HARTINGS*, F. SALAMINI**, E. SCHMELZER**

*) Istituto Sperimentale per la Cerealicoltura, Sezione di Bergamo, Via Stezzano 24, 24126 Bergamo(Italy)[email protected]**) Max-Planck Institut für Züchtungsforschung, Köln, Germany

maize cuticle, trichome development, glossy mutants, SEM and TEM analysis

The surfaces of many plants are covered with a cuticle secreted by epidermal cells, which deservesseveral protective roles and consists of a reticulated cuticle membrane layer covered with amorphousepicuticular waxes.

The maize Glossy1 (Gl1) gene is one of the many loci involved in epicuticular wax biosynthesis onseedling leaves. Mutations at this locus confer a glossy phenotype to the first 5 to 6 leaves in contrastto the dull appearance of their wt counterparts.

To gain insights into Gl1 functions, transcriptional analysis and microscopic inspection of mutantepidermis were performed. From the expression profile it turned out that Gl1 function is not restrictedto the juvenile developmental phase of the maize plant. This suggests a broader role of the geneproduct than previously predicted on the basis of the visual phenotype of gl1 mutants. Moreover,ultrastructural analysis of leaf epidermis indicated a pleiotropic effect of the gl1 mutation on juvenileleaf development. Besides reduction in wax synthesis, SEM analysis revealed alterations of leaftrichomes, namely decreased trichome size and increased trichome frequency, on gl1 seedling leaves.Analysis of the cuticle with TEM highlighted a strong reduction of cuticle membrane thickness inmutant seedlings. Similarly, mutations in the Arabidopsis Wax2 gene, a putative Gl1 homologue, altercuticle membrane synthesis, epicuticular wax production and trichome morphology. However, in wax2mutants abnormal cuticle development triggers post-genital organ fusion of adult organs soon afteremergence. This trait is not associated with gl1 mutations pointing to a different role of maize andArabidopsis cuticles in leaf development.



rolD ONCOGENE STIMULATES MERISTEMS OF BOTH AXILLARYSHOOTS AND LATERAL AND ADVENTITIOUS ROOTS IN ARABIDOPSISPLANTS

M.L. MAURO*, G. FALASCA**, S. D'ANGELI**, D. ZAGHI**, M.M. ALTAMURA**, P.COSTANTINO*

*) Dept. of Genetics and Molecular Biology, University "La Sapienza" of Rome, P.le Aldo Moro 5,Roma**) Dept. of Plant Biology, University "La Sapienza" of Rome, P.le Aldo Moro 5, Roma

rolD, arabidopsis, meristems, flowering, rooting

We are previously demonstrated that rolD oncogene from Agrobacterium rhizogenes stimulates intobacco plants an early and enhanced flowering process through formation of supranumerary axillarymeristems which develop as sylleptic inflorescences.

The effect of rolD gene on flowering is also shown in some other plants such as tomato, andosteospermum oecklonis.

Our study in Arabidopsis confirms the effect of rolD on flowering, because the main event observed inthe transgenic plants is an enhancement of coinflorescence formation.. The histological analysis showsthat, since very early in plant development, meristems are formed at the axils of the rosette leaves.These meristems develop as vegetative buds, the most of which becoming coinflorescences. Moreoverplants expressing rolD also show an altered development of the root apparatus. In particular secondary(lateral) and adventitious roots are promoted. Thin cell layers cultured in vitro confirm this role ofrolD in enhancing post-embryonic root meristems. Infact the rhizogenic response increases both, in thepresence of auxin and under hormone free conditions.

It is known that secondary and adventitious roots are specifically regulated by auxin. To shed morelight on the role of rolD on post-embrionic rhizogenesis, complementation analyses between someauxin mutants, with a reduced secondary roots formation and rolD plants are in progress.

Since the rolD gene product has an ornithine ciclodeaminase activity, a key step in the prolinesynthesis, we discuss the possible role of this aminoacid in the rolD -induced meristem formationduring plant development.



SENESCENCE IN LEAVES AND CELL CULTURES OF M. TRUNCATULA

R. DE MICHELE*, A. FERRARINI***, E. BARIZZA*, F. CARIMI**, M. ZOTTINI*, M.DELLEDONNE***, M. TERZI*, F. LO SCHIAVO*

*) Department of Biology, University of Padova, Via G. Colombo 3, 35131 Padova (Italy)[email protected]**) Istituto di Genetica Vegetale, Sezione di Palermo - CNR, Corso Calatafimi 414, 90128 Palermo(Italy)***) Dipartimento Scientifico e Tecnologico, Università degli Studi di Verona, Strada Le Grazie 15,37134 Verona (Italy)

Medicago truncatula, senescence, cDNA-AFLP, cell culture

Leaves at two different developmental stages, mature and senescent (25% of yellow leaf area), havebeen collected from climate chamber-grown plants of Medicago truncatula. mRNA was extractedfrom these leaves and used for cDNA-AFLP analysis to compare the gene expression in mature vs.senescent leaves. By using 32 different combinations of primers we identified about 500 fragments ofcDNA differently expressed during leaf senescence. These fragments are being cloned and will beeventually sequenced.

At the same time a cell line (JR) of M. truncatula var. Jemalong has been selected and characterised inits morphogenetical features and physiological growing parameters and for its differentiativecapacities: fresh weight, dry weight and cell death percentage. Studying senescence in cell cultureoffers several advantages. The natural senescence process has been characterized in the JR line,observing the variation of cell death (Evan Blue staining) and DNA analysis (“laddering”).

The differentially expressed genes, isolated in the cDNA-AFLP experiments, will be analysed inleaves and cell culture.



IDENTIFICATION OF A PUTATIVE ETHYLENE RECEPTOR GENEEXPRESSED DURING APPLE DEVELOPMENT AND RIPENING

S. STELLA*, F. COSTA*, W.E. VAN DE WEG**, C. MALIEPAARD**, S. SANSAVINI*

*) Dipartimento di Colture Arboree, University of Bologna, 40127 Bologna, [email protected]**) Plant Research International, Wageningen, P.O.Box 16, 6700AA, The Netherlands

ethylene receptor genes, gene expression, molecular mapping, QTL, functional markers

Ethylene is a small readily diffusible phytohormone regulating many aspects of plant life-cycle, suchas seed germination, steam elongation, fruit ripening, abscission, leaf and flower senescence, and plantresponsiveness to environmental stimuli (Abeles et al, 1992). In climacteric fruit, ethylene plays a keyrole during ripening, controlling many aspects such as flesh softening, flavour, and pigmentproduction. Its biological activity can be regulated at both levels of synthesis and perception, and iscarried on by a cascade of the signal transduction starting with ethylene’s binding to its receptors. Theplant’s ability to modulate the different responses to ethylene in a tissue-specific manner indicates thatethylene’s perception is a complex highly regulated process. We have isolated in apple (Malus xdomestica Borkh) a putative gene (Md-ERS1), a member of the ERS (Ethylene Responsive Sensor)receptor family, showing a high similarity to other homologous and heterologous sequences (Pyruscommunis, Prunus persica and Arabidopsis thaliana). Md-ERS1 was tested on ten apple cultivarsshowing an allelic variation between ‘Prima’ and ‘Fiesta’, parents of a crossing population in which ahighly saturated map was constructed (Maliepaard et al.,1998) . This allelic polimorphism allowed themarker’s positioning in the linkage group 3 of the ‘Fiesta’ map. A minor QTL for fruit firmness wasalso previously identified on linkage group 3 (Maliepaard et al., 2001). In order to study thesignificance of Md-ERS1 on the QTL position , we analysed the mapped QTL using two differentanalysis methods (interval mapping and MQM mapping). The firmness data was collected in fourdifferent places in Europe. The results showed an increase in the LOD value coincident with the mapposition of Md-ERS1 and was always “conserved” in the genomas of the apple grown in the other sites.In order to evaluate the role of Md-ERS1 on fruit ripening, its transcript level was analyzed in applesamples characterized by different stage of maturation, and a differential expression of this gene wasobserved.



EXPRESSION PATTERN OF PHOTOSYNTHESIS-RELATED GENES IN ADURUM WHEAT “STAY GREEN” MUTANT

P. RAMPINO*, G. SPANO**, S. PATALEO*, G. MITA***, J. A. NAPIER****, N. DI FONZO**,P.R. SHEWRY****, C. PERROTTA*

*) Department of Biological and Environmental Sciences and Technologies, University of Lecce, ViaProv.le per Monteroni, 73100 Lecce, Italy**) Experimental Institute of Crop Research, Foggia, Italy***) ISPA-CNR Lecce, Italy****) Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, U K

Triticum durum, stay green mutant, photosynthesis-related genes, senescence

Elucidation of the molecular basis of senescence is not only of fundamental interest but also relevant tothe genetic improvement of crop plants. A powerful tool for the analysis of the senescence process isthe molecular characterisation of mutants that are defective in an aspect of senescence, such as the“stay-green” mutants. “Stay-green” mutants are characterised by the persistence of the green colour ofleaves for longer than in parental genotypes. In particular, some “stay green” mutants can arise fromdelays in the initiation of senescence and its rate of progress. Consequently they continue tophotosynthesise for longer than usual, they are therefore said to be “functional stay-green” (Thomasand Howarth, 2000). A strong relationship between the extension of the photosynthetic capacity andgrain yield was observed in cereals such as maize and shorgum. In agronomic terms, these mutantshave higher kernel weights and this observation has been exploited by maize breeders. Althoughbreeders have extensively used such material for yield improvement, little is known about theunderlying genetics and molecular biology of the trait(s), even though detailed analyses have beenperformed in maize and sorghum (Tao et al., 2000; Thomas and Smart, 1993).

One approach to determine the molecular basis of the “stay-green” phenotype is to identify thedifferences between mutant and parental plants in their regulation of specific genes during thesenescence process.

A mutant of durum wheat cv trinakria (designated 504) was characterised by delayed leaf senescenceand analysis of photosynthetic parameters showed that it was functionally “stay green”. An in vitrosystem was established to mimic senescence by incubating wheat leaves in the dark allowing theexpression of genes known to be differentially regulated during senescence to be determined.Moreover, cdnas found to be differentially expressed in the mutant were cloned and sequenced,showing homologies with photosynthesis-related genes.

RT-PCR analyses were carried out using RNA samples from mutant and control plants, during bothnatural and artificially induced senescence, confirming the altered expression profiles of this genes inthe stay green mutant 504.

References

Tao, Y.Z., Henzell, R.G., Jordan, D.R., Butler, D.G., Kelly, A.M., Mcintyre, C.L., 2000. Identificationof genomic regions associated with stay-green in sorghum by testing RILs in multiple environments.Theoretical and Applied Genetics, 100, 1225-1232.Thomas and Howarth, 2000. Five ways to stay green. Journal of Experimental Botany, 51, 329-337.Thomas and Smart, 1993. Crop that stay green. Annals of Applied Biology, 123, 193-203.



HORMONAL REGULATION OF MTKNOX TRANSCRIPTION FACTORS

E. DI GIACOMO, F. SESTILI, G. FRUGIS, D. MARIOTTI

Istituto di Biologia e Biotecnologia Agraria (IBBA) - CNR, Area della Ricerca di Roma, Via SalariaKm. 29,300, 00016 Monterotondo Scalo (Roma), IT

transcription factors, Medicago truncatula, KNOX, legume, hormones

Homeobox genes are known to play a role in developmental regulation. The knotted-like homeobox(KNOX) genes fall into two classes. The class I KNOX genes like KN1, STM, and KNAT1 areexpressed in shoot meristems and loss of function mutations in different species revealed that KNOXgenes are involved in the acquisition and maintenance of meristem identity. The function of class IIKNOX genes is still undetermined.

Recent evidence suggests that KNOX transcription factors may control meristem development byregulating the balance of activities of multiple hormones. However, little is known about a possiblefeed-back regulation of KNOX gene expression by hormones.

We identified five KNOX genes in M. truncatula, a model species for legume genetics, which belongeither to class I (MtKNOX1 and MtKNOX2) or to class II (MtKNOX3, MtKNOX4, MtKNOX5).

The expression of MtKNOX genes was investigated by RT-PCR, Northern blotting analysis and in situhybridisation. All MtKNOX transcripts were found in leaves, stems, shoot apex and roots althoughclass I MtKNOX expression was very poor in leaves.

In order to investigate a possible regulation of MtKNOX by different hormones, we analysed geneexpression in response to auxin, cytokinins, gibberellins, ethylene, abscisic acid and jasmonate.



CHARACTERIZATION AND HORMONAL REGULATION OF KNOX GENESIN PEACH (PRUNUS PERSICA [[[[L.]]]] BATSCH)

E. CONDELLO*,**, D. GIANNINO**, G. TESTONE*,***, L. BRUNO**, M. B. BITONTI**, D.MARIOTTI*

*) Institute of Biology and Agricultural Biotechnology – CNR, Branch of Rome, Via Salaria km29,300, 00016 Monterotondo Scalo, Roma, Italy**) Dept. of Ecology, University of Calabria, Ponte Bucci, 87030 Arcavacata di Rende, Cosenza, Italy***) Dept. of Basic and Applied Biology, University of L’Aquila, Via Vetoio, 67010 Coppito,L’Aquila, Italy

transcription factors, Prunus persica, KNOX, hormones

The knotted-like homeobox genes (KNOX) are known to play fundamental roles in the regulation ofplant development and morphogenesis. They have been clustered into two classes according tohomeodomain similarities and gene expression patterns. The class I KNOX genes such as KN1, STM,and KNAT1 are expressed mainly in the shoot apical meristems, whereas messages of class II membersare abundant in several tissues. Moreover, complex interactions between hormones and KNOX geneshave emerged in the determination of meristem identity, organ differentiation and development. So far,the role and function of KNOX genes have been poorly investigated in peach, a model for fruit treespecies. Using tissue specific RT-PCR and based on the conservation of functional domains, weattempted to clone and characterize new members from both classes at the structural and expressionlevels. To elucidate any possible relation to hormones, KNOX transcripts were monitored in responseto the exogenous supply of distinct phyto-hormones.



BIOCHEMICAL AND MOLECULAR CHARACTERIZATION OFFLAVONOID PATHWAY IN STRAWBERRY FRUITS

J.R.M. ALMEIDA, F. MOURGUES, C. ROSATI

ENEA C. R. Trisaia, BIOTEC-GEN, S.S. 106 km 419+500, 75026 Rotondella (MT), [email protected]

polyphenols, plant pigments, Fragaria x ananassa, fruit quality, antioxidants

Flavonoids are polyphenol compounds with strong antioxidant properties, widely distributed in fruitsand accumulated at high levels in berries. A comprehensive study of the flavonoid metabolism in fruitsof cultivated strawberry (Fragaria x ananassa Duch.) genotypes was undertaken to correlate flavonoidaccumulation pattern with sequence polymorphisms and differences in gene expression of flavonoidpathway genes.

cDNA fragments of the structural genes coding for flavanone 3-hydroxylase (FHT), dihydroflavonol4-reductase (DFR), anthocyanidin synthase (ANS), flavonoid 3-O-glucosyltransferase (FGT), flavonolsynthase (FLS), a putative leucoanthocyanidin 4-reductase (LAR) and anthocyanidin 4-reductase(ANR) were amplified by PCR with specific or degenerate primers, based on sequences informationfrom Genbank. Genomic walking and/or 3’/5’RACE experiments were carried out for each gene togather information on number location and length of introns, sequence polymorphisms, and promotersequence.

Analysis of anthocyanin and flavan-3-ol+proanthocyanidin (fl3ol+PA) levels in fruits of 14 strawberryaccessions identified three “extreme” classes of high-anthocyanin, high-fl3ol+PA and low-anthocyanin/fl3ol+PA fruits. RNA was extracted from white, turning and red fruits of accessionsbelonging to each class. Real Time PCR experiments were carried out to quantify transcript levels offlavonoid genes throughout ripening. The activation of transcription at turning stage was generallyobserved for most genes and genotypes. Most interestingly, gene expression patterns were consistentwith observed chemotype, indicating a correlation between coordinated expression of flavonoid genesand the accumulation of relevant flavonoid classes.

The analysis of the promoter region of each gene is being carried out in accessions with divergentflavonoid patterns, to reveal possible polymorphisms in recognition sites of transcription factorsregulating the expression of flavonoid genes in fruits. Furthermore, full coding sequences are beingcloned in suitable expression vectors, for the biochemical characterization of the recombinant proteinsproduced in bacteria or yeast systems. The advance in knowledge from our work should point out thebest strategies for the modification of total or specific flavonoid levels in strawberry by geneticengineering.



cDNA CLONING AND EXPRESSION OF CAROTENOID BIOSYNTHETICGENES IN SUNFLOWER

M. SALVINI*,**, A. CONTI**, A. BERNINI**, M. FAMBRINI**, V. MICHELOTTI**, C.PUGLIESI**

*) Scuola Normale Superiore, Piazza Dei Cavalieri 7, I-56124 Pisa, (Italy)**) Dipartimento di Biologia delle Piante Agrarie, Sezione di Genetica, Università di Pisa, ViaMatteotti 1b, I-56124 Pisa (Italy)[email protected]

Helianthus annuus, carotenoid biosynthesis, phytoene synthase, z-carotene desaturase

The biochemistry of carotenoid biosynthesis has been well established. The fundamental steps ofcarotenoid biosynthesis are the assembly of C40 backbone, the desaturation and cyclization and finally,the xanthophyll formation. The first committed step of carotenoid synthesis, the head-to-headcondensation of geranylgeranyl diphosphate molecules to produce phytoene (colourless), is mediatedby the enzyme phytoene synthase. Membrane-localized enzymes carry out subsequent steps of thepathway leading to the coloured carotenoids (Sandmann, 2002, Physiol. Plant., 116: 431-440). Genesencoding some of the carotenogenic enzymes have been isolated in bacteria, algae, fungi and higherplants (Sandmann, 2002, Physiol. Plant., 116: 431-440; Giuliano et al. 2002, Trends Plant. Sci., 7: 427-429). However, the regulatory mechanisms that control carotenoid biosynthesis are poorly understood.In this study, we report the isolation of cDNAs corresponding to phytoene synthase (HaPsy) and z-carotene desaturase (HaZds) genes of sunflower and the analysis of their expression in relation todifferent organs and light-dependent pigment accumulation in chloroplasts.

The reconstructed full-length sequence (1916 bp) of the HaZds cDNA contains a 1761 bp CDS, 62-nucleotides of 5’-untranslated region (UTR), and 77-nucleotides of 3’-UTR (Conti et al., 2004, Plant &Cell Physiol. 45: 445-455). The predicted protein (64.9 kDa) consists of 587 amino acid residues witha putative transit sequence for plastid targeting in the N-terminal region and a typical amino oxidasedomain that includes the flavin adenosine dinucleotide (FAD) binding motif. The sunflower Zds genecomprises 14 exons and 13 introns scattered in a ca. 5.0 kb region. Also, sunflower Zds showed a highconservation of the distribution and size of the exons with rice Zds gene (Conti et al., 2004). The full-length sequence (1598 bp) of the HaPsy cDNA contains a 1242 bp CDS, 172-nucleotides of 5’-untranslated region (UTR), and 170-nucleotides of 3’-UTR (Salvini et al., 2004, J. Plant Physiol., inpress). The predicted protein (46.8 kDa) displayed a sequence of 414 amino acid residues with aputative transit sequence for plastid targeting in the N-terminal region. The phylogenetic analysisdemonstrated that both HaPsy and HaZds were clustered to marigold (Tagetes erecta) Psy and Zdsgenes, for which showed an overall amino acidic identity of 93.7 % and 96.6%, respectively.

Both HaZds and HaPsy were highly expressed in cotyledons and leaves; by contrast, their transcriptlevels were comparatively lower in both stems and roots. In addition, HaZds and HaPsy transcriptlevels were influenced by leaf expansion, which suggested that their expression are regulated duringthe process of leaf development. The light-dependent enhanced carotenoid production in sunflower

chloroplasts is concurrent with an increase of both HaZds and HaPsy transcript accumulation. Theseresults stand in contrast to those obtained in pepper (Simkin et al. 2000, J. Agric. Food Chem., 48:4676-4680) and in tomato (Simkin et al. 2003, Z. Naturforsch., 58c: 371-380). A possible explanationfor these species to species differences may result from the very dissimilar experimental conditionsused in these studies. Moreover, although the control of gene expression at the transcriptional level is akey regulatory mechanism, one or more post-transcriptionally control points must be decisive in theregulation of carotenogenesis; therefore one can suggest that some of these mechanisms are prevalentover others in some species.



A PLANT-DERIVED SIGNAL PEPTIDE DRIVES THE HPV16 E7ONCOPROTEIN TO THE SECRETORY PATHWAY OF MAMMALIANCELLS- IMPLICATIONS FOR THE IMPROVEMENT OF A DNA VACCINE

S. MASSA*, P. SIMEONE**, A. VENUTI**, R. FRANCONI*

*) ENEA, BIOTEC GEN, CR Casaccia, Via Anguillarese 301, 00060 Rome, Italy**) Istituti Fisioterapici Ospitalieri (IFO) - Istituto Regina Elena per lo Studio e la Cura dei Tumori,Via delle Messi d'Oro 156, 00158 Rome, Italy

HPV16, E7 oncoprotein, DNA vaccine, secretion

The “high risk” types of Human papillomaviruses (HPVs 16, 18) generally cause lesions that mayprogress to cervical cancers. The HPV16-E7 early protein has been identified as the majortransforming protein, therefore representing a prime target for a selective anti-cancer therapy.

A number of clinical trials are testing the potential of therapeutic vaccination with tumor antigen-encoding DNA vaccines. A drawback of DNA vaccines is their limited potency, due to their intrinsiclack of an in vivo amplifying and spreading ability and to the sub-optimal intracellular processing andpresentation of tumor antigens (expecially true for HPV16-E7).

Plant-derived sequences may offer an alternative approach to enhance HPV16-E7 protein expressionand "visibility" to the immune system, possibly increasing the efficacy of DNA vaccination againstHPV-associated tumors.

The ectopic production of HPV16-E7 protein in Nicotiana benthamiana plants (and, consequently, theefficacy of a plant-based vaccine) was shown to be enhanced when it was expressed in the secretorypathway by means of the signal peptide coding sequence of the polygalacturonase inhibiting proteinfrom Phaseolus vulgaris (PGIPss-E7, unpublished results).

We wanted to investigate if this plant signal peptide would be able to play the same role in amammalian expression system in vitro and amplify the immune response due to an "accumulation"effect in vivo.

To verify our hypothesis, a novel N-terminal gene fusion between PGIPss and a mutagenized,harmless version of the tumor antigen (PGIPss-E7GGG) was cloned into a mammalian expressionvector to assay its expression and eventual secretion by transfection of HEK 293 and COS-7 cell lines.To further increase the accumulation of the mutated E7 into the cell, a construct bearing a C-terminalER retention signal, KDEL, was also produced.

Preliminary data indicate an enhanced expression with the PGIPss-E7GGG plasmid in the above-mentioned mammalian expression systems and show that the E7GGG protein can be also detected inthe conditioned medium. These DNA constructs are being tested to assess whether they induce anykind of immune response and enhanced protection from cancer in vaccinated mice challenged with a

tumorigenic dose of E7-expressing tumor. Data on biological effects will be presented.



DETECTION OF CIS-ACTING REGULATORY VARIATIONS THATAFFECTS LIGNINE GENES IN EUCALYPTUS

A. MISSIAGGIA*, S. RADOVIC***, D. GRATTAPAGLIA**, M. MORGANTE***

*) Departamento de Genetica e Melhoramento de Plantas, ESALQ-USP, São Paulo, Brasil**) EMBRAPA-Recursos Genéticos e Biotecnologia, Brasilia, Brazil***) Dipartimento di Scienze Agrarie ed Ambientali, Università degli Studi di Udine, Via delleScienze 208, 33100 Udine, Italy

Eucalyptus, lignine, gene expression, SNPs.

The study of candidate genes that putatively affect trait expression, has been an important strategy tocharacterize a QTL and understand the biochemical or developmental pathways affecting the trait ofinterest. Biosynthetic (structural) genes of the lignification pathway constitute a unique opportunity fortesting the candidate gene approach and characterizing the QTL involved in the control of woodquality in woody plants.

Functional polymorphism in genes can be classified as non-synonymous and synonymous, reflectingnucleotide variants with and without effect on phenotype at the level of protein primary structure, andcan be recognized directly from DNA sequence. Variation in non-coding cis-regulatory DNAsequences, which affects gene expression levels, has also been proposed as a major component of thegenetic basis for phenotypic evolution, but it is mostly unknown.

A method to detect changes in transcript levels, due to cis acting sequence differences, without thenecessity to recognize specific regulatory variants that can be hundreds or even thousands of basesupstream from the transcription unit, has been developed in Eucalyptus, based on studies aboutmethods for measurement of allelic levels of gene expression in humans and mouse (Yan et al.,Science 2002; Cowles et al., Nature Genetics 2002).

The assay involves the study of two alleles of a gene under identical conditions and the comparison ofthe transcript expression level associated with each. To distinguish between the transcripts derivedfrom each of the two parental alleles we have used SNP markers (Single Nucleotide Polymorphisms)in the transcript itself. The assay involves RT-PCR amplification in the region surrounding the SNPmarker, followed by a SBE (Single Base Extension) using a primer adjacent to the variant base in thepresence of fluorescently labeled nucleotides and detection on a capillary DNA sequencer. The ratio ofthe levels of the two alleles is inferred by comparison with known mixtures of both parental genomicDNA used as reference standard, which after PCR amplification, are subjected to the same SNP assay.

Eucalyptus is an outcrossing species with a high level of heterozygosity so that a high degree ofnucleotide diversity affecting both coding sequence and expression level of genes could be detected.For this study we set a survey of polymorphisms among a list of candidate genes involved in thelignine biosynthetic pathway including PAL (phenylalanine ammonia lyase), COMT (Caffeate/5hydrolxyferulate o-Methyltransferase), CCR (Cinnamoyl CoA Redutase), CAD (Cinnamyl Alcoholdehydrogenase). We analysed the selected loci in the parentals and in 16 offsprings from the a cross

between Eucalyptus grandis X Eucalyptus urophylla. By using the SNP markers detected in thetranscribed regions we could distinguish between transcripts derived from each of the two parentalalleles in order to reveal abundance variation in the mRNA related to each single allele.


Poster Abstract – C.59

COLD STRESS AND CIS-ACTING REGULATORY VARIATION IN MAIZE

S. RADOVIC*, G. PEA**, M.E. PE'**, M. MORGANTE*

*) Dipartimento di Produzione Vegetale e Tecnologie Agrarie, Università degli Studi di Udine, Viadelle Scienze 208, 33100 Udine, Italy**) Dipartimento di Scienze Biomolecolari e Biotecnologie, Università di Milano, Via Celoria 126,20133 Milano, Italy

maize, gene expression, SNPs, cold stress

Functional polymorphism in genes can be classified as coding variation, altering the amino-acidsequence of the encoded protein, or regulatory variation affecting the level or the pattern of expressionof the gene. While the frequencies and consequences of coding polymorphism can be recognizeddirectly from the DNA sequence, the extent to which variations in non-coding cis-regulatory DNAalters gene expression in populations is mostly unknown. However, it has been suggested thatregulatory variations are important in modulating gene function since alterations in gene regulation areproposed to influence disease susceptibility, and to be the primary substrate for the evolution of thespecies. If this hypothesis is correct, it implies that cis-acting regulatory variation is a commonphenomenon.

Our recent studies of allele-specific expression among non-imprinted genes coming from two maizeinbred lines (B73 and H99) in the F1 hybrid lines (B73XH99 and H99XB73) confirmed the postulatedpresumption, since more than 58% of the genes tested showed greater than 1.5 fold differences inexpression among the alleles, with no difference among reciprocal hybrids. In addition, we observedsignificant variations in allelic expression ratios across different tissues.

We now examined the influence of abiotic stress on allele-specific expression in the two reciprocalhybrids, B73XH99 and H99XB73, using Single nucleotide polymorphism (SNP) assay developed formaize to measure the relative expression of each allele of a gene in a heterozygous individual. Themethod utilizes SNP markers in the transcript itself to distinguish between the transcripts derived fromeach of the two parental alleles.

RNA samples were isolated separately from the aerial part of seedlings and their roots after 24 and 72hours cold treatment, as well as from their respective controls. Preliminary results reveal that coldinduces significant variations in expression ratios for the alleles that initially showed no disparity inexpression, and it enhanced alteration in the relative expression ratios for those alleles that alreadyshowed differential expression patterns. In addition, transcriptional changes induced by cold variedbetween the aerial parts of the seedlings and their roots. We also analyzed the nucleotide sequencepolymorphisms in the 5' upstream regulatory region of some of the genes and tried to identify anyassociation between polymorphisms and differences in allelic expression.

Our findings suggest that cis-acting regulatory variation in addition to being a widespreadphenomenon in maize, is also relevant to stress response, with different alleles/haplotypes respondingdifferentially to stress in different parts of the plant. The heterozygous state found in hybrids for many

genes could therefore represent a buffering mechanism to improve stress tolerance.

Results also support the use of SNP assay as a valid quantitative method that indirectly scans for theinfluence of cis-acting effects on gene expression which helps us not only to appreciate the extent offunctionally important regulatory variations but also to focus on candidate haplotypes that havedifferences in expression for detailed molecular characterization of specific polymorphisms.



ISOLATION OF GENES CONTROLLING MEGASPOROGENESIS INMEDICAGO SATIVA L.

S. CAPOMACCIO, D. ROSELLINI, F. VERONESI

Dipartimento di Biologia Vegetale e Biotecnologie Agroambientali, Facoltà di Agraria, Università diPerugia, Borgo XX Giugno 74, 06121 Perugia

alfalfa, cDNA-AFLP, female sterility, gene expression, RT-PCR

Reproduction control is crucial for both seed production and plant breeding practice. The goal of thisstudy is to isolate and characterize genes involved in megasporogenesis and female sterility in alfalfa.A female sterile mutant was previously cytologically and phenotipically characterized (Rosellini et al.2003, Sexual Plant Reprod. 15:271-279), revealing a female-specific arrest of sporogenesis associatedwith ectopic, massive callose deposition within the nucellus.

Several sporophytic mutations influencing both male and female meiosis have been described, butthere are only two reports of mutations determining the block of female meiosis only (Siddiqi et al.2000, Development 127: 197-207; Motamayor et al. 2000, Sex. Plant Reprod. 12: 209-218). Thealfalfa mutation we are studying determines female-specific megasporogenesis abnormalities with amechanism that appears different from those described to date.

The cDNA-AFLP technique was employed to isolate genes differentially expressed between wild typeand female-sterile full-sib alfalfa plants. Ninety six ESTs were generated, and most of them published(GenBank from CB165074 to CB165159). BLAST analysis have revealed similarities with genesinvolved in the cell cycle, development and callose metabolism.

Four clones were selected to proceed with further studies: CB165076 similar to Arabidopsis thalianaeucaryotic initiation translation factor eIF4G III; CB165091 similar to a soybean 1,3 beta-glucanase;CB165105 similar to the A. thaliana transcription factor SCARECROW gene regulator; andCB165125 similar to an A. thaliana MAPKKK.

Full length mRNA sequences were obtained using the Rapid Amplification of cDNA Ends (RACE)technique. RT-PCR was performed to confirm differential expression during flower development usingthe 18s rRNA as control gene. In situ hybridization experiments are carried out to spatiallycharacterize the expression of the selected genes.



CHARACTERISATION OF THE PHENYLALANINE AMMONIA-LYASEGENES IN ARTICHOKE

A. DE PAOLIS*, D. PIGNONE**, G. SONNANTE**

*) Institute of Sciences of Food Production - CNR, Sect. Lecce**) Institute of Plant Genetics, CNR, Bari

phenolic compounds, PAL, Cynara cardunculus, genomic library

The formation of free radicals is connected to the normal aerobic cell methabolism. The consumptionof oxygen during the cell growth determines the formation of several oxygen free radicals, whoseinteraction with lipid molecules produces new free radicals: hydroperoxides and several peroxides.This group of radicals can interact with biological systems determining citotoxicity.

It has been demonstrated that artichoke extracts are rich in phenolic compounds which protect cellmembranes, show antibacterial, antioxidative, anti-HIV, bile expelling, hepatoprotective, urinative,and choleretic activities as well as the ability to inhibit cholesterol biosynthesis and LDL oxidation.

The first enzymatic step in the pathway of phenilpropanoid biosynthesis is the deamination ofphenylalanine by means of the enzyme phenylalanine ammonia-lyase (PAL). This enzyme has beencharacterised in several plant species and it is usually coded by more than one gene.

In order to study the organisation of the gene/genes coding for PAL in artichoke, we have used primersdesigned on the basis of the information retrieved from the sequence databases.

Primers were designed in two exons flanking the single intron included in the PAL gene. Afteramplification and gel separation, two DNA fragments were identified of about 550 and 1000 bprespectively. The two fragments were isolated and sequenced and revealed to share a high homologywith the PAL genes from other Asteraceae, the family which the artichoke belongs to.

The DNA fragments obtained are being used as a molecular probe to screen a genomic library in orderto isolate the PAL gene/genes in artichoke and characterise its/their organisation.

The isolation of mRNA from various tissues and the synthesis of cDNA will allow to follow thepatterns of expression of PAL genes in artichoke.



MOLECULAR CLONING AND COMPARISON OF DIHYDROFLAVONOL 4-REDUCTASE GENE SEQUENCES BETWEEN COLOURED ANDUNCOLOURED CULTIVARS OF SWEET ORANGE (CITRUS SINENSIS L.)

A. R. LO PIERO, I. PUGLISI, A. CONSOLI, G. PETRONE

Università di Catania, Dipartimento di Scienze Agronomiche, Agrochimiche e delle ProduzioniAnimali, Via S. Sofia 98, 95123 Catania

dihydroflavonol 4-reductase, promoter region, blond and blood orange cultivars

Anthocyanins are the main pigments in flowers and fruits where they serve as visual signals that attractinsects and animals for pollination and seed dispersal. They are also present in autumn foliage andrapidly developing shoots of tropical trees accomplishing a photoprotective role. The red-colour ofseveral orange cultivars (Tarocco, Moro and Sanguinello) is mostly caused by the presence, both in thefruit flesh and rind, of anthocyanin pigments not usually found in Citrus genus and, more specifically,absent in blonde orange cultivars such as Navel, Ovale, and Valencia. Anthocyanin biosynthesis hasbeen extensively investigated resulting in the elucidation of the biosynthetic pathway in which thevarious anthocyanin pigments are synthesised from phenylalanine. Recently, genes active in the“early” steps of anthocyanins biosynthesis encoding chalcone synthase (CHS), chalcone isomerase(CHI) and flavanone 3-hydroxylase (F3OH), have been cloned from Valencia orange young seedscDNA library and their expression was monitored during satsuma mandarin fruits development, bynature uncoloured (Moriguchi et al., Physiol. Plantarum, 2001, 111:66). In our previous works thepartial cDNA clones coding for CHS, CHI, F3OH, DFR, ANS and UFGT were isolated from Taroccoorange flesh cDNA by PCR amplification (Lo Piero et al., Proc. of the XLV It. Soc. of AgriculturalGenetics, 2001); the expression of the “late” genes dfr and ans was also monitored either in colouredor uncoloured orange cultivars by real-time RT-PCR (Russo et al., Proc. of the XLVII It. Soc. ofAgricultural Genetics, 2003). Those preliminary data showed that a strong reduction in DFRexpression occurs in the blonde cultivars suggesting that the phenotypic change from coloured andcolourless might be correlate either with mutation in a regulatory gene controlling the expression of dfror with different features of the promoter regions between the coloured and colourless orangecultivars. It has been found that the start point for the control of the anthocyanins pathway is differentin the most studied species. In maize the regulatory genes C1 and R seemed to regulate thetranscription of all the structural genes encoding the enzymes directly involved in the whole metabolicreactions, whereas in snapdragon and petunia the key regulatory points appeared to be localized downin the pathway, respectively at F3OH and DFR level. In this study we isolated the complete codingsequence of Citrus sinensis DFR (Genbank n°AY519363) as well as the promoter sequences of the dfrgenes from pigmented (Tarocco) and unpigmented (Navel) orange cultivars. The deduced aminoacidssequences were highly similar to those reported for DFRs of Vitis, Malus and Arabidopsis. In the 5’upstream regions of both Navel and Tarocco DFRs several elements to which transcription factors canbind were found. In particular, three core elements for Dof transcription factor (one zing finger) whichhas been reported to be involved in tissue specificity and light regulation ( Yanagisawa and Schmidt,The Plant Journal, 1999, 17:209), a GAmyb recognition element and a consensus sequence for themyb-like P transcription factor that in maize activates chs, chi and dfr genes leading to the production

of phlobaphene. As there were no significative differences in promoter sequences between Navel andTarocco dfr, a mutation in a regulatory gene controlling the expression of DFR is thought to beinvolved in the lack of anthocyanins biosynthesis in blonde orange cultivars.



ISOLATION OF AN ACYLTRANSFERASE SEQUENCE PUTATIVELYINVOLVED IN CYNARIN BIOSYNTHESIS IN CYNARA CARDUNCULUS L.

C. COMINO*, A. ACQUADRO*, E. PORTIS*, A. HEHN**, F. BOURGAUD**, S. LANTERI*

*) Di.Va.P.R.A. Plant Genetics and Breeding, University of Turin, Via L. da Vinci 44, I-10095Grugliasco (Turin), [email protected]**) UMR 1121 INPL-INRA Agronomie Environnement 2 avenue de la forêt de Haye, 54505Vandoeuvre-lès-Nancy, France

cynarin, Cynara cardunculus, acyltransferase isolation

Cynara cardunculus is a diploid (2n=2x=34) species which includes globe artichoke (var. scolymusL.), cultivated cardoon (var. altilis DC) and their progenitor wild cardoon [var. sylvestris (Lamk)Fiori].

Cynara cardunculus is source of biopharmaceuticals and its leaf extracts have been widely used inherbal medicine as hepatoprotectors and choleretics since ancient times. The chemical components ofthe leaves have been found rich in compounds originating from the metabolism of phenylpropanoidswhich (i) protect proteins, lipids and DNA from oxidative damage caused by free radicals, (ii) inhibitcholesterol biosynthesis and contribute to the prevention of arteriosclerosis and other vasculardisorders, (iii) inhibit HIV integrase, a key player in HIV replication and its insertion into host DNA,and (iv) possess antibacterial activity.

The major phenolic compounds in artichoke extracts are di-caffeoylquinic acids (e.g. cynarin) whichare present mainly in Cynara species, and its precursor chlorogenic acid, one of the most widespreadsoluble phenolic compound in the plant kingdom. While chlorogenic acid has practically no marketvalue, cynarin is a molecule of large pharmaceutical interest because it exhibits a strong antioxidantactivity.

In extracts from tobacco stem a protein: hydroxycinnamoyl-CoA:shikimate/quinatehydroxycinnamoyltransferase (HCT) has been purified, which controls the biosynthesis of chlorogenicacid. The HCT: (i) catalyses the esterification of caffeic acid with the 3-hydroxyl group of quinic acidin order to originate chlorogenic acid, (ii) catalyzes the esterification of p-coumaroyl-CoA with quinicacid in order to generate p-coumaroyl quinate. Furthermore, in tobacco, the cDNA responsible forHCT production has been isolated and sequenced.

Artichoke presumably possesses an extra HCT enzyme that converts chlorogenic acid into cynarin andthe aim of this research was to isolate the corresponding cDNA. mRNAs were extracted from globeartichoke leaves and the cDNAs generated by reverse transcription. Degenerate CODEHOP primerswere designed on conserved region of acyltransferase protein in order to amplify the HCT cDNA byPCR. The resulting DNA fragments were resolved by agarose gel electrophoresis and a band of 700 bpisolated, cloned into a plasmid and sequenced.

A translated database search (Blast x) revealed high similarity (79% identity and 86% homology) withthe tobacco HCT (cDNA: AJ507825) and global alignment (CLUSTAL W) revealed the artichoke HCT-like sequence clustering with one of the four main acyltransferase groups (i.e. anthranilate N-hydroxycinnamoyl/benzoyltransferase). Further work will try to isolate full length HCT-like cDNAsby using the isolated sequence as a probe to screen a cDNA library. After successful isolation, thecDNA candidates will be heterologously expressed in yeast or in bacteria, with the objective tocharacterize the cDNA coding for the enzyme converting chlorogenic acid into cynarin.



A GENE CODING FOR CYTOSOLIC GLUTAMMINE SYNTHETASE INLENTIL

G. SONNANTE, M. ORLANDO, D. PIGNONE

Istituto di Genetica Vegetale – CNR, Via Amendola 165/A, 70126 Bari

sequence variation, GS, twin genes, genepool, Lens culinaris

Glutammine synthetase (GS) is an enzyme which catalyses the first step of the fixation of ammoniuminto organic molecules. Its isoenzymes are coded by a small gene family, whose members aredifferently expressed on the basis of the plant age, organ, tissue and nitrogen nutrition. Two forms ofthe GS are known in plants: the cytosolic GS, usually coded by different nuclear genes, and thechloroplast GS coded by a single nuclear gene.

In leguminous plants, cytosolic GS is abundant in roots and its function is related to fixation ofammonia into amino acids.

With the aim at isolating and characterising GS genes in lentil, a preliminary study was started takingthe pea GS as a reference.

In pea, three classes of GS have been identified differing for function, localisation and chromosomelocation. The GS3 class includes two twin genes (GS3A and GS3B) coding for cytosolic GS. Thesetwo genes are very similar, but they differ especially in the length and sequence of introns 5 and 10,located in different positions of the genes.

In order to verify the presence of corresponding twin genes in lentil, and to evaluate the variationexisting in the genepool of lentil, primers were designed in the coding regions flanking the two introns.

After amplification and gel separation, a single band was observed for each primer combination used.These results may indicate that, differently from pea, the gene coding for GS3-like protein is present ina single form in lentil.

In order to assess the genetic structure of the GS3 gene, similar amplifications were carried out in otherspecies of Lens and in the related species Lathyrus sativus. The analysis of the sequences obtained inthese species allowed to identify intron length variation, in particular in Lens ervoides and Lath.sativus. The present data are also useful to assess relationships within the genus Lens.



CHARACTERIZATION OF CHLORELLA PYRENOIDOSA L-ASCORBIC ACIDACCUMULATING MUTANTS: IDENTIFICATION OF AN ENHANCEDBIOSYNTHETIC ENZYME ACTIVITY AND CLONING OF THE PUTATIVEGENE FROM ARABIDOPSIS THALIANA INTO TOBACCO CELLS

A. DI MATTEO*, R. HANCOCK**, H.A. ROSS**, L. FRUSCIANTE*, R. VIOLA**

*) Department of Soil, Plant and Environmental Sciences, University of Naples “Federico II”, ViaUniversità 100, 80055 Portici (Italy)[email protected]**) Scottish Crop Research Institute, Dundee, Scottland (UK)

ascorbic acid, GDP-L-galactose pyrophosphatase, Chlorella pyrenoidosa, Arabidopsis thaliana, BY-2

Plants are the main source of vitamin C in the human diet. In plants, L-ascorbic acid (AsA) is essentialfor photosynthetic activity via the detoxification of superoxide anions and hydrogen peroxide inchloroplasts in the absence of catalase. AsA is also crucially involved in the regeneration of α-tocopherol and zeaxanthin and the pH-mediated modulation of PS II activity. Although the major AsAbiosynthetic pathway in plant was elucidated, yet relatively little is known about the control of AsAbiosynthesis in plant.

We have undertaken a comparative study of L-ascorbic acid (AsA) biosynthesis in wild-type Chlorellapyrenoidosa and two mutant strains containing enhanced AsA content. Both mutant strainssynthesized AsA more efficiently from the distant precursors D-glucose or D-mannose whilst enhancedbiosynthesis was not observed from the immediate precursors L-galactose or L-galactono-1,4-lactone.In vitro assay of individual biosynthetic steps revealed that only one enzyme exhibited enhancedactivity in both mutant strains, a putative GDP-L-galactose pyrophosphatase. Both mutant strains werefound to contain higher concentrations of free L-galactose than the wild-type strain, suggesting thatGDP-L-galactose pyrophophatase and/or L-galactose-1-phosphate phosphatase activity were alsoenhanced in vivo.

In order to simplify cloning of the pyrophosphatase gene, Arabidopsis thaliana was used as modelplant. The activity was partially purified and a number of proteins were N-terminal sequenced byEdman degradation, one of which corresponded to a nucleotide pyrophosphatase-like protein.Nucleotide pyrophosphatase protein concentrations correlated well with enzyme activity as judged byCoomassie staining on polyacrylamide gels. Genomic sequence of the putative nucleotidepyrophosphatase has been cloned from Arabidopsis thaliana genome. Interestingly, its over-expressioninto BY-2 tobacco cells leaded approximately to both a three-fold increase of the correspondingenzyme activity and AsA content. Current efforts are focused on in vitro characterization of therecombinant enzyme and in vivo analysis of protein function.

Documents

PARALLEL SESSION C – GENOMICS AND PROTEOMICS · PARALLEL SESSION C – GENOMICS AND PROTEOMICS ... Laboratoire de Biologie Végétale et Microbiologie, ... (AFLP-TP) analisys. Roots