Fusarium head blight (FHB) is a horribly destructive fungal diseases of barley. FHB is caused by a species complex of Fusaria, of which Fusarium graminearum Schwabe is the main causal species of FHB epidemics in Canada. Field surveys show that two or more Fusarium species often co-exist within the same field or grain sample and F. poae is reported as another important species in barley in Ontario and Quebec. The aim of this study was to observe the pathogenicity of F. graminearum, F. poae and a co-inoculation of both species causing FHB in barley. Two susceptible barley cultivars were spray-inoculated at 10 to 14 days after heading. Phenotypic disease severity was rated on a scale of 0-9 at 4, 7, 14, 21, and 28 days after inoculation. There was a significant difference in FHB severity between F. graminearum and F. poae, where infection with F. graminearum produced more severe disease ratings. F. poae generated lower disease ratings and was not statistically different from the control. When heads were co-inoculated with both Fusarium species, the resulting FHB severity was unchanged 54relative to heads inoculated with F. graminearum only. The ratio of F. graminearum to F. poae genomic DNA was also no different than when heads were inoculated with F. graminearum alone, as quantified with ddPCR using markers specific to each species. Metabolomic analysis of sample extracts corroborated our other findings where F. graminearum appeared to outcompete F. poae in barley also at the mycotoxin level. Barley cultivar did not have a significant effect on visual FHB disease ratings and fungal DNA detection, however there were some chemical differences between cultivars in response to challenge by both F. graminearum and F. poae.

La caractérisation phénotypique de l’espèce Torulaspora delbrueckii en conditions œnologiques, à partir de l’étude d’un grand nombre de souches, a permis de mettre en évidence une grande variabilité au sein de cette espèce. En effet, les souches de T.delbrueckii présentent des différences au niveau des durées de phase de latence et de fermentation, des capacités biotiques mais aussi des productions d’éthanol (maximum 12% vol.). Cette variabilité se retrouve également pour la production d’acidité volatile, de glycérol et de certains arômes. Ce travail confirme les faibles productions d’acidité volatile et de glycérol de cette espèce et met en évidence une réponse au stress osmotique différente de celle de l’espèce Saccharomyces cerevisiae. Au final, l’espèce T. delbrueckii présente une grande « pureté » de fermentation et produit peu de composés indésirables comme le sulfure d’hydrogène, les phénols volatils, l’acetoïne, l’acétaldéhyde et le diacétyle. La réalisation de co-inoculations T. delbrueckii / S. cerevisiae sur moûts liquoreux, permet une réduction systématique de l’acidité volatile des vins, en comparaison à une fermentation pure de l’espèce S. cerevisiae, quelque soit la souche de T. delbrueckii utilisée. De plus, la souche T. delbrueckii OXT1 1//2 a permis de complexifier la composition aromatique d’un moût sec issu du cépage Sauvignon blanc (esters fermentaires +25%, phényl-2-éthanol +51% et thiols volatils +31%). Enfin, la mise au point d’un fermenteur à double compartiment, avec une séparation physique des levures tout en conservant l’homogénéité du milieu de culture, a permis d’aborder l’étude des interactions entre ces 2 espèces. Des inhibitions de type « cell-cell contact » ont ainsi été mises en évidence. / The phenotypic characterization of Torulaspora delbrueckii species in enological conditions, using a large number of strains, showed up a wide variability within this species. Indeed, the strains of T. delbrueckii present different fermentative capacities for lag phase and fermentation durations, biotic capacities and ethanol production (up to 12 % vol.). This phenotypic variability is also obtained for the production of volatile acidity, glycerol and certain aroma compounds. This work confirms the low productions of volatile acidity and glycerol of this species and reveals an osmotic stress response distinct from that of Saccharomyces cerevisiae. Finally, the T. delbrueckii species presents a high fermentation purity and produces low levels of undesirable compounds like hydrogen sulfide, volatile phenols, acetoin, acetaldehyde and diacetyl. The realization of co-inoculations T. delbrueckii / S. cerevisiae on high sugar must, allows a systematic reduction of the volatile acidity of wines, in comparison to a pure fermentation of S. cerevisiae. Moreover, co-inoculation with the T. delbrueckii strain OXT1 1 // 2 allowed the increase of aromatic complexity of a Sauvignon blanc dry wine (esters + 25%, 2-phenylethanol +51 % and volatile thiols + 31 %). At last, the development of a fermentor with double compartment allowing a physical separation of yeasts and preserving the homogeneity of culture medium, allowed us to approach the study of yeast interactions. Inhibitions of type “cell-cell contact” have been demonstrated.

Les infections multiples sont courantes dans la nature et sont considérées comme très importantes dans l'évolution des caractéristiques biologiques des parasites. Théoriquement, les infections multiples devraient entraîner une évolution de la virulence à la fois comme stratégie adaptative et comme stratégie plastique. Dans cette thèse, j'utilise Hyaloperonospora arabidopsidis, un parasite naturel d'Arabidopsis thaliana, qui s'est avéré pratique pour des études en écologie évolutive, pour étudier: i) les infections multiples consécutives à la co-inoculation et à l'inoculation séquentielle; succès de l'infection et succès de transmission d’une souche individuelle (génotypage par PCR) et des phénotypes d'infection, y compris virulence entre inoculation unique et mixte, iii) effet du délai d'inoculation et d'ordre des souches inoculées sur les phénotypes d'infection et le succès de l'infection individuelle. Ici, j'ai trouvé une fréquence plus élevée de co-infection à la suite de l'inoculation séquentielle que à la suite de la co-inoculation des mêmes combinaisons de souches. L'inoculation mixte de certaines combinaisons de souches a entraîné une modification des phénotypes d'infection, souvent avec un succès d'infection plus faible chez certaines souches à la suite des inoculations en mélanges qu’en inoculation simple. Ce résultat implique une interférence entre les souches dans l'inoculum mixte. La virulence globale de l'infection après l'inoculation mixte n'était pas toujours supérieure à celle de l'infection à souche simple. De plus, les souches uniques utilisées dans ces expériences ne différaient pas toujours les unes des autres en termes de virulence. Le seul test d'un mélange de génotypes à trois souches a provoqué une virulence globale plus élevée que les trois infections à souche unique respectives. Une plus grande virulence globale dans ce cas pourrait être due à la plasticité des souches parasitaires inoculées, à la réponse à la présence d'autres souches dans l'inoculum mixte ou à l'effet de multiples souches supprimant le système de défense de l'hôte. Lorsque les souches ont été inoculées de manière séquentielle et non ensemble, le succès de l'infection de souches individuelles différait entre les différents ordres d'inoculation, ce qui pourrait être dû à des effets indirects via le système de défense de l'hôte. En résumé, l'inoculation séquentielle a semblé réduire l'interférence entre les souches parasitaires, avec un effet de décalage temporel et d'ordre de la souche inoculée sur le succès de l'infection de souches individuelles. Une interférence dans un inoculum mixte peut générer différents succès d'infection et phénotypes d'infection à partir des inoculations individuelles respectives. J'ai trouvé un cas évident de virulence globale plus élevée dans les infections causées par des inoculations mixtes. Par conséquent, une virulence globale plus élevée peut se produire malgré le fait que nous ne trouvions pas de meilleures performances de génotypes plus virulents dans des infections à la suite d'inoculations mixtes. Ainsi, ces résultats ne permettent pas de prédire l’évolution de la virulence supérieure parmi ces combinaisons de souches testées. Cependant, la plasticité des phénotypes des souches inoculées dans l'inoculum mixte a généré une virulence globale de l'infection plus élevée. Ces résultats peuvent aider à comprendre comment les génotypes de parasites répondent aux infections mixtes. / Multiple infections are common in nature, and are considered very important in the evolution of parasite life-history traits. Theoretically, multiple infections should lead to evolution of higher levels of virulence both as an adaptive and as a plastic strategy. In this thesis I use Hyaloperonospora arabidopsidis, a natural parasite of Arabidopsis thaliana, which has proven a useful tool for unlocking some evolutionary ecology questions, to investigate: i) multiple infections following co-inoculation and sequential inoculation, ii) number of infected plants, infection success and transmission success of individual strain (genotyping via PCR), and infection phenotypes including virulence between after single- and mixed inoculation, iii) effect of time lag of inoculation and order of inoculated strain on infection phenotypes and individual strain infection success. Here I found that sequential inoculation contributed higher frequency of co-infection than co-inoculation of the same strain combinations. Mixed inoculum of some strain combinations led to modification of overall infection phenotypes, often with poorer infection success of individual strains compared with that of the more infectious strains. This result implies interference between strains in mixed inoculum. Overall virulence of infection after mixed inoculation was not always higher than that of single strain infection. Furthermore the single strains used in these experiments did not always differ from each other in virulence. The one test of a three-strain mixture of genotypes caused higher overall virulence than the three respective single strain infections. Higher overall virulence in this case might be caused by plasticity of inoculated parasite strains reponse to the presence of other strains in mixed inoculum or an effect of multiple strains suppressing the host defence system. When strains were inoculated sequentially instead of together, infection success of individual strains differed between different orders of inoculation, which could be due to indirect effects via the host defence system. In summary, sequential inoculation seemed to reduce interference between parasite strains, with effect of time lag and order of inoculated strain on infection success of individual strains. Interference in mixed inoculum can generate different infection successs and infection phenotypes from the respective single inoculations. I found one clear case of higher overall virulence in infections caused by mixed inoculations. Thus higher overall virulence can occur despite our not finding higher performance of more virulent genotypes from infections following mixed inoculations. Thus these finding do not predict the evolution of higher virulence among these strain combinations tested. However, plasticity of phenotypes of inoculated strains in mixed inoculum did generate higher overall virulence of infection. These findings can help to understand how the parasite genotypes respond to in mixed infections.

Submitted by Jorge Silva (jorgelmsilva@ufrrj.br) on 2017-02-08T16:51:22Z No. of bitstreams: 1 2016 - Rennan do Amaral Bastos.pdf: 2824478 bytes, checksum: 5a974a7444eaef272210a147add547a0 (MD5) / Made available in DSpace on 2017-02-08T16:51:22Z (GMT). No. of bitstreams: 1 2016 - Rennan do Amaral Bastos.pdf: 2824478 bytes, checksum: 5a974a7444eaef272210a147add547a0 (MD5) Previous issue date: 2016-02-04 / The objective of this work was to evaluate the production of biomass and grains in common bean (Phaseolus vulgaris) in co-inoculation with rhizobia and plant growth promoting bacteria. The work was divided into three chapters. The first evaluated the compatibility between rhizobia and plant growth promoting bacteria and their effects on plant growth. In Petri dishes assay, Azospirillum amazonense and A. brasilense were co-inoculated with Rhizobium tropici at three application levels, and no antagonism was detected. The experiment in Leonard jars had a 5x3+3 factorial with eight replications, consisting of five bacteria (Bradyrhizobium diazoefficiens, B. elkanii, A. amazonense, Azospirillum brasilense and Escherichia coli) and three application levels (104, 106 and 108 CFU mL-1), all co-inoculated with R. tropici plus three additional treatments (control, nitrogen, and inoculated with R. tropici). Co-inoculation increased nodulation, N concentration and content in the shoot. The experiment in pots with soil had 3x2x2+4 factorial with four replications, consisting of three growth promoters (B. diazoefficiens, B. elkanii and A. brasilense), two application levels (104 and 106 CFU mL-1), two N sources (R. tropici and mineral N), and four additional control treatments (absolute control, nitrogen, inoculated with rhizobia and co-inoculated with AZOTOTAL?). Inoculation with A. brasilense associated with N fertilization increased the production of leaves and shoots. The second chapter aimed to evaluate the growth and production of common bean under co-inoculation. An experiment in pots with soil in a factorial 2x3+4 with 8 repetitions were performed by combining the inoculation of R. tropici with A. brasilense or B. diazoefficiens in three application levels (104, 106 and 108 CFU mL-1) plus the control treatments (absolute, nitrogen, inoculated with rhizobia and co-inoculated with AZOTOTAL?). Four replications were harvested at 45 days after sowing, and four at grain maturity. Co-inoculated plants showed shoot mass, number of pods, shoot N content, grain yield, number of seeds and pods per plant higher than the control inoculated with rhizobia. The leaf chlorophyll Falker index was similar in co-inoculated treatments and under mineral N fertilization, but lower than the exclusive inoculation R. tropici. In the third chapter it was evaluated the growth and production of common bean under different strategies of mineral N fertilization associated with co-inoculation with R. tropici and A. brasilense. The experiment had seven treatments with four replications: control; inoculation with R. tropici without N application; control with mineral N; co-inoculation with R. tropici and A. brasilense without N fertilization or application of 20 kg ha-1 of N at sowing; inoculation with R. tropici and 20 kg ha-1 of N at sowing; inoculation with A. brasilense and 20 kg ha-1 of N at sowing. Plants were harvested at 45 days after sowing and at grain maturity. Co-inoculation combined with the N fertilization resulted in higher mass values of leaves, stem, shoot, dry mass of grains, number of grains per plant and accumulation of N in pods and grains. / O objetivo do trabalho foi avaliar a produ??o de biomassa e de gr?os em feijoeiro comum (Phaseolus vulgaris) sob co-inocula??o de riz?bio e bact?rias promotoras de crescimento vegetal. O trabalho foi dividido em tr?s cap?tulos. O primeiro teve como objetivo avaliar a compatibilidade entre riz?bio e bact?rias promotoras de crescimento e seus efeitos no crescimento das plantas. Em ensaio em placas de Petri, Azospirillum amazonense e A. brasilense foram co-inoculados com Rhizobium tropici em tr?s doses de aplica??o, verificando-se aus?ncia de antagonismo. O experimento em vasos de Leonard teve esquema fatorial 5x3+3 com oito repeti??es, composto por cinco bact?rias (Bradyrhizobium diazoefficiens, B. elkanii, Azospirillum amazonense, A. brasilense e Escherichia coli) e tr?s doses de aplica??o (104, 106 e 108 UFC mL-1), todas co-inoculadas com R. tropici, e tr?s tratamentos adicionais (testemunhas absoluta, nitrogenada, e inoculada com R. tropici). A co-inocula??o aumentou a nodula??o, teor e conte?do de N na parte a?rea. O experimento em vasos com solo teve esquema fatorial 3x2x2+4 com quatro repeti??es, composto por tr?s promotores de crescimento (B. diazoefficiens, B. elkanii e A. brasilense), duas doses (104 e 106 UFC mL-1), duas fontes de N (R. tropici e N mineral), e quatro tratamentos adicionais (testemunhas absoluta, nitrogenada, inoculada com riz?bio e co-inoculada com AZOTOTAL?). A inocula??o com A. brasilense associada ? aduba??o nitrogenada aumentou a produ??o de folhas e de parte a?rea. O segundo cap?tulo teve como objetivo avaliar o crescimento e produ??o do feijoeiro sob co-inocula??o. Foi realizado um experimento em vasos com solo em esquema fatorial 2x3+4 com oito repeti??es, combinando a inocula??o de R. tropici com A. brasilense ou B. diazoefficiens e tr?s doses de aplica??o (104, 106 e 108 UFC mL-1), acrescidos das testemunhas absoluta, nitrogenada, inoculada com riz?bio e co-inoculada com AZOTOTAL?. As plantas foram coletadas aos 45 dias ap?s plantio e na matura??o dos gr?os. As plantas co-inoculadas apresentaram massa de parte a?rea, n?mero de vagens, conte?do de N na parte a?rea, massa de gr?os, n?mero de gr?os e de vagens por planta, superiores ? testemunha inoculada com riz?bio. O ?ndice de clorofila Falker foi similar nos tratamentos co-inoculados e sob aduba??o nitrogenada, mas inferior ao da inocula??o com R. tropici. No terceiro cap?tulo avaliou-se o crescimento e produ??o do feijoeiro sob diferentes estrat?gias de aduba??o nitrogenada associada ? co-inocula??o. O experimento consisitu de sete tratamentos com oito repeti??es: testemunha absoluta; inocula??o com R. tropici sem aplica??o de N; testemunha com N mineral; co-inocula??o com R. tropici e A. brasilense sem N ou com 20 kg ha-1 de N em semeadura; inocula??o com R. tropici e 20 kg ha-1 de N em semeadura; inocula??o com A. brasilense e 20 kg ha-1 de N em semeadura. As plantas foram coletadas aos 45 dias e na matura??o. A co-inocula??o combinada com a aduba??o nitrogenada resultou em maiores valores de massa de folhas, caule, parte a?rea, massa seca de gr?os, n?mero de gr?os por planta e ac?mulo de N nas vagens e gr?os.

Orientador: Marcelo Andreotti / Resumo: Os sistemas integrados de produção agropecuária são alternativas viáveis para recuperação de áreas degradadas, rotação de culturas e produção de palhada com efeitos sobre as propriedades físicas, químicas e biológicas do solo. Esta pesquisa objetivou, em um LATOSSOLO VERMELHO distrófico, avaliar o desempenho produtivo da soja em sucessão ao capim Marandu, que foi submetido à três manejos: adubado com 50 kg ha-1 de N (Ureia) a cada corte antecessor, simulando o pastejo (0,30 m de altura), totalizando 4 cortes com intervalo de um para o outro de aproximadamente 30 dias, ou com pulverização foliar a cada corte com Azospirillum brasilense, e numa condição sem adubação nitrogenada e sem pulverização com Azospirillum após cada corte (controle), e em duas condições de semeadura da soja, com sementes inoculadas apenas com Bradyrhizobium japonicum ou co-inoculada com Azospirillum brasilense. A pesquisa constituiu-se de dois experimentos (um em área de sequeiro e outro em área irrigada por aspersão – pivô central) e foi realizada no ano agrícola 2016/2017 na Fazenda de Ensino, Pesquisa e Extensão (FEPE), localizada no município de Selvíria-MS. A cultura anterior ao experimento, em ambas as áreas, foi a Urochloa brizantha cv Marandu advinda do consórcio com milho, onde após a colheita do milho, o capim permaneceu na área até a semeadura da soja BMX POTÊNCIA RR. O delineamento experimental foi o de blocos casualizados, com quatro repetições, em esquema fatorial 3 x 2, constituídos do cul... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Crop livestock are viable alternatives for the recovery of degraded areas, crop rotation and straw production with effects on the physical, chemical and biological properties of the soil. The aim of the research in a dystrophic RED LATOSOLO, was to evaluate the productive performance of soybean in succession to the Marandu grass, which was submitted to three treatments: fertilized with 50 kg ha-1 of N (Urea) to each predecessor cut, simulating grazing (0,30 m high), totaling 4 cuts with interval from one to another of approximately 30 days, or with leaf spraying at each cut with Azospirillum brasilense, and in a condition without nitrogen fertilization and without spraying with Azospirillum after each cut (control), and under two conditions of soybean sowing, inoculated only with Bradyrhizobium or co-inoculated with Azospirillum brasilense. The research consisted of two experiments (one in the dry area and the other in a overhead irrigation area - central pivot) and was carried out in the agricultural year 2016/2017 at the Fazenda de Ensino, Pesquisa e Extensão (FEPE), located in the municipality of Selvíria-MS. The culture before the experiment, in both areas, was Urochloa brizantha cv. Marandu from the maize consortium, and after maize harvesting, the grass remained in the area until sowing of soybean BMX POTÊNCIA RR. The experimental design was a randomized complete block design, with four replications, in a 3 x 2 factorial scheme, consisting of the soybean crop in success... (Complete abstract click electronic access below) / Mestre

Submitted by Osvaldo Araujo (osvaldo.araujo.agro@gmail.com) on 2018-03-27T18:26:29Z No. of bitstreams: 1 Arauj Junior_o_ilha (3).pdf: 1043962 bytes, checksum: dce6e72d8aff0d848771cdd45f1576d6 (MD5) / Approved for entry into archive by Cristina Alexandra de Godoy null (cristina@adm.feis.unesp.br) on 2018-03-27T19:27:27Z (GMT) No. of bitstreams: 1 araujojunior_o_me_ilha.pdf: 1043962 bytes, checksum: dce6e72d8aff0d848771cdd45f1576d6 (MD5) / Made available in DSpace on 2018-03-27T19:27:27Z (GMT). No. of bitstreams: 1 araujojunior_o_me_ilha.pdf: 1043962 bytes, checksum: dce6e72d8aff0d848771cdd45f1576d6 (MD5) Previous issue date: 2018-02-22 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Os sistemas integrados de produção agropecuária são alternativas viáveis para recuperação de áreas degradadas, rotação de culturas e produção de palhada com efeitos sobre as propriedades físicas, químicas e biológicas do solo. Esta pesquisa objetivou, em um LATOSSOLO VERMELHO distrófico, avaliar o desempenho produtivo da soja em sucessão ao capim Marandu, que foi submetido à três manejos: adubado com 50 kg ha-1 de N (Ureia) a cada corte antecessor, simulando o pastejo (0,30 m de altura), totalizando 4 cortes com intervalo de um para o outro de aproximadamente 30 dias, ou com pulverização foliar a cada corte com Azospirillum brasilense, e numa condição sem adubação nitrogenada e sem pulverização com Azospirillum após cada corte (controle), e em duas condições de semeadura da soja, com sementes inoculadas apenas com Bradyrhizobium japonicum ou co-inoculada com Azospirillum brasilense. A pesquisa constituiu-se de dois experimentos (um em área de sequeiro e outro em área irrigada por aspersão – pivô central) e foi realizada no ano agrícola 2016/2017 na Fazenda de Ensino, Pesquisa e Extensão (FEPE), localizada no município de Selvíria-MS. A cultura anterior ao experimento, em ambas as áreas, foi a Urochloa brizantha cv Marandu advinda do consórcio com milho, onde após a colheita do milho, o capim permaneceu na área até a semeadura da soja BMX POTÊNCIA RR. O delineamento experimental foi o de blocos casualizados, com quatro repetições, em esquema fatorial 3 x 2, constituídos do cultivo da soja em sucessão ao capim-marandu submetido à três manejos e em duas condições de semeadura da soja, inoculada apenas com Bradyrhizobium ou co-inoculada com A. brasilense. A co-inoculação via semente de Bradyrhizobium e Azospirillum não incrementou o crescimento, os componentes da produção e produtividade da soja sob Sistema Plantio Direto (SPD), independentemente se em área de sequeiro ou irrigada. Em ambos os manejos, sequeiro e irrigado, a pulverização foliar de A. brasilense no capim Marandu é viável para o SIPA promovendo produção de forragem satisfatória em relação à adubação nitrogenada, bem como este residual de A. brasilense no solo, resultante da aplicação foliar no capim Marandu antecessor à soja, aumentou a massa de nódulos da soja, mostrando ser uma alternativa viável para potencializar a fixação biológica de N2, principalmente em condições de sequeiro. / Crop livestock are viable alternatives for the recovery of degraded areas, crop rotation and straw production with effects on the physical, chemical and biological properties of the soil. The aim of the research in a dystrophic RED LATOSOLO, was to evaluate the productive performance of soybean in succession to the Marandu grass, which was submitted to three treatments: fertilized with 50 kg ha-1 of N (Urea) to each predecessor cut, simulating grazing (0,30 m high), totaling 4 cuts with interval from one to another of approximately 30 days, or with leaf spraying at each cut with Azospirillum brasilense, and in a condition without nitrogen fertilization and without spraying with Azospirillum after each cut (control), and under two conditions of soybean sowing, inoculated only with Bradyrhizobium or co-inoculated with Azospirillum brasilense. The research consisted of two experiments (one in the dry area and the other in a overhead irrigation area - central pivot) and was carried out in the agricultural year 2016/2017 at the Fazenda de Ensino, Pesquisa e Extensão (FEPE), located in the municipality of Selvíria-MS. The culture before the experiment, in both areas, was Urochloa brizantha cv. Marandu from the maize consortium, and after maize harvesting, the grass remained in the area until sowing of soybean BMX POTÊNCIA RR. The experimental design was a randomized complete block design, with four replications, in a 3 x 2 factorial scheme, consisting of the soybean crop in succession to the Marandu grass submitted to three treatments and in two sowing conditions of soybean, inoculated only with Bradyrhizobium or co-inoculated with A.brasilense. The co-inoculation of Bradyrhizobium and Azospirillum in the seeds did not increase the growth, the components of the production and productivity of the soybeans under no-tillage system, irrespective of dry or irrigated areas. In both dry and irrigated managements, the A. brasilense leaf spray in the Marandu grass is viable for the SIPA promoting satisfactory forage production in relation to the nitrogen fertilization, as well as this residual of A. brasilense in the soil, resulting from the foliar application in the Marandu grass predecessor to soybean, increased the nodule mass of the soybean, proving to be a viable alternative to potentiate the biological N2 fixation, mainly in dry conditions.

L’ocratossina e le ammine biogene sono due metaboliti biologici che possono essere ritrovati nei vini. Il primo di questi contaminanti è stato studiato recentemente per la sua elevata tossicità sebbene non sia rintracciabile frequentemente nei vini. Le ammine biogene sono presenti in ogni tipo di vino in differenti concentrazioni, oltretutto, alcune di loro, ad alte concentrazioni, possono causare reazioni allergiche. Gli obiettivi di questo PhD riguardano tre differenti aspetti. Indagare metodi biologici di per la decontaminazione di ocratossina A durante la vinificazione; in particolare lo studio si è focalizzato sull’uso di un ceppo Lactobacillus plantarum utilizzato come starter malolattico. Valutare la presenza di Lactobacillus spp., isolati da mosto e vino, in grado di produrre ammine biogene, usando tecniche molecolari come la reazione a catena della polimerasi (PCR) per rilevare i geni codificanti gli enzimi responsabili della sintesi di questi composti. Testare la capacità di un L. plantarum di effettuare la malolattica effettuando l’inoculo in differenti fasi della vinificazione e valutare il trend delle ammine biogene già presenti nel mosto. / Two of the major biological metabolites present in wine are the ochratoxin and the biogenic amines. The first of these contaminants was studied in recent decades because of its toxicity in humans, although its presence is not frequent in wines. The biogenic amines are present in every types of wine in different concentration, and some of them, in high concentrations, can cause allergenic reactions in humans. The objectives of this PhD regard three different aspects. Investigate the biological methods to reduce ochratoxin A in wine during winemaking; in particular the study is focused to use a Lactobacillus plantarum strain as malolactic starter. Investigate the presence of Lactobacillus spp., isolated from must and wine, able to produce the amines, using molecular techniques as polymerase chain reaction (PCR) to detect the genes that encode for the enzymes responsible of the synthesis of these compounds. Test the ability of a L. plantarum to perform MLF in relationship with inoculation time and assess the trend of biogenic amines already present in must.