Biotechnological strategies to improve safety and quality in food products

The purpose of this PhD research is to facilitate the development of green and successful strategies for the control of undesirable microorganisms in food products. It is well known that many stress resistant bacteria are able to contaminate food products and produce their spoilage or, worse still, be a potential source of human illness. In the last decade, illnesses resulting from food borne pathogens have been higher than in the past and have become one of the most widespread public health problems in the world. Contextually, contaminations with spoilage microorganisms remain a major threat for the industry and food-based market, so much so, that consumers are not only paying more attention to the risk of foodborne pathogens but also the safety of chemical preservatives that are used to control undesirable microorganisms. It is, therefore, essential to find a satisfactory solution and useful strategy to prevent or reduce the incidences of pathogens or spoilage microorganisms. In the last two decades, much attention has been focused on food bio preservation, a “green strategy” that can assure shelf-life extension and food safety using microorganisms or their antimicrobial compounds. Lactic acid bacteria could be considered an ideal choice for application as protective cultures in food products and, more specifically, Lactobacillus plantarum is the most versatile and widespread species. Several screening processes are developed in order to select the most appropriate effective strains to be used as protective cultures, including the production of bacteriocins, BLIS, organic acids, hydrogen peroxide as well as short chain fat acids. However, these screening programmes are labour intensive and a large number of strains isolated from different food matrices are assessed, thereby requiring more expensive investments in order to avoid unsatisfactory results. These findings call for a more simplified and useful approach when searching for new protective strains, taking into account that food stress conditions strongly influence the development of specific microbial strains. It would be extremely interesting to ascertain the effect of different environments on the selection of strains able to exert antimicrobial activities and, therefore, this research looks at the correlation between the Lb. plantarum strain isolated from hard environments and the ability to produce antimicrobial compounds. In addition, the main antimicrobial compound produced by producers Lb. plantarum strains and its mode of action was also investigated. This dissertation assesses the observations and the main significant results were reported in six chapters. Chapter 1 is an overview of the biocontrol strategies developed. More specifically, advancements in control strategies based on natural compounds and living organisms and/or their antimicrobial products (biocontrol, or bio preservation) were highlighted. These natural preservation methods are regarded as health-friendly by consumers, and are expected to have a lower impact on food nutritional and sensory properties. In addition, they may reduce the processing costs and, at the same time, extend the product shelf life period. However, until now, several issues, such as the high minimal inhibitory concentration levels, the stability of antimicrobial compounds, the knowledge of action mode, as well as the relation between microbial growth and compound formation kinetics, still remain unclear, making the individuation of a simplified screening procedure necessary. In Chapter 2 are reported the objectives of PhD research. Chapter 3 considers the relation between antimicrobial properties of Lb. plantarum strains and their source of isolation. For this purpose, a total of 110 Lb. plantarum strains were used as antagonistic strains (producers) against 33 undesirable microorganisms (indicators), including both moulds and bacteria. The antimicrobial activity exerted by cells, cell free supernatants (CFS), neutralized CFS (nCFS) or CFS added with α-chymotrypsin, proteinase K, and trypsin (pCFS) of the producer strains was evaluated by the spot-on-the-lawn and by the agar well diffusion assay. Moreover, the inhibition effects expressed by cell free supernatant (CFS) and by neutralized cell free supernatant of selected strains was evaluated in culture-broth expressed against strains belonging to Ps. fluorescens, B. thermosphacta and L. innocua. The preliminary results achieved by the evaluation of the antimicrobial effects expressed by CFS and the correspondent neutralized CFS support the hypothesis that the inhibition was due to the production of extracellular compounds having neither acid (such as lactic acid, that represent the principal extracellular metabolites produced by Lb. plantarum) nor proteinaceous nature. In addition, the results evidenced that the inhibitory effect produced by certain Lb. plantarum strains also remains at higher pH values. Therefore, the comparison between the inhibitory effects produced by CFS and lactic acid could provide more information on the antimicrobial compound. Moreover, in order to better appreciate differences between lactic acid and CFS, the most lactic acid resistant strains among the indicators should be chosen. L. innocua strains are well known for their acid stress resistance. The relative results evidenced that the inhibitory effect of CFS from Lb. plantarum H_BB1 against L. innocua was due to the synergic presence of more than one inhibitory substance. A more in depth, further investigation evidenced that in addition to lactic acid, the CFS might also possess another compound of acid nature. On the basis of this evidence, PLA in the cells free supernatant from strains (able to produce antimicrobial effects), was evaluated. The results highlighted significant differences among the assayed strains showing that PLA production is strain-dependent. In addition, for the first time, a relation between PLA-producing strains and isolation environment of the strains was highlighted. In fact, those environments characterised by harsh conditions (high ethanol levels, low pH and high sugar levels), such as wines and honey, harboured a higher number of antagonistic strains than other fermented matrices (e.g. cheese, sourdoughs or fermented sausages). This could be due to selective pressures which are more accentuated in wines and honey than in the other food matrices researched. The most important scientific enrichment produced by the activities in Chapter 1 is attributable to results highlighting that the choice of the source of isolation could be an important preliminary tool for the individuation of antagonistic strains. However, the correlation between Lb. plantarum PLA formation ability and their isolation sources would lead to opening new frontiers in understanding the PLA formation process. PLA formation seems to be linked to stress response mechanisms performed by Lb. plantarum. However, no information with regards the LAB stress response and PLA production is available in literature and little information is reported on the relation between the microbial growth phase and PLA formation. Even if the prevailing opinion in the scientific community believes that PLA formation is related to LAB growth arrest, the linkage to metabolic pathways involved in its stationary phase has not been clarified. Little information, if any, can be found on the optimal pH condition of PLA metabolic pathway in the Lactobacillus species. With this in mind, the research reported in Chapter 4 focuses on the effect of growth phase and on the PLA formation by Lb. plantarum H_BB1. Moreover, cultivation conditions that were able to assure the highest PLA levels were investigated. The production of PLA by Lb. plantarum H_BB1 was preliminarily investigated in MRS broth. The comparison between PLA behaviour and growth curves evidenced that the PLA accumulation begins immediately after the end of the lag phase and reached the highest levels between the exponential and the stationary phase. As far as I know, these results show, for the first time, that the PLA production is strictly related to the growth and to the exponential phase of Lb. plantarum. More specifically, the results obtained in the present study suggest that the PLA could assume new metabolic meanings. In fact, the accumulation from the beginning of the exponential phase highlight typical behaviour of a primary metabolite. On the other hand, the highest production rate between the exponential and stationary phase suggests that PLA production could assume a key role in the acid stress response. In fact, in the transition between exponential and stationary phase, pH showed values similar to the pKa of lactic acid. On the basis of the above reported statements, the evaluation of ecological factors on the PLA formation process appear essential. More precisely, sub-lethal pH could positively affect some metabolic pathways in Lb. plantarum. For this purpose, PLA production by Lb. plantarum was assayed in different cultural conditions (MRS acidified to pH 4.0 and to pH 3.5) and the results suggest that the metabolic pathway involved in PLA formation is tied to the energetic metabolism of growing cells. Key reactions of PLA formation, such as the regeneration of NAD+ levels, the transamination reaction (where the -ammino group is transferred to a keto acid acceptor) and the deamination reactions with NH3 and amino acceptor regeneration, found different linkages with typical metabolic activities of growing cells. The results evidenced that, in no way, could the PLA formation be related to cell growth arrest. Whereas, its formation could represent an adaptation response of growing cells to acid stress. In fact, evaluating the behaviour of ratio between PLA (g/L) and biomass (g/L) levels, the highest performances were detected when the strain was cultivated in MRS pre-acidified to pH 4.0. On the basis of the above reported results, the PLA could be considered a “primary-like metabolite” of Lb. plantarum in sub-optimal pH condition and may open new horizons to the development of an advanced optimal design for maximum PLA production. Moreover, the evidences reported in state-of-the-art Chapter 1, together with data specified in Chapters 2 and 3 suggest the opportunity to use PLA as anti-Listeria compounds. To date, antimicrobial activity of PLA, including anti-Listeria ability, was well recognized. Whereas, little information is available on the PLA anti-Listeria mechanism. Some results reported in literature suggest that the PLA mode of action could be similar to lactic acid. Nevertheless, considering the chemical structure of PLA, an action mode which is different from the lactic acid and more similar to phenolic acids could also be hypothesized. In Chapter 5, the anti-Listeria mechanism of 3-phenyllactic acid was investigated and, hence, the antimicrobial effect of PLA was evaluated to different pH. Moreover, the PLA anti-Listeria has been compared with those expressed by the lactic acid and the better studied hydroxybenzoic and hydroxycinnamic acids. Listeria innocua was chosen as the indicator to investigate the antimicrobial mechanism of PLA. Listeria innocua is regarded as a non-pathogenic indicator for the presence of Listeria monocytogenes in foods. This study has evidenced that PLA, which, for many years, has been considered an antifungal metabolite, is also able to inhibit bacteria cells. Moreover, very low concentrations were required to produce anti-Listeria activity. MIC values of about 0.47 or 0.94 mg/mL appear compatible with the maximum PLA production revealed in Lb. plantarum cultures. In fact, this PhD research (chapters 2 and 3), recognizes that Lb. plantarum strains produced up to 0.12 or 0.23 mg/mL and this production level could be increased two or even tenfold when specific cultural strategies were applied. Therefore, a resolution to the gap between PLA required to assure antimicrobial activity and the PLA levels detected in fermentation batches seems possible. This gap has long proven to be a serious obstacle when applying PLA producing bacteria as protective or as anti-Listeria cultures in food characterized by neutral or sub-acid pH. Moreover, the relation between pH values and anti-Listeria activity of PLA was clarified. A relation between MICs and pH values was found and a significant reduction in PLA anti-Listeria activity was detected at the highest pH values tested. In order to understand the PLA mode of action, the anti-Listeria effect produced by PLA at pH 5.5 was compared to the effect illustrated by lactic acid to the same pH value. Lactic acid was the best choice as its antimicrobial mechanism is well known. The results evidenced that the antimicrobial action of PLA was substantially different from LA. The different and more successful effect produced by PLA must be due to its amphiphilic properties resulting from the hydrophobic group-benzene ring and hydrophilic group-carboxy in its chemical structure. These properties would allow an interaction with the lipid and protein in cytoplasmic membrane as well as an interaction with genomic materials. Therefore, a comparison between PLA and the more studied phenolic acids (hydroxybenzoic and hydroxycinnamic) should be researched. Furthermore, to better understand the antimicrobial effect of PLA on L. innocua, further experiments were conducted using three phenolic compounds (gallic, ferulic and caffeic acid). The results generated from both an MIC and MBC survival test, highlighted that the gallic acid and PLA showed the most performing anti-Listeria activity. However, the differences between PLA and GA in death kinetic parameters suggest that PLA produces anti-Listeria activity through a specific mechanism which is somewhat different from those usually adopted by other phenolic compounds. Antimicrobial activity of phenolic acids involves several mechanisms of action such as permeability destabilization or the rupture of the cytoplasmic membrane as well as enzymes inhibition through nonspecific interaction. It is possible to hypothesize that PLA utilizes more than one of these pathways but differently from the other phenolic compounds. More exhaustive results were obtained by the evaluation of the effect of phenolic acids on surface charge and loss of cellular content. In fact, zeta potential measurements demonstrated that after phenolic acids exposure, the cells become more (P<0.05) negatively charged when exposed to PLA. While in the presence of other phenolic compounds (GA, CA, FE) no variation in charge was detected. This fact may open new horizons to the understanding of the PLA anti-Listeria mechanism. It is possible to surmise that the PLA anti-Listeria action is also associated with the affinity with cell surface and the interaction PLA-cell surface could contribute to the damage of cellular structures. The rupture of cellular structures was also supported by the results of the cellular content loss. The results could help to explain the differences in the anti-Listeria mechanism of phenolic compounds. Hydroxybenzoic and hydroxycinnamic acid seem to induce an alteration in membrane permeability without causing its rupture. Whereas, PLA having the main targets in cellular surface and in cytoplasmic membrane, leads to a severe rupture of the cellular structures. All these evidences contribute to the enrichment of scientific knowledge in the anti-Listeria mechanism of PLA and highlighted that PLA effectiveness is superior to that expressed by other preservative acids. Finally, in Chapter 6 are reported the general conclusions of PhD research.

Le attività condotte nella presente tesi di dottorato hanno inteso individuare strumenti, tali da consentire un più agevole sviluppo di strategie “verdi” per il controllo di microrganismi indesiderati nei prodotti alimentari. Negli ultimi due decenni, molta attenzione è stata, infatti, focalizzata sulle "strategie verdi" che utilizzando microrganismi o composti antimicrobici possono assicurare l'estensione della shelf-life ed il controllo di microrganismi sempre più pericolosi come Listeria monocytogenes. Finora, sono stati sviluppati diversi processi di screening per selezionare ceppi più efficaci ed appropriati da utilizzare come culture protettive. Lactobacillus plantarum è sicuramente la specie più versatile e diffusa e produttore di acido 3-fenil-lattico, il quale sembra essere uno dei composti bio-conservanti più efficaci ed interessanti. Tuttavia, detti programmi di screening sono piuttosto laboriosi e spesso necessitano di valutare un gran numero di ceppi, isolati da diverse matrici alimentari. Pertanto, sono necessari investimenti particolarmente costosi al fine di evitare risultati insoddisfacenti. Inoltre, fino ad oggi, la concentrazione minima inibitoria, la stabilità dei composti antimicrobici, la conoscenza del modo d'azione, nonché la relazione tra la crescita microbica e la cinetica di formazione dei composti, non sono del tutto noti. È necessario, dunque, un approccio più semplificato ed utile per la selezione di nuovi ceppi protettivi, tenendo conto che le condizioni di stress alimentare possono influenzare fortemente lo sviluppo di specifici ceppi microbici. Sarebbe, quindi, estremamente interessante accertare l'effetto di diversi ambienti d’isolamento sulla selezione di ceppi in grado di esercitare attività antimicrobiche. Tuttavia, la correlazione tra la resistenza del ceppo alle condizioni di stress e la capacità di produrre effetti antimicrobici è stata poco studiata. Pertanto, attraverso le azioni previste dal progetto di dottorato è stata preliminarmente valutata la correlazione tra i ceppi di Lb. plantarum isolati da ambienti ostili e la capacità di assicurare un'azione inibitoria contro microrganismi indesiderati, nonché la correlazione tra matrice di isolamento e la capacità di questi ceppi di produrre un composto antimicrobico specifico (acido 3-fenillactico). I risultati oltre a confermare che la capacità di produrre PLA sia un carattere ceppo-dipendente hanno evidenziato una chiara relazione tra i ceppi produttori di PLA e le caratteristiche chimico-fisiche della matrice di isolamento. In dettaglio, matrici caratterizzate da caratteristiche chimico fisiche più restrittive (alti livelli di etanolo o di zucchero e bassi valori di pH), come il vino o il miele, permettono l’isolamento di ceppi con maggiori proprietà antagonistiche rispetto ad altre matrici fermentate, come formaggi, impasti acidi o insaccati fermentati. Sulla base di tali risultati sono stati valutati gli effetti della fase di crescita e delle condizioni di coltivazione sulla formazione di acido 3-fenil-lactico (PLA) prodotto da Lb. plantarum. Infatti, l’incremento ed il miglioramento della formazione di PLA rappresenta un argomento molto attuale e cruciale per assicurare una maggiore produzione di PLA nell'industria alimentare. La formazione di PLA potrebbe essere legata a meccanismi di risposta allo stress messi in atto da Lb. plantarum. Tuttavia, non sono disponibili informazioni sulla relazione tra la risposta dello stress dei LAB e la produzione di PLA, così come sono riportate poche informazioni sulla relazione tra la fase di crescita microbica e la formazione di PLA. La comunità scientifica ritiene che la formazione di PLA sia correlata all'arresto della crescita de LAB, tuttavia, non sono state spiegate le vie metaboliche coinvolte in tale produzione in fase stazionaria. Inoltre, poche informazioni sono riportate sulla condizione ottimale di pH del metabolismo di produzione del PLA nelle specie di Lactobacillus. Anche in tal caso sono emersi risultati fortemente innovativi. In dettaglio, l’accumulo nelle prime ore della fase esponenziale evidenzia, infatti, un comportamento tipico di un metabolita primario. D’altra parte, la più alta quantità prodotta tra la fase esponenziale e la fase stazionaria, suggerisce che la produzione di PLA potrebbe assumere un ruolo importante nella risposta allo stress acido. Dal presente studio è emerso, infatti, che un pH sub-ottimale (pH 4,0), migliora la formazione di PLA da parte di uno specifico ceppo di Lb. plantarum. Infine, è stato studiato il meccanismo anti-Listeria dell'acido 3-fenil-lattico. A tal fine, l'effetto antimicrobico del PLA è stato valutato a differenti valori di pH e l’attività anti-Listeria del PLA è stata confrontata con quella espressa dall'acido lattico e dagli acidi idrossibenzoici e idrossicinnamici, meglio studiati. I risultati ottenuti hanno evidenziato che il PLA esplica una spiccata attività antimicrobica nei confronti di Listeria innocua. L’attività anti-Listeria del PLA, ha dimostrato di essere pH dipendente, evidenziando un comportamento tipico di altri acidi deboli e acidi organici. È possibile ipotizzare che la modalità di azione del PLA sia correlata alla sua forma indissociata, la quale è in grado di attraversare la membrana microbica ed esplicare un maggiore effetto antimicrobico. L' azione antimicrobica del PLA è risultata sostanzialmente diversa da quella esplicata dell’acido lattico. L’effetto nettamente differente e più efficace del PLA potrebbe essere legato alle sue proprietà anfifiliche, conferitegli dall'anello benzoico idrofobo, e al gruppo carbossilico idrofilo presenti nella sua struttura chimica. Queste sue proprietà permetterebbero un'interazione con le molecole lipidiche e le proteine della membrana citoplasmatica ed un’interazione con il materiale genomico. Dal confronto con altri composti fenolici, i risultati evidenziano che l’attività anti-Listeria del PLA, viene attuata attraverso un meccanismo specifico, alquanto diverso da quello impiegato dagli altri composti fenolici. Gli acidi idrossicinnamici e l’acido idrossibenzoico sembrano indurre un'alterazione nella permeabilità della membrana senza causarne la rottura. Mentre, il PLA, interagendo con la superficie cellulare e la membrana citoplasmatica, determina una grave rottura delle strutture cellulari. Quanto riportato contribuisce, dunque, all'arricchimento delle conoscenze scientifiche sul meccanismo anti-Listeria del PLA, ed evidenziano una maggiore efficacia del PLA rispetto agli altri acidi preservanti impiegati nel presente studio.