PROPRIETA' FUNZIONALI DELLA BARRIERA INTESTINALE DI INSETTO E MODULAZIONE DELLA PERMEABILITA' PARACELLULARE.

Control of insect pests is still mainly performed using broad spectrum chemical pesticides even though the limits of this strategy are presently well known: the reduction in pest population is associated with an unfavourable alteration of food quality and safety, with a strong impact on nontarget species and with the rise of a widespread resistance in target insects. The most recent approach to the Integrated Pest Management is based on the detection of new genes encoding for polypeptides with potential insecticide activity, with a particular attention to biopesticides derived from insect antagonists and plants (Whetstone e Hammock, 2007). However, efficient delivery methods are essential in deploying bioinsecticides. For an effective oral delivery, it will be crucial to develop strategies to facilitate their passage through the midgut barriers, i.e. the peritrophic membrane (PM) and the midgut epithelium (ME). This can be achieved by altering the sieving properties of the PM and by increasing the rate of absorption by the ME. In the framework of a coordinated effort towards the development of new delivery strategies my laboratory, in collaboration with other research groups, discovered that the recombinant Chitinase A (ChiA) of Autographa californica multicapsid nucleopolyhedrovirus (AcMNPV) determined in vitro structural alterations on lepidopteran larvae PM and a strong increase of its permeability to molecules (Rao et al., 2004). Moreover, when this enzyme is delivered either with artificial diet or with transgenic plants to lepidopteran larvae, a significant negative effect on insect biological performance and survival was observed. I demonstrated that PMs isolated from fifth instar Heliothis virescens larvae fed on transgenic tobacco plants expressing ChiA starting from the first day of the fourth instar show an increased flux of methylene blue. The PMs were mounted in Ussing chambers and incubated in the presence of methylene blue, a good tracer of PM permeability, in the endoperitrophic compartment. The flux of methylene blue through the PMs isolated from larvae reared on ChiA-expressing tobacco plants was significantly higher than that of controls (PM isolated from larvae reared on control tobacco plants). To demonstrate that the increased permeability was due to the hydrolytic activity of AcMNPV ChiA on the PM chitin mesh, this enzyme was extracted and purified from ChiA-expressing tobacco plants and its activity tested on the permeability of PMs of H. virescens larvae isolated in Ussing chambers. PM incubation with ChiA (40 μM) in the endoperitrophyc compartment caused a significant increase of TMOF (Trypsin Modulating Oostatic Factor) flux compared to control. Thus, the use of this enzyme, and of any other similar enhancer, can be of great interest in the development of effective delivery strategies. Once crossed the PM, the macromolecules have to pass the ME. They can reach the haemocoel either through the cellular pathway, crossing the two polarized plasma membranes of the epithelial cells, or through the paracellular pathway, along the aqueous channels formed by the junctional complexes. The latter, a preeminent pathway for the delivery of small peptides (Fiandra et al. 2009), is the main objective of my study. In my lab a number of functional properties of the smooth septate junction (SJ) of the lepidopteran larval midgut has been recently elucidated in vitro, demonstrating that its permeability can be modulated by cAMP and/or a fine regulation of cytosolic Ca2+ concentration (Fiandra et al., 2006). During my PhD, I have studied one of the signaling pathway(s) that may lead to the intracellular release of Ca2+ responsible for the increase of the junction permeability. I took advantage of a medium-chain fatty acid (C10) known to modulate the mammalian tight junction by activation of a Ca2+ ions-mediated intracellular signaling pathway (Cano-Cebrian et al., 2005). I demonstrated that the addition of C10 (20 mM) to the luminal side of Bombyx mori larval midguts isolated in conventional Ussing chambers, caused a decrease of the paracellular (shunt) electrical resistance (Rsh) (i.e. an increase of the junction ion conductance) and an increase of the paracellular fluxes of proctolin and fluorescein, two molecules that cross the midgut epithelium only through the paracellular route. Afterward, I evaluated if the effect of C10 on the shunt electrical resistance in lepidopteran larvae was due to the activation of a Ca2+ ions-mediated intracellular signaling pathway like in the mammalian tight junction, i.e. if C10 induced a raise of the intracellular calcium concentration. B. mori midgut cells, obtained by enzymatic disaggregation of the tissue, were pre-loaded with the cell permeant Ca2+-sensitive fluorescent dye Fluo-3AM, incubated in the absence (control) or in the presence of C10 (2 mM) and observed by fluorescence microscopy. Control cells did not show an appreciable fluorescence intensity during the entire experimental period. On the contrary, midgut cells incubated with C10 for 5 min showed a strong fluorescence signal, no more observable after 10 min of incubation. These data demonstrate that C10 causes a transient increase of the free cytosolic calcium concentration and therefore that its effect on the shunt resistance can be due, like in mammals, to the activation of a Ca2+ ions-mediated intracellular signaling pathway. Then, I analyzed which is the signalling pathway activated by C10 that leads to the raise of the intracellular calcium concentration in lepidopteran midgut cells. I examined whether specific inhibitors of major proteins involved in signal cascade activated by C10 in mammalian cells were able to abolish the decrease of the paracellular resistance induced by C10 on B. mori larval midgut. The inhibitors used are: U-73122, inhibitor of phospholipase C (PLC), W-7 inhibitor of calcium-calmodulin binding, ML-7 and PEPTIDE-18, which inhibit the kinase responsible for phosphorylation of the myosin light chain (MLCK). Then I calculated the value of the paracellular resistance in the midgut of B. mori larvae mounted in Ussing chamber in the presence of C10 or in the presence C10 and inhibitors. The results obtained indicate that all the inhibitors tested are able to abolish the effect caused by the C10 on Rsh. Summarizing the data, I demonstrated that C10 triggers an IP3- dependent signalling cascade by activation of phopholipase C. As a result, Ca2+ is released from the intracellular stores and a calmodulin-dependent kinase is activated. This event leads to the phosphorylation of the myosin light chain by myosin light chain kinase. Myosin light chain phosphorylation presumably causes a modification of the cytoskeleton organization connected to the SJ and thereby an increase of the paracellular permeability to ions and small organic molecules. Even though the study of the permeability through the peritrophic membrane and the paracellular pathway represents the main part of my PhD project, I have also collaborated to the characterization of the early events involved in Junonia coenia densovirus (JcDNV) infection of the lepidopteran pest Spodoptera frugiperda. A few years ago my laboratory has begun a collaboration with Dr. Mylene Ogliastro (“Laborotoire de Biologie Integrative et Virologie” INRA-UMII - Institute National de la Researche Agronomique-Université de Montpellier II-, France) with the aim to clarify some aspects of JcDNV pathogenicity. Densoviruses (DNVs) are parvoviruses highly pathogenic for arthropods, mostly insects, at larval stages, including agronomical pests and insects vector-borne diseases (Bergoin, 2008). DNVs have a limited host-range and are not pathogenic to vertebrates, characteristics that make them particularly interesting as potential pest control agents alternative to chemical pesticides. During the second year of my PhD I spent a period in Montpellier, in Dr. Ogliastro’s lab, where I collaborated to the elucidation of the mechanism used by JcDNV to enter midgut columnar cells. Midguts of S. frugiperda larvae were isolated, cut longitudinally and interposed as a sheet between the two half chambers of the Ussing perfusion apparatus. Midguts were preincubated for 30 min in the absence (control) or in the presence of different drugs: dynasore, (400 μM) that inhibits clathrin-coated vesicle formation blocking the GTPase dynamine, the sterol-sequestering drug methyl cyclodextrine (60 μM) that inhibits lipid raft-dependent endocytosis and the fungal toxin wortmannin (10 mM), a PI3K inhibitor. Virus was then added to the luminal compartment and the incubation lasted 10 min. To verify if these drugs inhibit JcDNV internalization we performed immunofluorescence assays. Midguts incubated with the drugs did not present virus particles inside the cells or, however, a lesser amount compared to control sample, a clear indication that clathrin mediated endocytosis, lipid-raft dependent endocytosis and PI3K-dependent pathways are involved in JcDNV internalization. These data indicate that JcDNV uptake into midgut columnar cells of S. frugiperda larvae is mediated by different endocytic mechanisms. This finding is not surprising, since a number of viruses have been shown to utilize more than one pathway for internalization.