Plasmonic Metasurfaces for Specific SERS Detection of Shiga Toxins.

The interest in the development of nanoscale plasmonic technologies has dramatically increased in recent years. The photonic properties of plasmonic nanopatterns can be controlled and tuned via their size, shape, or the arrangement of their constituents. In this work, we propose a 2D hybrid metallic polymeric nanostructure based on the octupolar framework with enhanced sensing property. We analyze its plasmonic features both numerically and experimentally, demonstrating the higher values of their relevant figures of merit: we estimated a surface-enhanced Raman spectroscopy (SERS) enhancement factor of 9 × 107 and a SPR bulk sensitivity of 430 nm/RIU. In addition, our nanostructure exhibits a dual resonance in the visible and near-infrared region, enabling our system toward multispectral plasmonic analysis. Finally, we illustrate our design engineering strategy as enabled by electron beam lithography by the outstanding performance of a SERS-based biosensor that targets the Shiga toxin 2a, a clinically relevant bacterial toxin. To the best of our knowledge, this is the first time that a SERS fingerprint of this toxin has been evidenced.

4-Aminopyrazolo[3,4-d]pyrimidine (4-APP) as a novel inhibitor of the RNA and DNA depurination induced by Shiga toxin 1.

Shiga toxin 1 (Stx1) catalyses the removal of a unique and specific adenine from 28S RNA in ribosomes (RNA-N-glycosidase activity) and the release of multiple adenines from DNA (DNA glycosylase activity). Added adenine behaves as an uncompetitive inhibitor of the RNA-N-glycosidase reaction binding more tightly to the Stx1-ribosome complex than to the free enzyme. Several purine derivatives and analogues have now been assayed as inhibitors of Stx1. Most of the compounds showed only minor differences in the rank order of activity on the two enzymatic reactions catalysed by Stx1. The survey highlights the importance of the amino group in the 6-position of the pyrimidine ring of adenine. Shifting (2-aminopurine) or substituting (hypoxanthine, 6-mercapto-purine, 6-methylpurine) the group greatly decreases the inhibitory power. The presence of a second ring, besides the pyrimidine one, is strictly required. Substitution, by introducing an additional nitrogen, of the imidazole ring of adenine with triazole leads to loss of inhibitory power, while rearrangement of the nitrogen atoms of the ring from the imidazole to the pyrazole configuration greatly enhances the inhibitory power. Thus 4-aminopyrazolo[3,4-d]pyrimidine (4-APP), the isomer of adenine with the five-membered ring in the pyrazole configuration, is by far the most potent inhibitor of both enzymatic reactions catalysed by Stx1. This finding opens perspectives on therapeutic strategies to protect endothelial renal cells once endocytosis of Stx1 has occurred (haemolytic uraemic syndrome). In the RNA-N-glycosidase reaction 4-APP binds, as adenine, predominantly to the Stx1-ribosome complex (uncompetitive inhibition), while inhibition of the DNA glycosylase activity by both inhibitors is of the mixed type.

Uncompetitive inhibition by adenine of the RNA-N-glycosidase activity of ribosome-inactivating proteins.

Ricin is a member of the ribosome-inactivating protein (RIP) family with RNA-N-glycosidase activity which inactivates eukaryotic ribosomes by specifically removing adenine from the first adenosine of a highly conserved GAGA loop present in 28S rRNA. Free adenine protects ribosomes in cell-free systems from inactivation by ricin. Protection by adenine is highly specific, since AMP, adenosine and modified adenines (1-methyladenine and ethenoadenine) were completely ineffective. Kinetic analysis of the behaviour of adenine as inhibitor of the RNA-N-glycosidase reaction catalysed by ricin, Shiga-like toxin I and momordin, two other members of the RIP family, established that inhibition was of the uncompetitive type, the inhibitor binding to the enzyme-substrate complex. Adenine did not protect ribosomes from alpha-sarcin, an RNAase that inactivates ribosomes by cleaving the phosphodiester bond located in the GAGA loop at one nucleotide distance from the adenosine depurinated by the RNA-N-glycosidases. Adenine at the concentration of 1 mM lowered 1.5-fold the toxicity of ricin and 3.7-fold that of Shiga-like toxin I on Vero cells in culture. The same concentration of adenine decreased 2.4-fold the inactivation of isolated ribosomes by ricin, 2.8-fold the inactivation by Shiga-like toxin I and 20-fold that by momordin.

Nucleotides U28-A42 and A37 in unmodified yeast tRNA(Trp) as negative identity elements for bovine tryptophanyl-tRNA synthetase.

Wild-type bovine and yeast tRNA(Trp) are efficiently aminoacylated by tryptophanyl-tRNA synthetase both from beef and from yeast. Upon loss of modified bases in the synthetic transcripts, mammalian tRNA(Trp) retains the double recognition by the two synthetases, while yeast tRNA(Trp) loses its substrate properties for the bovine enzyme and is recognised only by the cognate synthetase. By testing chimeric bovine-yeast transcripts with tryptophanyl-tRNA synthetase purified from beef pancreas, the nucleotides responsible for the loss of charging of the synthetic yeast transcript have been localised in the anticodon arm. A complete loss of charging akin to that observed with the yeast transcript requires substitution in the bovine backbone of G37 in the anticodon loop with yeast A37 and of C28-G42 in the anticodon stem with yeast U28-A42. Since A37 does not prevent aminoacylation of the wild-type yeast tRNA(Trp) by the beef enzyme, a negative combination apparently emerges in the synthetic transcript after unmasking of U28 by loss of pseudourydilation.

Differential up-regulation by tRNAs of ribosome-inactivating proteins.

Some plant ribosome-inactivating proteins (RIPs) with RNA-N-glycosidase activity on 28S RNA require, for the inactivation of ribosomes, the presence of macromolecular cofactors present in post-ribosomal supernatants. In the case of gelonin one of the cofactors is tRNATrp lacking one or two nucleotides at the 3'-CCA end [Brigotti, M., Carnicelli, D., Alvergna, P., Pallanca, A., Lorenzetti, R., Denaro, M., Sperti, S. and Montanaro, L. (1995) Biochem. J. 310, 249-253]. In the present study it is shown that tRNAs are involved in the up-regulation of all the cofactor-requiring RIPs up to now identified (agrostin, barley RIP, PAP and tritin, besides gelonin). Polyacrylamide gel electrophoresis shows that tRNA fractions with different mobilities stimulate different RIPs. With the identification of agrostin, the cofactor-requiring RIPs (italics) add to five out of a total of thirteen investigated: barley RIP, bryodin-R, gelonin, lychnin, momordin, momorcochin-S, PAP, saporin-6, tritin [Carnicelli, D., Brigotti, M., Montanaro, L. and Sperti, S. (1992) Biochem. Biophys. Res. Commun. 182, 579-582], agrostin, luffin, trichokirin and trichosanthin (present study).

Identification of the tRNAs which up-regulate agrostin, barley RIP and PAP-S, three ribosome-inactivating proteins of plant origin.

Ribosome-inactivating proteins (RIP) are RNA-N-glycosidases widely diffused in plants which depurinate ribosomal RNA at a specific universally conserved position, A4324 in rat ribosomes. A small group of RIPs (cofactor-dependent RIPs) require ATP and tRNA to reach maximal activity on isolated ribosomes. The tRNA which stimulates gelonin was identified as tRNA(Trp). The present paper reports the identification of three other tRNAs which stimulate agrostin (tRNA(Ala)), barley RIP (tRNA(Ala), tRNA(Val)) and PAP-S (tRNA(Gly)), while for tritin-S no particular stimulating tRNA emerged. The sequences of tRNA(Val) and tRNA(Gly) correspond to the already known ones (rabbit and man, respectively). The tRNA(Ala) (anticodon IGC) identifies a new isoacceptor. Only the stimulating activity of the tRNA(Ala) for agrostin approaches the specificity previously observed for the couple gelonin-tRNA(Trp).

Primer tRNA(Trp) of RSV-transformed or RAV-1-infected cells up-regulates the antiribosomal activity of gelonin.

Some ribosome-inactivating proteins (RIPs) with RNA-N-glycosidase activity on 28S rRNA require, for maximal inactivation of ribosomes, the presence of tRNA. tRNA(Trp) specifically up-regulates gelonin, the RIP from Gelonium multiflorum. The same tRNA is the primer of the reverse transcriptase of Rous sarcoma virus (RSV) and of its mutant (RAV-1) which lacks the src gene. Here we demonstrate that gelonin is more active in inhibiting endogenous protein synthesis by lysates of RSV-transformed or RAV-1-infected cells and that such increase in activity correlates with the increased amount of primer tRNA(Trp) in the cells.

Partial purification of two proteins which sensitize ribosomes to gelonin: sensitization is not linked to phosphorylation of ribosomal proteins.

Inactivation of ribosomes by gelonin, from Gelonium multiflorum, requires ATP and extraribosomal protein(s) present in the rabbit reticulocyte lysate [SPERTI, S. et al. (1991) Biochem. J. 277, 281-284]. On the anion exchanger Mono Q the activity responsible for the sensitization of ribosomes to gelonin resolves in two peaks which both display a kinase activity on ribosomal proteins. However, staurosporin, an inhibitor of several protein kinases, strongly inhibits phosphorylation of ribosomal proteins without affecting the gelonin-promoting activity of Mono Q peaks. All the evidence collected contradicts a direct link between sensitization to gelonin and phosphorylation of ribosomes.

Shiga toxin 1: damage to DNA in vitro.

Shiga toxins share with plant ribosome-inactivating proteins the same enzymatic mechanism of action: the removal of a specific adenine from 28S RNA when acting on ribosomes and the removal of multiple adenines when acting on DNA in vitro. The activity on DNA, only recently reported, is particularly evident, and has been studied mostly at acidic pH. For the in vitro activity, on both ribosomes and DNA, Shiga toxins require activation by trypsin, urea and dithiothreitol which release the enzymatically active A(1) fragment. Activation by the classical procedure leaves large amounts of urea and DTT which interfere in the DNA depurination assay and completely abolish depurination at physiological pH. A consistent release of [3H]adenine from DNA at neutral pH is instead observed when the toxin is activated in vitro by an improved method which removes most of the drastic reagents required for proteolytic cleavage and reduction. Damage to single-stranded DNA by Shiga toxin 1 (Stx1) primarily involves depurination. A spontaneous DNA breakdown appears in fact only after extensive base removal, a behavior similar to that observed with uracil-DNA glycosylase, a simple glycosylase devoid of lyase activity. NaCl inhibits the activity of Stx1, probably by minimizing the sliding distance traveled by the enzyme along DNA in search of its target sites and promoting dissociation of the substrate-enzyme complex.

The RNA-N-glycosidase activity of Shiga-like toxin I: kinetic parameters of the native and activated toxin.

Shiga toxin and Shiga-like toxins are ribosome-inactivating proteins with RNA-N-glycosidase activity which remove a specific adenine from 28S RNA. The toxins are composed of an A subunit non-covalently associated to a multimer of receptor-binding B subunits. Near the COOH-terminus of the A subunit, a disulfide-bonded loop contains two trypsin-sensitive arginine residues. Proteolytic nicking at these sites, followed by reduction, removes from the A subunit the C-terminal end together with the associated B subunits. The requirement of such cleavage for biological activity of Shiga toxin and Shiga-like toxins has been recently questioned. The present paper reports the kinetic constants of the adenine release from highly purified Artemia salina ribosomes catalysed by Shiga-like toxin I and by its A subunit before and after treatment with trypsin, urea and dithiothreitol or urea and dithiothreitol alone. All reactions had approximately the same Km (1 microM). The Kcat was 0.6 min-1 for the untreated holotoxin and 6 min-1 for the isolated A subunit, respectively. The trypsin treatment increased 1000-fold the Kcat of the holotoxin (770 min-1) and 100-fold the Kcat of the A subunit (640 min-1). The same Kcat (693 min -1) was also observed when the A subunit was treated only with urea and dithiothreitol. Thus the full activity of Shiga-like toxin I required not only removal of the B subunits but also activation of the A subunit itself. Such activation could be largely induced in vitro by drastic loosening of the molecule induced by urea and dithiothreitol, but in vivo would probably require a proteolytic cleavage of the toxin. Inactivation of ribosomes by Shiga-like toxin I did not require sensitization of ribosomes by ATP and macromolecular cofactors present in postribosomal supernatants.

A survey of adenine and 4-aminopyrazolo[3,4-d]pyrimidine (4-APP) as inhibitors of ribosome-inactivating proteins (RIPs).

The inhibitory power of adenine and 4-aminopyrazolo[3,4-d]pyrimidine (4-APP) on the RNA-N-glycosidase activity catalyzed by bacterial (Shiga toxin 1) and plant (ricin, gelonin, momordin, bryodin-R, PAP-S, luffin, trichosantin, saporin 6 and barley) RIPs has been compared. The behavior of the two inhibitors is largely variable. While Shiga toxin 1 is preferentially inhibited by 4-APP, plant RIPs are either preferentially inhibited by adenine, or equally inhibited by the two compounds or, finally, only slightly more by 4-APP. Sequence variabilities involved in these different behaviors are discussed. The experimental data clearly indicate that, in spite of the same mechanism of action, RIPs differ widely in the ability to fit small ring molecules in the active cleft. While the strong inhibitory power of 4-APP on Shiga toxin 1 opens perspectives of therapeutic interventions, the ineffectiveness of the compound on ricin precludes its use as a suitable antidote in poisoning.

Interaction of diphtheria toxin fragment A and of elongation factor 2 with cibacron blue.

Diphtheria toxin fragment A interacts with Cibacron blue in solution, although it is not retained by blue Sepharose columns. Difference spectral titration of fragment A with the dye gives a dissociation constant of the order of 10(-5) M and a 1:1 stoichiometry for the complex. In equilibrium dialysis experiments Cibacron blue behaves as a competitive inhibitor of the binding of NAD to diphtheria toxin fragment A. The dye inhibits in a non-competitive way the fragment A-catalysed transfer of ADP-ribose from NAD to elongation factor 2 (EF2). By affinity chromatography on blue Sepharose a binding of EF2 and of ADP-ribosyl-EF2 with the dye is also demonstrated. GDP, GTP and GDP(CH2)P are able to displace EF2 from blue Sepharose.

Ribosome-bound elongation factor 2 escapes phosphorylation by Ca2+/calmodulin-dependent protein kinase III.

The activity of eukaryotic elongation factor 2 is regulated by phosphorylation catalysed by a highly specific Ca2+/calmodulin-dependent protein kinase. Phosphorylated EF2 binds to ribosomes with decreased affinity. The present evidence indicates that EF2 prebound to ribosomes is protected from phosphorylation, just as earlier evidence indicated that ribosome-bound EF2 is protected from ADP-ribosylation catalysed by diphtheria toxin. Ribosome-inactivating proteins ricin and gelonin, by interfering with the EF2-ribosome interaction, allow full phosphorylation of EF2.

Inactivation of ribosomes by an inhibitor of protein synthesis from Salmonella enteritidis.

Sonic extracts of Salmonella enteritidis contain a factor which inhibits protein synthesis in cell-free systems by irreversibly inactivating ribosomes. The extent of the inactivation of ribosomes depends on the system used to assay protein synthesis, natural mRNA translation being more strongly inhibited than poly(U) translation. The inhibitory power of the Salmonella factor is destroyed by trypsin and by 5% mercaptoethanol. Placental RNase inhibitor is unable to protect ribosomes from inactivation.

A polymerase chain reaction (PCR) method for the identification of collagen adhesin gene (CNA) in Staphylococcus-induced prosthesis infections.

Staphylococci are well-recognized pathogens of foreign body-associated infections. The pathogenesis of such infections involves an initial step of contact between the colonizing bacterium and the biomaterial, with subsequent colony formation. Several studies have been devoted to identify adhesion mechanisms for these bacteria. Slime in particular has been extensively investigated. Recently, considerable attention has been given to the host protein receptors that have been shown in in vitro assays to serve as substrates for bacterial adherence. To determine the importance of the collagen adhesin as virulence factor in Staphylococcus-induced prosthesis infection, a simple and reliable method using a polymerase chain reaction (PCR) was devised to identify collagen adhesin gene (cna). By using lysates of ten strains from orthopedic prostheses (5 Staphylococcus aureus and 5 Staphylococcus epidermidis) and two 20-oligonucleotides as primers, a 192-bp region of the cna gene was amplified by PCR and detected by agarose gel electrophoresis. Results obtained by this method were in accordance with those obtained by the in vitro phenotypic characterization of binding ability to collagen of Staphylococcus strains.

Dyskerin expression influences the level of ribosomal RNA pseudo-uridylation and telomerase RNA component in human breast cancer.

Dyskerin is a nucleolar protein, altered in dyskeratosis congenita, which carries out two separate functions, both fundamental for proliferating cells. One function is the pseudo-uridylation of ribosomal RNA (rRNA) molecules, necessary for their processing, and the other is the stabilization of the telomerase RNA component, necessary for telomerase activity. A significant feature of dyskeratosis congenita is an increased susceptibility to cancer; so far, however, no data have been reported on dyskerin changes in human tumours. Therefore, in this study, the distribution of dyskerin in a large series of human tumours from the lung, breast, and colon, as well as from B-cell lymphomas, was analysed by immunohistochemistry. Dyskerin proved never to be lost or delocalized outside the nucleolus. A quantitative analysis of dyskerin mRNA expression was then performed in 70 breast carcinomas together with the evaluation of telomerase RNA component levels and rRNA pseudo-uridylation. Dyskerin mRNA levels were highly variable and directly associated with both telomerase RNA component levels and rRNA pseudo-uridylation. Dyskerin gene silencing in the MCF-7 human breast carcinoma cell line reduced telomerase activity and rRNA pseudo-uridylation. Significantly, patients with low dyskerin expression were characterized by a better clinical outcome than those with a high dyskerin level. These data indicate that dyskerin is not lost in human cancers and that the levels of its expression and function are associated with tumour progression.

RNA present in post-ribosomal supernatants makes ribosomes susceptible to inactivation by gelonin and alpha-sarcin.

The remarkable resistance of isolated ribosomes to gelonin is overcome by cofactors present in post-ribosomal supernatants. In rat liver post-ribosomal supernatant RNA is the cofactor responsible of the sensitization of ribosomes. Isolated RNA, which consists mostly of deacylated tRNA, accounts for less than 10 per cent of the activity of the original supernatant. The activity of the supernatant is completely destroyed by micrococcal nuclease and RNAase A and also by proteinase K, suggesting that some protein enhances the effect of RNA. RNA has a role also in the sensitization of ribosomes to alpha-sarcin, an RNAase which inactivates ribosomes by hydrolyzing a single phosphodiester bond in the same region of 28S rRNA which is the target of the N-glycosidase activity of gelonin.

Differential requirement of ATP and extra-ribosomal proteins for ribosome inactivation by eight RNA N-glycosidases.

The requirement of ATP and extra-ribosomal proteins for the inactivation of ribosomes by eight plant RNA N-glycosidases [ribosome-inactivating proteins (RIPs)] was investigated. Tritin, pokeweed antiviral protein and barley RIP depend, as gelonin [Sperti, S., Brigotti, M., Zamboni, M., Carnicelli, D. and Montanaro, L. (1991) Biochem. J., 277, 281-284], on the presence of ATP and extra-ribosomal proteins for full inactivation of ribosomes, while bryodin, lychnin, momordin, momorcochin and saporin inactivate isolated Artemia salina ribosomes suspended in buffer saline.

Elongation factor 2 from Artemia salina embryos and its affinity for ribosomes.

Crude extracts from Artemia salina undeveloped embryos do not contain detectable elongation-factor-2 (EF2) kinase and endogenous ADP-ribosylating activities. Accordingly, EF2 purified from this source is an enzyme relatively free from phosphorylated and ADP-ribosylated forms. Endogenous ADP-ribosyltransferase activity appears only after purification of EF2. The affinities of EF2 and of ADP-ribosyl-EF2 for ribosomes from A. salina undeveloped embryos have been calculated by measuring the ability of the factors to inhibit the N-glycosidase activity of ricin on ribosomes.

Oligonucleotides complementary to the alpha-sarcin domain of 28S rRNA inhibit cell-free protein synthesis.

Oligodeoxynucleotides complementary to the alpha-sarcin domain of rat 28S rRNA inhibit cell-free protein synthesis. The poly(U) translation system containing Artemia salina ribosomes was more sensitive to inhibition than the system containing rat liver ribosomes. The 21-mer, which was the most effective of the 7 oligonucleotides tested, hybridized with naked 28S rRNA. Hybridization with whole ribosomes, assayed by S1 nuclease protection, occurred only at high ionic strength or with ribosomes actively engaged in protein synthesis.