Search

Locked nucleic acids (LNAs) are formed by bicyclic ribonucleotides where the O2??? and C4??? atoms are linked through a methylene bridge and the sugar is blocked in a 3???-endo conformation (Singh, S.K. et al., Chem Commun, 1998, 455-6). They represent a promising tool for therapeutic and diagnostic applications, characterized by higher thermal stability and nucleases resistance with respect to their natural counterparts (Doessing, H. and Vester, B., Molecules, 2011, 16, 4511-26; Kaur, H. et al., Chemical reviews, 2007, 107, 4672-97). However, structural descriptions of LNA-containing quadruplexes are rather limited, and up to now no crystal structure was reported. Here we present the first crystallographically derived model of an all LNA-substituted quadruplex forming sequence 5???-TGGGT-3???. Structural details derived from this first high resolution crystallographic model of an all-LNA quadruplex add important information to our knowledge of the effect of modified nucleotides in quadruplex structure and stability. In particular, we observe a regular terminal thymine-tetrad, the first to be observed in crystallographic structure of G-quadruplexes and a likely contributor to the high thermal stability of this molecule. Moreover, the unusual stacked arrangement of the two quadruplexes in the asymmetric unit suggests how sequential quadruplexes can arrange in the crowded cell environment, while at the same time the structure provides an opportunity in the design of LNA-based blocks for developing scaffolds within nanotechnology applications.

Locked nucleic acids (LNAs) are formed by bicyclic ribonucleotides where the O2’ and C4’ atoms are linked through a methylene bridge and the sugar is blocked in a 3’-endo conformation [1]. They represent a promising tool for therapeutic and diagnostic applications, characterized by higher thermal stability and nucleases resistance with respect to their natural counterparts [2,3]. However, structural descriptions of LNA-containing quadruplexes are rather limited, and up to now no crystal structure was reported. Here we present the first crystallographically derived model of an all LNA-substituted quadruplex forming sequence 5’-TGGGT-3’. The presence of two closely packed quadruplexes in the asymmetric unit made structural determination particularly changeling, but eventually a very good model at 1.7 Å resolution has been obtained. Structural details derived from this first high resolution crystallographic model of an all-LNA quadruplex add important information to our knowledge of the effect of modified nucleotides in quadruplex structure and stability. Moreover, the unusual stacked arrangement of the two quadruplexes in the asymmetric unit suggests how sequential quadruplexes can arrange in the crowded cell environment, while at the same time the structure provides an opportunity in the design of LNA-based blocks for developing scaffolds within nanotechnology applications. [1] S.K. Singh, P. Nielsen, A.A. Koshkin, J. Wengel Chem Commun, 1998, 455-456. [2] H. Doessing, B. Vester Molecules, 2011,16, 4511-4526. [3] H. Kaur, B.R. Babu, S. Maiti Chemical reviews, 2007,107, 4672-4697.

Human α-thrombin is a multifunctional serine protease that plays a pivotal role in hemostasis. Two binding sites on its surface, the fibrinogen recognition site (exosite I) and the heparin binding site (exosite II), control its catalytic function. Promising thrombin inhibitors are aptamers, short single-stranded oligonucleotides able to bind target proteins with high selectivity and affinity (Bock et al, Nature, 1992, 355, 564-6). Among them a 15-mer aptamer recognizing exosite I (TBA) (Macaya et al, PNAS, 1993, 90, 3745-9) and a 29-mer binding exosite II (HD22) aptamer (Tasset et al, JMB, 1997, 272, 688-98) were selected for their high affinity towards thrombin. Structural data on thrombin-aptamer complexes are essential to understand recognition mechanism(s) of TBA and HD22 and to design novel aptamers with improved pharmaceutical properties. We have performed an X-ray structural characterization of the complexes of thrombin with TBA (Russo Krauss et al, NAR, 2012, 40, 8119-28), TBA variants (Russo Krauss et al, NAR, 2011, 39, 7858-67) and mutants or a shorter variant of HD22 (HD22-27mer). This study has been complemented with a CD analysis of aptamers both free and in complex with thrombin, in the presence of Na+ or K+. All together our results have provided a clear picture of the recognition process between thrombin and its aptamers. Moreover, they allow to define the effect that ions have on structure and flexibility, and therefore on thrombin affinity and inhibition, of the various aptamers. Notably, the thrombin-HD22-27mer complex has provided the first crystallographic model of an aptamer with a mixed duplex-quadruplex fold and reveals unexpected novel structural features.

Aptamers are single-stranded DNA or RNA sequences able to recognize a wide variety of natural and synthetic targets including amino acids, peptides, proteins and even cells with high affinity and specificity [1]. They are considered a sort of nucleic acid version of antibodies with improved properties. Aptamers display a large number of structural arrangements, in particular the interesting G-quadruplex architecture that is adopted by many of them. Several cases have been also reported where aptamers are supposed to adopt a mixed duplex/quadruplex structure [1,2]. Interestingly, in every case, the presence of both structural motifs is needed for strong and specific binding of the target. Also potent second-generation thrombin aptamers adopt a duplex/quadruplex bimodular folding [2]. They have been proved to recognize thrombin exosite II [2], whereas best studied thrombin binding aptamers bind exosite I [3,4], and show very high affinity and specificity towards thrombin [2]. A sound model of these oligonucleotides, either free or in complex with thrombin, is not yet available. Here we present a structural study of one of these aptamers, HD22-27mer. Interestingly, the crystal structure of the aptamer in complex with thrombin displays a novel architecture, in which a double helical stem is enchained to a pseudo-G-quadruplex. Our results also underline the role of residues that join the duplex and the quadruplex motifs and control their recruitment in thrombin binding. Finally, our crystallographic model shed light on the contrasting circular dichroism studies reported in the literature [5], suggesting a major role of ions in determining the aptamer properties. [1] B. Gatto, M. Palumbo, C. Sissi Current Medicinal Chemistry, 2009, 16, 1248-1265; [2] D.M. Tasset, M.F. Kubik, W. Steiner Journal of Molecular Biology, 1997, 272(5), 688-698;. [3] I. Russo Krauss, A. Merlino, C. Giancola, A. Randazzo, L. Mazzarella, F. Sica Nucleic Acids Res., 2011, 39(17), 7858–7867 [4] I. Russo Krauss, A. Merlino, A. Randazzo, E. Novellino, L. Mazzarella, F. Sica Nucleic Acids Res., 2012, 40(16), 8119-28 [5] G. Marson, M. Palumbo, C. Sissi Curr Pharm Des., 2012, 18(14), 2027-2035

The aim of anticoagulant/antithrombotic therapy in cardiovascular disease is to prevent fibrin deposition and platelet aggregation, halting current and future ischemic episodes. The limitations to the effectiveness of the most commonly used agents (heparin, coumadin and aspirin) have led to the development of new anticoagulant compounds. In particular, aptamers represent an attractive approach because of their high specificity and low immunogenicity. The best known example is the Thrombin Binding Aptamer (TBA), namely 5‘GGTTGGTGT-GGTTGG3’[1]. TBA and its derivatives [2] adopt a G-quadruplex structure [2-6] and inhibit thrombin activity by blocking the fibrinogen binding site (exosite I) [3-6]. Biological properties of TBA are strictly dependent on its tertiary structure. Other antithrombotic aptamers have been identified using the SELEX process and partially characterized. Among them HD22, namely 5’GTCCGTGGTAGGGCAGGTTGGGGTGAC3’, is particularly interesting. It presents a 15-nucleotide core sequence that has striking similarity to TBA, and it has been reported to adopt a mixed duplex/quadruplex structure [7]. Remarkably HD22 binds thrombin with much higher affinity than TBA and has been shown to bind exosite II instead of exosite I [7]. No structural data on thrombin-HD22 complex has been reported so far. We have solved the X-ray structure of the thrombin-HD22 complex, with the aim to understand the molecular details of the interaction between the two molecules and to investigate the differences with respect to thrombin-TBA complex, whose structure we have recently determined at high resolution. These results could help the design of a new class of aptamers with an improved capability to modulate thrombin function. [1] L.C. Bock, L.C. Griffin, J.A. Latham, E.H. Vermaas, J.J. Toole, Nature, 355, 1992, 564. [2] L. Martino, A. Virno, A. Randazzo, A. Virgilio, V. Esposito, C. Giancola, M. Bucci, G. Cirino, L. Mayol, NAR, 34, 2006, 6653. [3] R.F. Macaya, P. Schultze, F.W. Smith, J.A. Roe, J. Feigon, PNAS USA, 90, 1993, 3745. [4] K. Padmanabhan, A. Tulinsky, Acta Crystallogr. D, 52, 1996, 272. [5] I. Russo Krauss, A. Merlino, A. Randazzo, L. Mazzarella, F. Sica, Acta Crystallogr F, 66, 2010, 961. [6] I. Russo Krauss, A. Merlino, C. Giancola, A. Randazzo, L. Mazzarella, F. Sica NAR, submitted [7] D.M. Tasset, M.F. Kubik, W. Steiner, JMB, 272(5), 1997, 688.

Glutathione (GSH) is the major antioxidant molecule in most living organisms. Besides its protective role in contrasting oxidative free radical species and in the removal of toxic metals, GSH is involved in redox homeostasis and cell signaling; furthermore, it is essential to maintain cellular proteins in their reduced state and acts as a post-translational regulator of protein function. The synthesis of GSH from its constituent amino acids involves two ATP-requiring enzymatic steps catalyzed by glutamate cysteine ligase (GshA) and glutathione synthetase (GshB), respectively. Its biosynthetic route has been described in details for a variety of organisms; however, nothing is known about this process in psychrophilic microorganisms. In particular, although various GshBs have been purified and sequenced, none of these enzymes belong to an organism living at low temperatures. Very recently, a recombinant form of GshB from the cold-adapted bacterium Pseudoalteromonas haloplanktis (PhGshB) has been purified. The activity of PhGshB was assayed through a direct method, measuring the release of inorganic phosphate from radiolabelled ATP. PhGshB is active at 10°C and its activity significantly increases with temperature, at least up to 30°C; the Km for ATP ranges between 0.14 and 0.25 mM in the 10 – 30°C interval. The crystal structure of PhGshB has been determined and refined at 2.3 Ǻ resolution. Structural information is combined with a characterization of the thermal stability of the enzyme performed by circular dichroism and fluorescence measurements. A comparison between our data and those obtained for the closely related GshB from the mesophilic Escherichia coli [1-3], sharing 69% identity with PhGshB, is presented in the attempt to understand the mechanisms of cold-adaptation of the psychrophilic enzyme.

Recently, the superfamily of animal, extracellular, pyrimidine-specific RNases, often called the RNase A superfamily, has been shown to include not only tetrapod enzymes, but also fish enzymes [1]. In particular, five RNases from zebrafish (Danio rerio) [1-3] and two from the Atlantic salmon (Salmo salar) [4] have been reported to have a very low RNase activity and to be endowed, like RNase 5 (human angiogenin), with powerful angiogenic activity. We have determined the X-ray structure of two zebrafish RNases [3]. In these proteins, like in human angiogenin, the putative binding subsite B1 of the pyrimidine base is partially obstructed by the side chain of Glu located in the C-terminal segment of the protein, and this structural feature well account for their low catalytic activity. More recently, the crystal structure of RNase-2 from Salmo salar (Ss2) has been also determined. Surprisingly, within an essentially unmodified RNase folding, the enzyme presents an extensive reorganization of the active site region with respect to other pancreatic RNases. In particular, although it has the highest catalytic activity among fish RNases, it presents an active site fully obstructed by a peptide segment at C-terminal region (CTR), and with the two catalytic histidines in direct contact. Thus the enzyme appears to be auto-inhibited in a completely different manner compared to the other angiogenins. Comparison of the structure of Ss2 with those of RNase complexes with substrate analogs suggests that Ss2 could adopt two distinct conformations: a closed form with the CTR blocking the substrate binding cleft (observed in the crystal structure) and an open conformation, where the CTR swings out forming an open cleft with the active site exposed. Overall, these data provide novel structural insights into the mechanism that modulates RNase activity of angiogenins. [1] E. Pizzo, P. Buonanno, A. Di Maro, S. Ponticelli, S. De Falco, N. Quarto, M.V. Cubellis and G. D'Alessio, J Biol Chem, 281, 2006, 27454. [2] S. Cho and J. Zhang, Mol Biol Evol, 24, 2007, 1259. [3] E. Pizzo, A. Merlino, M. Turano, I. Russo Krauss, F. Coscia, A. Zanfardino, M. Varcamonti, A. Furia, C. Giancola, L. Mazzarella, F. Sica, G. D'Alessio, Biochem J., 433(2), 2010, 345 [4] E. Pizzo, M. Varcamonti, A. Di Maro, A. Zanfardino, C. Giancola, G. D'Alessio, FEBS J, 275, 2008, 1283.

The stability of the folded structure of a protein molecule is generally limited to a narrow range of environmental parameters. Depending on temperature, pH and types of co-solutes and on the evolutionary history of the protein itself, the tertiary structure may become less tightly bound and undergo to a more or less extensive unfolding that can lead to either unspecific aggregation or ordered adducts with other molecules or even to oligomeric aggregates endorsed with new functions. In the same way, specific sequence modification in hot points of the protein may lead to an “open” monomeric species which can interact in a complementary fashion with a similarly opened molecule (domain swapping) to produce a thermodynamic stable or kinetically trapped transient dimeric form (swapped dimer)[1]. The dimer is stabilized by the pre-existing surfaces of the swapped domains, whereas the flexible loops connecting these domains give rise to a somewhat ‘destructured quaternary assembly’ of the chains that strongly increases the potentiality for higher aggregation [2]. Bovine pancreatic ribonuclease (RNase A) is able to swap both the C-terminal - strand (residues 116-124) and the N-terminal -helix (residues 1-13), thus forming two different dimers [3-4]. Cis-trans isomerization of the Asn113-Pro114 peptide group was observed when the protein forms the C-terminal swapped dimer [4]. To study the effect of the substitution of Pro114 on the swapping process of RNase A, we have prepared and characterized the P114A monomeric and dimeric variants of the enzyme. Surprisingly, the mutation not only affects the swapping of the C-terminal b-strand but also that of the N-terminal helix and the stability of the two swapped dimers. These results, which are discussed on the bases of the X-ray structures of the monomeric and C-terminal swapped dimeric forms of P114A and on the available biochemical and structural data on RNase A dimers, reveal a cross-talk between the chain termini in the swapping process of RNase A.

The ability to retain ribonucleolytic activity in the presence of the ribonuclease inhibitor (RI) is one of the determinant of ribonucleases (RNases) cytotoxicity. Bovine seminal ribonuclease (BSRNase), the only dimeric protein among the pancreatic-like ribonucleases, shows an high selective toxicity to tumor cells1. The native enzyme is an equilibrium mixture of two forms, with (MxM) or without (M=M) swapping of the N-terminal arms2-4. The molecular envelope and the exposed surface of the two isomers are practically indistinguishable and their diversity is almost completely buried in the interior of the protein5. The cytotoxic and antitumor activity of the enzyme is a peculiar property of the swapped dimer1. Indeed this form, despite the breakage of the interchain disulfide bonds, likely to occur in the cytosolic compartment, generates a metastable dimer (NCDBS) with a compact structure that allows the molecule to escape RI6. To generate RNases suitable for antitumor treatment in humans we have dimerized two low immunogenic proteins, bovine pancreatic RNase (RNase A) and its human homologue (HP-RNase). To achieve this, two different strategies have been used. First, we have introduced few selected structural features of the seminal enzyme in these RNases. Differently from NCD-BS, the crystal structure analysis of the NCD form of the RNase A mutants has revealed that they adopt an opened quaternary structure, which can be easily sequestered by RI7. All the dimeric variants, except one, do exhibit a cytotoxic activity at a much lower level than BS-RNase. On the contrary, the dimeric HP-RNase variant (HHP2-RNase), in which the putative swapping determinants of the seminal enzyme have been introduced, was found to be extremely cytotoxic8. Surprisingly, this activity is associated also with the unswapped form of the protein. The crystal structure of this mutant has provided a reasonable explanation for this result9. Secondly, in order to force the swapping in HP-RNase, we have deleted five residues in the loop connecting the N-terminal helix to the core of the protein (desHP-RNase). This last variant is enzymatically active10 and has a very low cytotoxic action. The crystal structure of desHP-RNase reveals a tetrameric association of two swapped dimers with a quaternary assembly compatible with RI binding. Interestingly, the analysis of packing interactions suggests a pathway of large-scale oligomerization. On this basis, fibril formation was predicted and observed.

Recently, four RNases from zebrafish (Danio rerio) have been identified and studied (ZF-RNase-1, -2, -3 and -4) [1-3]. Kazakou et al. [2], after an extensive survey of zebrafish DNA sequences, which indicated widespread polymorphism among the known ZF-RNase homologues, determined the X-ray structure of ZF-RNase-1 and -3 at basic pH. They also describe a polymorphic variant of the gene encoding ZF-RNase-2. More recently, we have proved that this variant is indeed a cell product by itself [4]. Here we report the X-ray structure of ZF-RNase-5 and that of ZF-RNase-1 at acidic and neutral pH. The structures have been directly compared with those described by Kazakou et al. [2] and with the structures of ZF-RNase-2 and -4, which have been obtained by homology modelling (Fig.1). The results have allowed to explain some differences in the catalytic activity of these proteins [4].

Pancreatic ribonucleases, thanks to their ability to degrade RNA, represent potential antitumor therapeutics. It has been suggested that RNases have to reach the cytosol and to avoid the binding of the ribonuclease inhibitor (RI) in order to exert their cytotoxic activity1. Among antitumor RNases, bovine seminal ribonuclease (BS-RNase) is an interesting case: it is a covalent homodimer with two interchain antiparallel disulphide bridges between Cys31 and Cys322-3, which in the reducing cellular compartment of cytosol convert into a non covalent dimeric form (NCD) stabilized by the swapping of N-terminal helices4. It has been suggested that NCD-BS avoids inhibition by RI because it adopts a compact quaternary structure that interferes with binding [4]. Structural studies have indicated a key role in determining this dimeric assembly for specific amino acid residues: Pro19 and Leu285. Using the homology mutation approach, few structural features of BS-RNase have been introduced in bovine pancreatic ribonuclease (RNase A), which does not show antitumor action. The introduction of Leu28, Cys31 and Cys32 (LCC) and that of the same residues plus Pro19 (PLCC) or the introduction of the hinge peptide sequence (residues 16-22) of BS-RNase and the two cysteines (GNPSCC) in the sequence of RNase A produced dimeric variants which do exhibit a cytotoxic activity, though at a much lower level than BS-RNase6-7. Structural determination of NCD forms of these mutants has allowed to rationalyze their biological activities: all the dimers have open quaternary structures, which can be easily sequestered by RI. The comparison of these structures with those previously reported for other RNase swapped dimers has suggested that an additional important role in determining the compact quaternary structure of NCD-BS can be played also by Gly167. Therefore, the GPLCC mutant of RNase A (with substitution S16G, A19P, Q28L, R31C and S32C) has been produced and found to exert a cytotoxic activity on SVT2 cells comparable to that of seminal enzyme. In order to verify whether NCDGPLCC has a quaternary assembly similar to NCD-BS, its structure has been determined from two different crystal forms obtained from similar crystallization conditions. Surprisingly, NCD-GPLCC shows two different quaternary structures, both of them distinct with respect to NCD-BS. As predicted7, the introduction of Gly16 determines a hinge peptide conformation similar to that observed in NCD-BS. However, in each structure this occurs only for one of the two hinge peptides, and the resulting quaternary assembly is still compatible with the RI binding. Altogether these data represent an encouraging step toward a rational design of a dimeric form of RNase A endowed with a strong cytotoxic activity and point up the necessity of a deeper characterization of its complex with RI, in terms of binding constants and residual ribonuclease activity.

The Protein Crystallography world: role of Prof. L. Mazzarellaq

Editoriale Biopolymers Special Issue dedicated to Prof. Lelio Mazzarella Editors Giulietta Smulevich & Filomena Sica Volume 91, Issue 12, December 2009, Page975

Cellular imaging techniques based on vibrational spectroscopy (Raman) are powerful tools in cell biology because the molecular composition of sub-cellular compartments can be visualized without the need for labeling [1]. Moreover resonance Raman (RR) spectroscopy is an ideal method to elucidate the structural characteristics of heme domains in hemoproteins, directly monitoring the iron-axial ligand bond strength. The information provided is important in understanding the mechanisms of protein control of heme reactivity [2]. In this context, Optical Tweezers (OT) have revealed a formidable tool to select and manipulate single particles, included cells or cellular organelles, allowing a rapid identification of the changes, at single cell level, in response to external agents, environmental stress, or in presence of cellular disorders [3]. Haemoglobinopathies are inherited disorders defined as qualitative or quantitative deficiency of haemoglobin (Hb) in red blood cells (RBCs). Herein we analyze anemic RBCs with two human Hb variants, HbS and HbC (Gluβ6Val and GlubβLys, respectively). A comparative study in a family (mother, father and daughter, respectively heterozygous for HbS, HbC and Hb C+S) shows that different Raman images and mechanical properties of healthy and anemic RBCs could help to evaluate the expression ratios between HbS and HbC, and correlate these data with varied clinical conditions. References [1]. van Manen H.J., et al., Proc. Natl. Acad. Sci. 2005, 102, 10159. [2]. Vitagliano L., et al., J. Am. Chem. Soc. 2008, 130, 10527. [3]. De Luca A.C., et al., Opt. Exp. 2008, 16, 7943. AV and LM acknowledge PRIN 2007 funds.

The aggregation of intracellular, normally soluble, proteins into insoluble fibers has been now seen as the common cause of many diseases. Neverthless, it is known that some proteins form fibers that are non-toxic and even functionally relevant, whereas others form toxic aggregates without forming fibers. The structures of all these protein aggregates is difficult to obtain by conventional structural methods such as X-ray crystallography, solution NMR and electron microscopy. Recently, to gain insight into the structural chainges that take place when proteins form fibrils, D. Eisenberg and his coworkers designed a bovine pancreatic ribonuclease (RNase A) variant that is able to form amyloid-like fibrils that contains 3D domain swapped molecules capable of enzymatic activity. Along this line, by deleting five residues in the hinge loop linking the N-terminal a-helix of human pancreatic RNase (HP-RNase) to the rest of the protein, we found that HP-RNase is able to form domain swapped dimers. X-ray structure analysis has revealed a tetrameric association of two swapped HP-RNase dimers, which have a different quaternary structure. The analysis of packing interactions in this structure suggests a pathway of large-scale oligomerization of the protein. On the basis of the observed supramolecular assembly, fibril formation was predicted. This expectation was confirmed by the observation of the protein aggregates by electron microscopy. The biophysical and biochemical characterization of those fibrils is now in progress. These data provide an interesting example in which a single deletion of a few residues can be the primary event irreversibly leading to a monomer to dimer transition by domain swapping and to protein aggregation by fibril formation.

X-ray crystallography is the most accurate physico-chemical method for determining the 3D structure of proteins. This method can be used also to provide important information about the dynamic aspects of a molecular structure, such as thermal displacement parameters or dynamic disordering effects. In addition, in this paper we show that low frequency motion can also be analyzed by this technique, provided that multiple structural models of the same protein, crystallized in different crystal environments, are available. For bovine pancreatic ribonuclease (RNase A), a large number of crystallographic studies have been carried out on crystals obtained in a variety of experimental conditions. These structural data were used to map the low-frequency β-sheet closure motion of this protein, which is considered to be important in determining its enzymatic activity. The force constant and the frequency derived from the X-ray crystallographic data are in good agreement with the values obtained from other experimental techniques.

Bovine pancreatic ribonuclease (RNase A) has been converted to a fibrillar form by expanding, with an insertion of ten glutamines, the peptide connecting the core domain of the protein with the C-terminal b-strand [1]. The designed amyloid-like fiber of RNase A contains 3D domain swapped molecules capable of enzymatic activity, thus supporting the amyloid-like fiber model in which a cross-b spine is decorated with 3D domain swapping. By deleting five aminoacid residues in the hinge loop linking the N-terminal helix of human pancreatic RNase (HP-RNase) to the rest of the protein, we found that HP-RNase is able to form domain swapped dimers. These dimers cannot be dissociated by heating at 70 degrees C, and have a great tendency to aggregate. X-ray structure analysis of the HP-RNase dimer reveals that the molecules give rise to a double-helical arrangement that is stabilized by the exchange of N-terminal fragments between subunits belonging to different helices. The observed supra-molecular assembly suggests that the cross b-spine could be not the only way to produce fibres.

Frataxin is a protein responsible for the Friedreich ataxia. Frataxin is involved in iron homeostasis and in iron-sulfur assembly. It has been suggested that frataxin acts as an iron chaperone in the monomeric form and as iron storage in the oligomeric state. We present a structural study of metal ion iteraction with this protein.

Bovine seminal ribonuclease (BS-RNase) is a covalent homodimeric enzyme homologous to pancreatic ribonuclease (RNase A), endowed with a number of special biological functions. It is isolated as an equilibrium mixture of swapped (MxM) and unswapped (M=M) dimers. The interchanged N termini are hinged on the main bodies through the peptide 16-22, which changes conformation in the two isomers. At variance with other proteins, domain swapping in BS-RNase involves two dimers having a similar and highly constrained quaternary association, mainly dictated by two interchain disulfide bonds. This provides the opportunity to study the intrinsic ability to swap as a function of the hinge sequence, without additional effects arising from dissociation or quaternary structure modifications. Two variants, having Pro19 or the whole sequence of the hinge replaced by the corresponding residues of RNase A, show equilibrium and kinetic parameters of the swapping similar to those of the parent protein. In comparison, the x-ray structures of MxM indicate, within a substantial constancy of the quaternary association, a greater mobility of the hinge residues. The relative insensitivity of the swapping tendency to the substitutions in the hinge region, and in particular to the replacement of Pro19 by Ala, contrasts with the results obtained for other swapped proteins and can be rationalized in terms of the unique features of the seminal enzyme. Moreover, the results indirectly lend credit to the hypothesis that the major role of Pro19 resides in directing the assembly of the non-covalent dimer, the species produced by selective reduction of the interchain disulfides and considered responsible for the special biological functions of BS-RNase.

Onconase (ONC), a member of the RNase A superfamily extracted from oocytes of Rana pipiens is used in treatment of various forms of cancer. ONC antitumor properties depend on its ribonucleolytic activity that is three-five order of magnitude lower than RNase A. The structural reasons for this very low ribonucleolityc activity are not yet clearly understood. The most damaging side effect from ONC treatment is renal toxicity, probably linked to the enzyme stability, which is unusually high for a protein isolated from a mesophilic source. In this context, we have prepared and determined the crystal structures of two ONC mutants (M23L and C87S,des103- 104), and performed molecular dynamics simulations of ONC and C87S,des103-104 with the aim of explaining on structural grounds the modifications of the activity and thermal stability of these mutants. Despite the strict similarity in the β-sheet architecture, ONC does not possess the β-sheet breathing motion characteristic of other RNaselike molecules and considered to be functionally important. The decreased flexibility provides a basis to explain the low affinity of ONC towards nucleotides and, more generally, its lower catalytic activity. The results also suggest the basis of the unusually high thermal stability of the enzyme.