Elucidation of the ribonuclease A aggregation process mediated by 3D domain swapping: a computational approach reveals possible new multimeric structures

Human RNase 1 can extensively oligomerize through 3D domain swapping thanks to the crucial contribution of its C-terminus

Human ribonuclease (RNase) 1 and bovine RNase A are the proto-types of the secretory "pancreatic-type" (pt)-RNase super-family. RNase A can oligomerize through the 3D domain swapping (DS) mechanism upon acetic acid (HAc) lyophilisation, producing enzymatically active oligomeric conformers by swapping both N- and C-termini. Also some RNase 1 mutants were found to self-associate through 3D-DS, however forming only N-swapped dimers. Notably, enzymatically active dimers and larger oligomers of wt-RNase 1 were collected here, in higher amount than RNase A, from HAc lyophilisation. In particular, RNase 1 self-associates through the 3D-DS of its N-terminus and, at a higher extent, of the C-terminus. Since RNase 1 is four-residues longer than RNase A, we further analyzed its oligomerization tendency in a mutant lacking the last four residues. The C-terminus role has been investigated also in amphibian onconase (ONC®), a pt-RNase that can form only a N-swapped dimer, since its C-terminus, that is three-residues longer than RNase A, is locked by a disulfide bond. While ONC mutants designed to unlock or cut this constraint were almost unable to dimerize, the RNase 1 mutant self-associated at a higher extent than the wt, suggesting a specific role of the C-terminus in the oligomerization of different RNases. Overall, RNase 1 reaches here the highest ability, among pt-RNases, to extensively self-associate through 3D-DS, paving the way for new investigations on the structural and biological properties of its oligomers.

Slow Evolution toward “Super-Aggregation” of the Oligomers Formed through the Swapping of RNase A N-Termini: A Wish for Amyloidosis?

Natively monomeric RNase A can oligomerize upon lyophilization from 40% acetic acid solutions or when it is heated at high concentrations in various solvents. In this way, it produces many dimeric or oligomeric conformers through the three-dimensional domain swapping (3D-DS) mechanism involving both RNase A N- or/and C-termini. Here, we found many of these oligomers evolving toward not negligible amounts of large derivatives after being stored for up to 15 months at 4 °C in phosphate buffer. We call these species super-aggregates (SAs). Notably, SAs do not originate from native RNase A monomer or from oligomers characterized by the exclusive presence of the C-terminus swapping of the enzyme subunits as well. Instead, the swapping of at least two subunits’ N-termini is mandatory to produce them. Through immunoblotting, SAs are confirmed to derive from RNase A even if they retain only low ribonucleolytic activity. Then, their interaction registered with Thioflavin-T (ThT), in addition to TEM analyses, indicate SAs are large and circular but not “amyloid-like” derivatives. This confirms that RNase A acts as an “auto-chaperone”, although it displays many amyloid-prone short segments, including the 16–22 loop included in its N-terminus. Therefore, we hypothesize the opening of RNase A N-terminus, and hence its oligomerization through 3D-DS, may represent a preliminary step favoring massive RNase A aggregation. Interestingly, this process is slow and requires low temperatures to limit the concomitant oligomers’ dissociation to the native monomer. These data and the hypothesis proposed are discussed in the light of protein aggregation in general, and of possible future applications to contrast amyloidosis.

Modeling and Structure Determination of Homo-Oligomeric Proteins: An Overview of Challenges and Current Approaches

Protein homo-oligomerization is a very common phenomenon, and approximately half of proteins form homo-oligomeric assemblies composed of identical subunits. The vast majority of such assemblies possess internal symmetry which can be either exploited to help or poses challenges during structure determination. Moreover, aspects of symmetry are critical in the modeling of protein homo-oligomers either by docking or by homology-based approaches. Here, we first provide a brief overview of the nature of protein homo-oligomerization. Next, we describe how the symmetry of homo-oligomers is addressed by crystallographic and non-crystallographic symmetry operations, and how biologically relevant intermolecular interactions can be deciphered from the ordered array of molecules within protein crystals. Additionally, we describe the most important aspects of protein homo-oligomerization in structure determination by NMR. Finally, we give an overview of approaches aimed at modeling homo-oligomers using computational methods that specifically address their internal symmetry and allow the incorporation of other experimental data as spatial restraints to achieve higher model reliability.

RNase A Domain-Swapped Dimers Produced Through Different Methods: Structure-Catalytic Properties and Antitumor Activity

Upon oligomerization, RNase A can acquire important properties, such as cytotoxicity against leukemic cells. When lyophilized from 40% acetic acid solutions, the enzyme self-associates through the so-called three-dimensional domain swapping (3D-DS) mechanism involving both N- and/or C-terminals. The same species are formed if the enzyme is subjected to thermal incubation in various solvents, especially in 40% ethanol. We evaluated here if significant structural modifications might occur in RNase A N- or C-swapped dimers and/or in the residual monomer(s), as a function of the oligomerization protocol applied. We detected that the monomer activity vs. ss-RNA was partly affected by both protocols, although the protein does not suffer spectroscopic alterations. Instead, the two N-swapped dimers showed differences in the fluorescence emission spectra but almost identical enzymatic activities, while the C-swapped dimers displayed slightly different activities vs. both ss- or ds-RNA substrates together with not negligible fluorescence emission alterations within each other. Besides these results, we also discuss the reasons justifying the different relative enzymatic activities displayed by the N-dimers and C-dimers. Last, similarly with data previously registered in a mouse model, we found that both dimeric species significantly decrease human melanoma A375 cell viability, while only N-dimers reduce human melanoma MeWo cell growth.

Biological Activities of Secretory RNases: Focus on Their Oligomerization to Design Antitumor Drugs

Ribonucleases (RNases) are a large number of enzymes gathered into different bacterial or eukaryotic superfamilies. Bovine pancreatic RNase A, bovine seminal BS-RNase, human pancreatic RNase 1, angiogenin (RNase 5), and amphibian onconase belong to the pancreatic type superfamily, while binase and barnase are in the bacterial RNase N1/T1 family. In physiological conditions, most RNases secreted in the extracellular space counteract the undesired effects of extracellular RNAs and become protective against infections. Instead, if they enter the cell, RNases can digest intracellular RNAs, becoming cytotoxic and having advantageous effects against malignant cells. Their biological activities have been investigated either in vitro, toward a number of different cancer cell lines, or in some cases in vivo to test their potential therapeutic use. However, immunogenicity or other undesired effects have sometimes been associated with their action. Nevertheless, the use of RNases in therapy remains an appealing strategy against some still incurable tumors, such as mesothelioma, melanoma, or pancreatic cancer. The RNase inhibitor (RI) present inside almost all cells is the most efficacious sentry to counteract the ribonucleolytic action against intracellular RNAs because it forms a tight, irreversible and enzymatically inactive complex with many monomeric RNases. Therefore, dimerization or multimerization could represent a useful strategy for RNases to exert a remarkable cytotoxic activity by evading the interaction with RI by steric hindrance. Indeed, the majority of the mentioned RNases can hetero-dimerize with antibody derivatives, or even homo-dimerize or multimerize, spontaneously or artificially. This can occur through weak interactions or upon introducing covalent bonds. Immuno-RNases, in particular, are fusion proteins representing promising drugs by combining high target specificity with easy delivery in tumors. The results concerning the biological features of many RNases reported in the literature are described and discussed in this review. Furthermore, the activities displayed by some RNases forming oligomeric complexes, the mechanisms driving toward these supramolecular structures, and the biological rebounds connected are analyzed. These aspects are offered with the perspective to suggest possible efficacious therapeutic applications for RNases oligomeric derivatives that could contemporarily lack, or strongly reduce, immunogenicity and other undesired side-effects.

Onconase dimerization through 3D domain swapping: structural investigations and increase in the apoptotic effect in cancer cells

Onconase® (ONC), a protein extracted from the oocytes of the Rana pipiens frog, is a monomeric member of the secretory 'pancreatic-type' RNase superfamily. Interestingly, ONC is the only monomeric ribonuclease endowed with a high cytotoxic activity. In contrast with other monomeric RNases, ONC displays a high cytotoxic activity. In this work, we found that ONC spontaneously forms dimeric traces and that the dimer amount increases about four times after lyophilization from acetic acid solutions. Differently from RNase A (bovine pancreatic ribonuclease) and the bovine seminal ribonuclease, which produce N- and C-terminal domain-swapped conformers, ONC forms only one dimer, here named ONC-D. Cross-linking with divinylsulfone reveals that this dimer forms through the three-dimensional domain swapping of its N-termini, being the C-terminus blocked by a disulfide bond. Also, a homology model is proposed for ONC-D, starting from the well-known structure of RNase A N-swapped dimer and taking into account the results obtained from spectroscopic and stability analyses. Finally, we show that ONC is more cytotoxic and exerts a higher apoptotic effect in its dimeric rather than in its monomeric form, either when administered alone or when accompanied by the chemotherapeutic drug gemcitabine. These results suggest new promising implications in cancer treatment.

Extensive deamidation of RNase A inhibits its oligomerization through 3D domain swapping

Bovine pancreatic ribonuclease A (RNase A) is the monomeric prototype of the so-called secretory 'pancreatic-type' RNase super-family. Like the naturally domain-swapped dimeric bovine seminal variant, BS-RNase, and its glycosylated RNase B isoform, RNase A forms N- and C-terminal 3D domain-swapped oligomers after lyophilization from acid solutions, or if subjected to thermal denaturation at high protein concentration. All mentioned RNases can undergo deamidation at Asn67, forming Asp or isoAsp derivatives that modify the protein net charge and consequently its enzymatic activity. In addition, deamidation slightly affects RNase B self-association through the 3D domain swapping (3D-DS) mechanism. We report here the influence of extensive deamidation on RNase A tendency to oligomerize through 3D-DS. In particular, deamidation of Asn67 alone slightly decreases the propensity of the protein to oligomerize, with the Asp derivative being less affected than the isoAsp one. Contrarily, the additional Asp and/or isoAsp conversion of residues other than N67 almost nullifies RNase A oligomerization capability. In addition, Gln deamidation, although less kinetically favorable, may affect RNase A self-association. Using 2D and 3D NMR we identified the Asn/Gln residues most prone to undergo deamidation. Together with CD spectroscopy, NMR also indicates that poly-deamidated RNase A generally maintains its native tertiary structure. Again, we investigated in silico the effect of the residues undergoing deamidation on RNase A dimers structures. Finally, the effect of deamidation on RNase A oligomerization is discussed in comparison with studies on deamidation-prone proteins involved in amyloid formation.

Bovine seminal ribonuclease triggers Beclin1-mediated autophagic cell death in pancreatic cancer cells

Among the large number of variants belonging to the pancreatic-type secretory ribonuclease (RNase) superfamily, bovine pancreatic ribonuclease (RNase A) is the proto-type and bovine seminal RNase (BS-RNase) represents the unique natively dimeric member. In the present manuscript, we evaluate the anti-tumoral property of these RNases in pancreatic adenocarcinoma cell lines and in nontumorigenic cells as normal control. We demonstrate that BS-RNase stimulates a strong anti-proliferative and pro-apoptotic effect in cancer cells, while RNase A is largely ineffective. Notably, we reveal for the first time that BS-RNase triggers Beclin1-mediated autophagic cancer cell death, providing evidences that high proliferation rate of cancer cells may render them more susceptible to autophagy by BS-RNase treatment. Notably, to improve the autophagic response of cancer cells to BS-RNase we used two different strategies: the more basic (as compared to WT enzyme) G38K mutant of BS-RNase, known to interact more strongly than wt with the acidic membrane of cancer cells, or BS-RNase oligomerization (tetramerization or formation of larger oligomers). Both mutant BS-RNase and BS-RNase oligomers potentiated autophagic cell death as compared to WT native dimer of BS-RNase, while the various RNase A oligomers remained completely ineffective. Altogether, our results shed more light on the mechanisms lying at the basis of BS-RNase antiproliferative effect in cancer cells, and support its potential use to develop new anti-cancer strategies.

The multiple forms of bovine seminal ribonuclease: structure and stability of a C-terminal swapped dimer

Bovine seminal ribonuclease (BS-RNase) acquires an interesting anti-tumor activity associated with the swapping on the N-terminal. The first direct experimental evidence on the formation of a C-terminal swapped dimer (C-dimer) obtained from the monomeric derivative of BS-RNase, although under non-native conditions, is here reported. The X-ray model of this dimer reveals a quaternary structure different from that of the C-dimer of RNase A, due to the presence of three mutations in the hinge peptide 111-116. The mutations increase the hinge peptide flexibility and decrease the stability of the C-dimer against dissociation. The biological implications of the structural data are also discussed.

Structural and functional relationships of natural and artificial dimeric bovine ribonucleases: new scaffolds for potential antitumor drugs

Protein aggregation via 3D domain swapping is a complex mechanism which can lead to the acquisition of new biological, benign or also malignant functions, such as amyloid deposits. In this context, RNase A represents a fascinating model system, since by dislocating different polypeptide chain regions, it forms many diverse oligomers. No other protein displays such a large number of different quaternary structures. Here we report a comparative structural analysis between natural and artificial RNase A dimers and bovine seminal ribonuclease, a natively dimeric RNase with antitumor activity, with the aim to design RNase A derivatives with improved pharmacological potential.

Double domain swapping in bovine seminal RNase: formation of distinct N- and C-swapped tetramers and multimers with increasing biological activities

Bovine seminal (BS) RNase, the unique natively dimeric member of the RNase super-family, represents a special case not only for its additional biological actions but also for the singular features of 3D domain swapping. The native enzyme is indeed a mixture of two isoforms: M = M, a dimer held together by two inter-subunit disulfide bonds, and MxM, 70% of the total, which, besides the two mentioned disulfides, is additionally stabilized by the swapping of its N-termini.

When lyophilized from 40% acetic acid, BS-RNase oligomerizes as the super-family proto-type RNase A does. In this paper, we induced BS-RNase self-association and analyzed the multimers by size-exclusion chromatography, cross-linking, electrophoresis, mutagenesis, dynamic light scattering, molecular modelling. Finally, we evaluated their enzymatic and cytotoxic activities.

Several BS-RNase domain-swapped oligomers were detected, including two tetramers, one exchanging only the N-termini, the other being either N- or C-swapped. The C-swapping event, confirmed by results on a BS-K113N mutant, has been firstly seen in BS-RNase here, and probably stabilizes also multimers larger than tetramers.

Interestingly, all BS-RNase oligomers are more enzymatically active than the native dimer and, above all, they display a cytotoxic activity that definitely increases with the molecular weight of the multimers. This latter feature, to date unknown for BS-RNase, suggests again that the self-association of RNases strongly modulates their biological and potentially therapeutic properties.

Investigation on PLK2 and PLK3 substrate recognition

Analyses of human phosphoproteome based on primary structure of the aminoacids surrounding the phosphor Ser/Thr suggest that a significant proportion of phosphosites is generated by a restricted number of acidophilic kinases, among which protein kinase CK2 plays a prominent role. Recently, new acidophilic kinases belonging to the Polo like kinase family have been characterized, with special reference to PLK1, PLK2, and PLK3 kinases. While some progress has been made in deciphering the PLK1-dependent phosphoproteome, very little is known about the targets of PLK2 and PLK3 kinases. In this report by using an in vitro approach, consisting of cell lysate phosphorylation, phosphoprotein separation by 2D gel electrophoresis and mass spectrometry, we describe the identification of new potential substrates of PLK2 and PLK3 kinases. We have identified and validated as in vitro PLK2 and PLK3 substrates HSP90, GRP-94, β-tubulin, calumenin, and 14-3-3 epsilon. The phosphosites generated by PLK3 in these proteins have been identified by mass spectrometry analysis to get new insights about PLKs specificity determinants. These latter have been further corroborated by an in silico analysis of the PLKs substrate binding region.

3DSwap: curated knowledgebase of proteins involved in 3D domain swapping

Three-dimensional domain swapping is a unique protein structural phenomenon where two or more protein chains in a protein oligomer share a common structural segment between individual chains. This phenomenon is observed in an array of protein structures in oligomeric conformation. Protein structures in swapped conformations perform diverse functional roles and are also associated with deposition diseases in humans. We have performed in-depth literature curation and structural bioinformatics analyses to develop an integrated knowledgebase of proteins involved in 3D domain swapping. The hallmark of 3D domain swapping is the presence of distinct structural segments such as the hinge and swapped regions. We have curated the literature to delineate the boundaries of these regions. In addition, we have defined several new concepts like ‘secondary major interface’ to represent the interface properties arising as a result of 3D domain swapping, and a new quantitative measure for the ‘extent of swapping’ in structures. The catalog of proteins reported in 3DSwap knowledgebase has been generated using an integrated structural bioinformatics workflow of database searches, literature curation, by structure visualization and sequence–structure–function analyses. The current version of the 3DSwap knowledgebase reports 293 protein structures, the analysis of such a compendium of protein structures will further the understanding molecular factors driving 3D domain swapping.

Database URL:http://caps.ncbs.res.in/3dswap

RNase A oligomerization through 3D domain swapping is favoured by a residue located far from the swapping domains

Bovine pancreatic ribonuclease A forms 3D domain-swapped oligomers by lyophilization from 40% acetic acid solutions or if subjected to various thermally-induced denaturation procedures. Considering that the intrinsic swapping propensity of bovine seminal RNase, the only member of the pancreatic-type RNase super-family that is dimeric in nature, is decreased from 70 to 30% if Arg80 is substituted by Ser (the corresponding residue in native RNase A), we introduced the opposite mutation in position 80 of the pancreatic enzyme. Our aim was to detect if the RNase A tendency to aggregate through domain swapping could increase. Aggregation of the S80R-RNase A mutant was induced either through the 'classic' acetic acid lyophilization, or through a thermally-induced method. The results indicate that the S80R mutant aggregates to a higher extent than the native protein, and that the increase occurs especially through N-terminal swapping. Additional investigations on the dimeric and multimeric species formed indicate that the S80R mutation increases their stability against regression to monomer, and does not significantly change their structural and functional features.

Insights into Protein Sequence and Structure-Derived Features Mediating 3D Domain Swapping Mechanism using Support Vector Machine Based Approach

3-dimensional domain swapping is a mechanism where two or more protein molecules form higher order oligomers by exchanging identical or similar subunits. Recently, this phenomenon has received much attention in the context of prions and neurodegenerative diseases, due to its role in the functional regulation, formation of higher oligomers, protein misfolding, aggregation etc. While 3-dimensional domain swap mechanism can be detected from three-dimensional structures, it remains a formidable challenge to derive common sequence or structural patterns from proteins involved in swapping. We have developed a SVM-based classifier to predict domain swapping events using a set of features derived from sequence and structural data. The SVM classifier was trained on features derived from 150 proteins reported to be involved in 3D domain swapping and 150 proteins not known to be involved in swapped conformation or related to proteins involved in swapping phenomenon. The testing was performed using 63 proteins from the positive dataset and 63 proteins from the negative dataset. We obtained 76.33% accuracy from training and 73.81% accuracy from testing. Due to high diversity in the sequence, structure and functions of proteins involved in domain swapping, availability of such an algorithm to predict swapping events from sequence and structure-derived features will be an initial step towards identification of more putative proteins that may be involved in swapping or proteins involved in deposition disease. Further, the top features emerging in our feature selection method may be analysed further to understand their roles in the mechanism of domain swapping.

Isoform specific phosphorylation of p53 by protein kinase CK1

The ability of three isoforms of protein kinase CK1 (alpha, gamma(1), and delta) to phosphorylate the N-terminal region of p53 has been assessed using either recombinant p53 or a synthetic peptide reproducing its 1-28 sequence. Both substrates are readily phosphoylated by CK1delta and CK1alpha, but not by the gamma isoform. Affinity of full size p53 for CK1 is 3 orders of magnitude higher than that of its N-terminal peptide (K (m) 0.82 muM vs 1.51 mM). The preferred target is S20, whose phosphorylation critically relies on E17, while S6 is unaffected despite displaying the same consensus (E-x-x-S). Our data support the concept that non-primed phosphorylation of p53 by CK1 is an isoform-specific reaction preferentially affecting S20 by a mechanism which is grounded both on a local consensus and on a remote docking site mapped to the K(221)RQK(224) loop according to modeling and mutational analysis.

Domain swapping in materials design

Peptide self-assembly can be used as a bottom-up approach to material fabrication. Although many different types of materials can be prepared from peptides, hydrogels are perhaps one of the most common. Gels typically result from the self-assembly of peptides into fibrillar networks. Controlling the structural morphology of these fibrils and the networks they form allows direct control over a given material's bulk properties. However, exerting this control is extremely difficult as the mechanistic rules that govern peptide self-assembly are far from being established. Conversely, several amyloidogenic proteins have been shown to self-assemble into fibrils using a mechanism known as domain swapping. Here, discrete units of secondary structure or even whole domains are exchanged (swapped) among discrete proteins during self-assembly to form extended networks with precise structural control. This review discusses several common mechanistic variations of domain swapping using naturally occurring proteins as examples. The possibility of using these principles to design peptides capable of controlled assembly and fibril formation leading to materials with targeted properties is explored.

Cystic fibrosis transmembrane regulator fragments with the Phe508 deletion exert a dual allosteric control over the master kinase CK2

Cystic fibrosis mostly follows a single Phe508 deletion in CFTR (cystic fibrosis transmembrane regulator) (CFTRDeltaF508), thereby causing premature fragmentation of the nascent protein with concomitant alterations of diverse cellular functions. We show that CK2, the most pleiotropic protein kinase, undergoes allosteric control of its different cellular forms in the presence of short CFTR peptides encompassing the Phe508 deletion: these CFTRDeltaF508 peptides drastically inhibit the isolated catalytic subunit (alpha) of the kinase and yet up-regulate the holoenzyme, composed of two catalytic and two non-catalytic (beta) subunits. Remarkable agreement between in silico docking and our biochemical data point to different sites for the CFTRDeltaF508 peptide binding on isolated CK2alpha and on CK2beta assembled into the holoenzyme, suggesting that CK2 targeting may be perturbed in cells expressing CFTRDeltaF508; this could shed light on some pleiotropic aspects of cystic fibrosis disease.

Kinetic analysis provides insight into the mechanism of ribonuclease A oligomer formation

Ribonuclease A forms a series of oligomers by 3D domain swapping, a possible mechanism for amyloid formation. Using experimental data, the Ribonuclease oligomerization process is analyzed to obtain estimates of individual equilibrium and microscopic rate constants. The results suggest several novel insights into Ribonuclease oligomer formation: (i) two dimers may combine to yield tetramers, (ii) the lower abundance of the cyclic trimer could be ascribed to the cis conformation of its Asn113-Pro114 peptide bonds, (iii) oligomers become the dominant species at very high protein concentrations or upon applying a modest tenfold increase in the equilibrium constants (iv) the rate constants for trimer and tetramer formation are faster than those of dimer formation and (v) glycosylation affects the relative populations of different trimer and tetramer species. By mass spectrometry, oligomers as large as tetradecamers are detected. These results are consistent with the proposal that 3D domain swapping is a mechanism for amyloid formation.

Oligomerization of ribonuclease A under reducing conditions

By lyophilization from 40% acetic acid solutions, bovine ribonuclease A forms well characterized, three-dimensional domain-swapped oligomers: dimers, trimers, tetramers, and higher order multimers. Each oligomeric species consists of at least two conformers. Identical oligomers also form by thermally-inducing the oligomerization of highly concentrated RNase A dissolved in fluids endowed with various denaturing power. Now, our question is: which might the influence of a reducing agent be on RNase A oligomerization, i.e., of conditions that decrease the stability of the protein and increase the mobility of its swapping domains? To address this question, we carried out experiments of RNase A oligomerization in the presence of increasing concentrations of dithiothreitol (DTT) under the two experimental conditions mentioned above. Results indicate that RNase A oligomers similar to those previously known form anyhow, but with a change of their relative proportions. The amounts of dimers and trimers decrease by increasing the concentration of DTT, while the yields of two tetramers remarkably increase. Moreover, in the presence of DTT RNase A forms labile and probably unstructured aggregates that can possibly drive the protein towards precipitation when the reducing agent's concentration increases. Taken together, these results point out once again (i) the important role of the 3D domain swapping mechanism in protein oligomerization, and (ii) the importance of the native structure of RNase A (and of proteins in general) in preventing an uncontrolled aggregation and precipitation in a reducing and highly crowded environment like that existing in a living cell.