Differential screening of phage-ab libraries by oligonucleotide microarray technology

A novel and efficient tagArray technology was developed that allows rapid identification of antibodies which bind to receptors with a specific expression profile, in the absence of biological information. This method is based on the cloning of a specific, short nucleotide sequence (tag) in the phagemid coding for each phage-displayed antibody fragment (phage-Ab) present in a library. In order to set up and validate the method we identified about 10,000 different phage-Abs binding to receptors expressed in their native form on the cell surface (10 k Membranome collection) and tagged each individual phage-Ab. The frequency of each phage-Ab in a given population can at this point be inferred by measuring the frequency of its associated tag sequence through standard DNA hybridization methods. Using tiny amounts of biological samples we identified phage-Abs binding to receptors preferentially expressed on primary tumor cells rather than on cells obtained from matched normal tissues. These antibodies inhibited cell proliferation in vitro and tumor development in vivo, thus representing therapeutic lead candidates.

High-avidity monoclonal antibodies against the human scavenger class B type I receptor efficiently block hepatitis C virus infection in the presence of high-density lipoprotein

The human scavenger class B type 1 receptor (SR-B1/Cla1) was identified as a putative receptor for hepatitis C virus (HCV) because it binds to soluble recombinant HCV envelope glycoprotein E2 (sE2). High-density lipoprotein (HDL), a natural SR-B1 ligand, was shown to increase the in vitro infectivity of retroviral pseudoparticles bearing HCV envelope glycoproteins and of cell culture-derived HCV (HCVcc), suggesting that SR-B1 promotes viral entry in an HDL-dependent manner. To determine whether SR-B1 participates directly in HCV infection or facilitates HCV entry through lipoprotein uptake, we generated a panel of monoclonal antibodies (MAbs) against native human SR-B1. Two of them, 3D5 and C167, bound to conformation-dependent SR-B1 determinants and inhibited the interaction of sE2 with SR-B1. These antibodies efficiently blocked HCVcc infection of Huh-7.5 hepatoma cells in a dose-dependent manner. To examine the role of HDL in SR-B1-mediated HCVcc infection, we set up conditions for HCVcc production and infection in serum-free medium. HCVcc efficiently infected Huh-7.5 cells in the absence of serum lipoproteins, and addition of HDL led to a twofold increase in infectivity. However, the HDL-induced enhancement of infection had no impact on the neutralization potency of MAb C167, despite its ability to inhibit both HDL binding to cells and SR-B1-mediated lipid transfer. Of note, MAb C167 also potently blocked Huh-7.5 infection by an HCV strain recovered from HCVcc-infected chimpanzees. These results demonstrate that SR-B1 is essential for infection with HCV produced in vitro and in vivo and suggest the possible use of anti-SR-B1 antibodies as therapeutic agents.

Modulation of the immune response induced by gene electrotransfer of a hepatitis C virus DNA vaccine in nonhuman primates

Induction of multispecific, functional CD4+ and CD8+ T cells is the immunological hallmark of acute self-limiting hepatitis C virus (HCV) infection in humans. In the present study, we showed that gene electrotransfer (GET) of a novel candidate DNA vaccine encoding an optimized version of the nonstructural region of HCV (from NS3 to NS5B) induced substantially more potent, broad, and long-lasting CD4+ and CD8+ cellular immunity than naked DNA injection in mice and in rhesus macaques as measured by a combination of assays, including IFN-gamma ELISPOT, intracellular cytokine staining, and cytotoxic T cell assays. A protocol based on three injections of DNA with GET induced a substantially higher CD4+ T cell response than an adenovirus 6-based viral vector encoding the same Ag. To better evaluate the immunological potency and probability of success of this vaccine, we have immunized two chimpanzees and have compared vaccine-induced cell-mediated immunity to that measured in acute self-limiting infection in humans. GET of the candidate HCV vaccine led to vigorous, multispecific IFN-gamma+CD8+ and CD4+ T lymphocyte responses in chimpanzees, which were comparable to those measured in five individuals that cleared spontaneously HCV infection. These data support the hypothesis that T cell responses elicited by the present strategy could be beneficial in prophylactic vaccine approaches against HCV.

A T-cell HCV vaccine eliciting effective immunity against heterologous virus challenge in chimpanzees

Three percent of the world's population is chronically infected with the hepatitis C virus (HCV) and at risk of developing liver cancer. Effective cellular immune responses are deemed essential for spontaneous resolution of acute hepatitis C and long-term protection. Here we describe a new T-cell HCV genetic vaccine capable of protecting chimpanzees from acute hepatitis induced by challenge with heterologous virus. Suppression of acute viremia in vaccinated chimpanzees occurred as a result of massive expansion of peripheral and intrahepatic HCV-specific CD8(+) T lymphocytes that cross-reacted with vaccine and virus epitopes. These findings show that it is possible to elicit effective immunity against heterologous HCV strains by stimulating only the cellular arm of the immune system, and suggest a path for new immunotherapy against highly variable human pathogens like HCV, HIV or malaria, which can evade humoral responses.

A novel adenovirus type 6 (Ad6)-based hepatitis C virus vector that overcomes preexisting anti-ad5 immunity and induces potent and broad cellular immune responses in rhesus macaques

Success in resolving hepatitis C virus (HCV) infection has been correlated to vigorous, multispecific, and sustained CD8(+) T-cell response in humans and chimpanzees. The efficacy of inducing T-cell-mediated immunity by recombinant serotype 5 adenovirus vector has been proven in many animal models of infectious diseases, but its immunogenicity can be negatively influenced by preexisting immunity against the vector itself. To evaluate the less prevalent adenovirus serotype 6 (Ad6) as an alternative vector for and HCV vaccine development, we have generated serotype 5 and 6 adenoviral vectors directing expression of the nonstructural region of HCV (MRKAd5-NSmut and MRKAd6-NSmut). Immunogenicity studies in mice showed that the two vectors induced comparable T-cell responses but that only MRKAd6-NSmut was not suppressed in the presence of anti-Ad5 immunity. In contrast, preexisting anti-Ad5 immunity dramatically blunted the immunogenicity of the serotype 5-based HCV vector. Furthermore, MRKAd6-NSmut showed equivalent potency, breadth, and longevity of HCV-specific T-cell responses in rhesus macaques as the corresponding Ad5-based vector over a wide range of doses and was capable of boosting DNA-primed animals even if administered at low doses. These data support the use of the MRKAd6-NSmut for anti-HCV immunotherapy and, more generally, for the Ad6 serotype as a better genetic vaccine vehicle than Ad5.

Searching for DNA-protein interactions by lambda phage display

We applied phage display technology to DNA-protein interaction studies. A cDNA expression library displayed on the surface of bacteriophage lambda was generated from the highly differentiated MMH E14 murine hepatic cell line. Selection of this library using the promoter sequence of the liver-enriched transcription factor HNF1alpha gene as ligate identified DNA-binding domains specifically interacting with different regions of this regulatory sequence. One of the selected phage showed 100% identity to a DNA-binding domain shared by differentiation specific element-binding protein, vasoactive intestinal peptide receptor-repressor protein and replication factor C and was further investigated. Specific binding of the selected protein domain was confirmed in a phage-independent context. By combining ELISA and South-Western assays using the selected phage and a bacterially expressed glutathione-S-transferase protein fused to the encoded DNA-binding domain, an array of multiple adjacent DNA-binding sites sharing a common consensus motif was identified. The strategy described represents a powerful tool to identify proteins that bind to DNA regulatory elements.

Biotin-tagged cDNA expression libraries displayed on lambda phage: a new tool for the selection of natural protein ligands

cDNA expression libraries displayed on lambda phage have been successfully employed to identify partners involved in antibody-antigen, protein- protein and DNA-protein interactions and represent a novel approach to functional genomics. However, as in all other cDNA expression libraries based on fusion to a carrier polypeptide, a major issue of this system is the absence of control over the translation frame of the cDNA. As a consequence, a large number of clones will contain lambda D/cDNA fusions, resulting in the foreign sequence being translated on alternative reading frames. Thus, many phage will not display natural proteins, but could be selected, as they mimic the binding properties of the real ligand, and will hence interfere with the selection outcome. Here we describe a novel lambda vector for display of exogenous peptides at the C-terminus of the capsid D protein. In this vector, translation of fusion peptides in the correct reading frame allows efficient in vivo biotinylation of the chimeric phage during amplification. Using this vector system we constructed three libraries from human hepatoma cells, mouse hepatocytic MMH cells and from human brain. Clones containing open reading frames (ORFs) were rapidly selected by streptavidin affinity chromatography, leading to biological repertoires highly enriched in natural polypeptides. We compared the selection outcome of two independent experiments performed using an anti-GAP-43 monoclonal antibody on the human brain cDNA library before and after ORF enrichment. A significant increase in the efficiency of identification of natural target peptides with very little background of false-positive clones was observed in the latter case.

Application of phage display peptide library to autoimmune diabetes: identification of IA-2/ICA512bdc dominant autoantigenic epitopes

Autoantigenic epitope mapping represents a critical issue in autoimmune diseases. The islet tyrosine phosphatase-like protein IA-2/ICA512bdc is a major autoantigen in type 1 diabetes (IDDM), but the epitopes responsible for autoantibody binding have been only partially defined. The aim of our study was to identify ICA512bdc epitopes, and in particular mini-epitopes, utilizing a novel strategy for autoimmune diseases. The study was performed in three sequential steps: (1) construction of a lambda-phage surface-displayed ICA512bdc cDNA library with the methodology of tagged random priming with peptides displayed as a fusion to the C terminus of the capsid protein D; (2) affinity selection of the resulting library, followed by immunoscreening, enzyme-linked immunosorbent assay and sequence analysis of positive clones, and (3) radioimmunoprecipitation to detect autoantibodies to the selected clones. This strategy resulted in the identification of two epitopes (IA-2 residues 761 - 964 and 929 - 979), which were recognized by 100 % and 62.9 % ICA512bdc-positive IDDM patients, respectively. Interestingly, the larger clone was detected also by a proportion (16.7 %) of new onset ICA512bdc-negative patients, thus suggesting that this region contains not only the main autoantigenic repertoire of ICA512bdc molecule, but is able to detect IA-2 autoantibodies in even higher percentages of patients. In addition, this study showed the existence of multiple epitopes located in the C-terminal domain of the IA-2 protein, one of which is formed by the 50 C-terminal amino acids, and provided evidence that the strategy used represents a valid tool for identification of epitopes within autoantigenic molecules.

Bacteriophage lambda display of complex cDNA libraries: a new approach to functional genomics

We describe the construction and characterization of two lambda surface displayed cDNA expression libraries derived from human brain and mouse embryo. cDNA inserts were obtained by tagged random-priming elongation of commercially available cDNA libraries and cloned into a novel lambda vector at the 3' end of the D capsid protein gene, which produced highly complex repertoires (1x10(8) and 2x10(7) phage). These libraries were affinity selected with a monoclonal antibody against the neural specific factor GAP-43 and with polyclonal antibodies that recognize the EMX1 and EMX2 homeoproteins. In both cases rapid identification of specific clones was achieved, which demonstrates the great potential of the lambda display system for generating affinity selectable cDNA libraries from complex genomes.

[Medico-legal problems in the relationship between psychopathology and work]

The Author treats the relationship between work and mental health from a legal medical viewpoint and referring to two different opposite situations: work as a pathogenic factor and work as a therapeutic factor. In the first case, the problem concerns the acknowledgment of the connection between work and mental disorders. It includes the need of a careful estimate of the causation and the meaning that it could assume during the damage evaluation process, the peculiar psycopathological question of the relationship between disturb and damage, and the difficulties in the estimation of this latter. Finally, it concerns the setting peculiarity, essential to a correct methodological approach during the evaluation. In the second case, the focus is the safeguard of both the subject with a psychiatric disorder and his environment, with the need to insert the occupational physician in a "net" of supportive interventions aimed at avoiding that the work became an aggravating agent or an additional risk factor. The Author emphasizes the implicit ethical component of the choice to introduce the psychiatric patient in the "working life"; this choice belongs both to the single practitioner and to the collectivity, and it is supported by the Constitution.

Osteopontin is an autoantigen of the somatostatin cells in human islets: identification by screening random peptide libraries with sera of patients with insulin-dependent diabetes mellitus

Random peptide libraries (RPLs) screening with IDDM sera has identified 5 disease-specific 'mimotopes' displayed on phage (phagotopes). We characterised one phagotope (CH1p), by raising a rabbit antibody against the peptide insert on phage, which was employed in immunohistochemistry, Western blotting and cDNA libraries screening. The CH1p mimotope was detected in somatostatin cells of human islets and experimentally raised anti-osteopontin antibodies or human sera positive for the phagotope, detected a similar subpopulation of islet cells. The screening of cDNA library identified a clone corresponding to human osteopontin. In summary, RPLs proved to be successful in the identification of a novel islet-related autoantigen (osteopontin), whose significance in disease remains to be established.

Efficient display of an HCV cDNA expression library as C-terminal fusion to the capsid protein D of bacteriophage lambda

We describe the construction and characterization of a hepatitis C virus (HCV) cDNA expression library displayed as a fusion to the carboxy terminus of the capsid protein D of bacteriophage lambda. cDNA inserts were obtained by tagged random-priming of the HCV genome and cloned into a lambda vector from which chimeric phage bearing both wild-type D protein and D fusion products on the capsid surface were produced. The resulting library was affinity-selected with anti-HCV human monoclonal antibodies recognizing linear or conformational epitopes, and human sera from HCV-infected patients. Selection was monitored by immuno-screening experiments, ELISA, and sequence analysis of positive clones. The performance of this library was compared with two additional HCV cDNA display libraries generated as N-terminal fusions to the III and VIII capsid proteins of filamentous phage M13. The results obtained demonstrate the great potential of the lambda display system for constructing complex cDNA libraries for natural ligand discovery.

Peptide and protein display on the surface of filamentous bacteriophage

The isolation of ligands that bind biologically relevant molecules is fundamental to the understanding of biological processes and to the search for therapeutics. Filamentous phage can be used to display foreign peptides and proteins in physical association with their DNA coding sequences. Repertoires larger than 10(8) phage clones expressing different peptide sequences can be prepared using molecular genetic techniques. The strategies utilizing this technology promise to provide not only new binding and possibly catalytic activities, but also lead structures for the development of new drugs and vaccines.

Identification of a novel type 1 diabetes-specific epitope by screening phage libraries with sera from pre-diabetic patients

We used random peptide libraries displayed on phage to search for ligands to insulin dependent diabetes mellitus-related antibodies and were able to identify several candidate disease-related peptides. One of them, clone 92, showed a significant difference in the frequency of reactivity with the sera of patients and normal controls. Human immunoglobulins immunopurified on phage 92 specifically stained the islets on human pancreatic sections. When injected into rabbits, the selected peptide elicited antibodies that also stained human and rat pancreatic sections, with a pattern similar to that observed with immunoglobulins purified from the sera of patients. No reactivity was observed in other tissues. Our results indicate that the peptide identified in this work mimics a novel, diabetes-related self-antigen.

Selection of biologically active peptides by phage display of random peptide libraries

Random peptide libraries displayed on phage are used as a source of peptides for epitope mapping, for the identification of critical amino acids responsible for protein-protein interactions and as leads for the discovery of new therapeutics. Efficient and simple procedures have been devised to select peptides binding to purified proteins, to monoclonal and polyclonal antibodies and to cell surfaces in vivo and in vitro.

Selection of phage-displayed peptides mimicking type 1 diabetes-specific epitopes

Phage display technology represents a powerful tool for the identification of peptides reacting with disease-related antibodies present in human sera. The application of this technology to type 1 diabetes could provide a set of novel reagents for diabetes prediction and could also lead to the identification of novel autoantigens or even of environmental factors possibly causing the disease. In the present study, sera of prediabetic and high risk individuals were used to select candidate peptides from phage-displayed random peptide libraries. Diabetes specific phage clones were then identified from these through screening and counter screening, using sera from diabetic and non-diabetic individuals. The results presented in this paper demonstrate the feasibility of this methodology to identify peptides reacting preferentially with antibodies present in the serum of diabetic patients.

Modelling antibody-antigen interactions: ferritin as a case study

In this work, we propose a model for the structure of the antigen-antibody complex formed by human H-ferritin and an antibody that specifically recognizes it. We cloned and sequenced the antibody gene, predicted the antibody three-dimensional structure, and reconstructed the H-ferritin-antibody complex using an automated docking procedure previously validated on known complexes. This procedure allowed us to identify one putative complex which we carefully analysed, in order both to evaluate its likelihood, in light of a set of experimental results described in the literature, and to predict precisely which are the sites of interaction between the two molecules. Our model is compatible with the experimentally determined characteristics of the complex. Some of the residues that form the predicted antigenic site of ferritin can be found in the amino acid sequence of peptides selected from a random peptide library because of their affinity for the ferritin monoclonal antibody. Furthermore, the structural difference between the antigenic site in human H-ferritin and the corresponding region in other species permits us to rationalize the inability of the antibody to recognize human L-ferritin and rat, chicken and mouse H-ferritin. Through the analysis of our model complex, we identify a number of other residues putatively involved in the interaction. This multidisciplinary approach shows that synergy between computational and experimental methods may bring further insight into the understanding of antibody-antigen recognition rules.

Derivation of vaccines from mimotopes. Immunologic properties of human hepatitis B virus surface antigen mimotopes displayed on filamentous phage

We have previously reported the identification, using human immune sera, of mimotopes of human hepatitis B virus surface Ag (HBsAg) displayed on filamentous phage. To test if these mimotopes could be useful in developing a vaccine against the human hepatitis B virus (HBV), we have compared the humoral immune response of animals immunized either with a recombinant HBsAg vaccine, or with mimotopes. Immunogens were prepared by fusing the mimotopes on different carrier molecules (phage coat protein pIII and pVIII, recombinant human H ferritin, HBV core peptide) and by synthesizing multiple antigenic peptides carrying the mimotopes' amino acid sequences. These immunogens were injected into mice and rabbits and sera were collected and tested for the presence of HBsAg-specific Abs. Our data confirm that mimotopes can induce a humoral immune response resembling that induced by the original Ag, and HBsAg mimotopes displayed on phage prove to be the best immunogens, inducing the most reproducible and potent immunization. Mimotopes that react as HBV subtype-specific Ags do not show this specificity as immunogen and induce a nonsubtype-restricted response. Furthermore, mimotopes displayed on phage elicit a strong response to HBsAg in a strain of mouse reported to show a low response to it. These results indicate that mimotopes identified from random peptide libraries through utilizing human immune sera could be important leads for the derivation of new vaccines.

Derivation of vaccines from mimotopes. Immunologic properties of human hepatitis B virus surface antigen mimotopes displayed on filamentous phage

We have previously reported the identification, using human immune sera, of mimotopes of human hepatitis B virus surface Ag (HBsAg) displayed on filamentous phage. To test if these mimotopes could be useful in developing a vaccine against the human hepatitis B virus (HBV), we have compared the humoral immune response of animals immunized either with a recombinant HBsAg vaccine, or with mimotopes. Immunogens were prepared by fusing the mimotopes on different carrier molecules (phage coat protein pIII and pVIII, recombinant human H ferritin, HBV core peptide) and by synthesizing multiple antigenic peptides carrying the mimotopes' amino acid sequences. These immunogens were injected into mice and rabbits and sera were collected and tested for the presence of HBsAg-specific Abs. Our data confirm that mimotopes can induce a humoral immune response resembling that induced by the original Ag, and HBsAg mimotopes displayed on phage prove to be the best immunogens, inducing the most reproducible and potent immunization. Mimotopes that react as HBV subtype-specific Ags do not show this specificity as immunogen and induce a nonsubtype-restricted response. Furthermore, mimotopes displayed on phage elicit a strong response to HBsAg in a strain of mouse reported to show a low response to it. These results indicate that mimotopes identified from random peptide libraries through utilizing human immune sera could be important leads for the derivation of new vaccines.

Identification of biologically active peptides using random libraries displayed on phage

The construction of new and increasingly diverse libraries, as well as the implementation of more powerful selection schemes, has led to the identification of linear peptides that mimic complex epitopes. Phage display techniques are allowing the selection of disease-related peptides, which reproduce the antigenic and immunogenic properties of natural antigens, using whole sera from patients. The range of applications of phage technology has been extended to include the search for peptides binding to molecules other than antibodies, such as cell receptors and enzymes.

Monoclonal antibodies that recognise filamentous phage: tools for phage display technology

We generated six hybridoma cell lines that secrete monoclonal antibodies (mAb) which specifically bind filamentous phage coat proteins. Two of these mAb recognise epitopes that include the N terminus of the coat protein III (pIII), while two others are specific for the N terminus of the major coat protein VIII (pVIII). These mAb are valuable tools to study phage assembly and structure. Furthermore, we describe two examples of how these mAb can be exploited in the construction and screening of peptide libraries displayed by the filamentous phase major coat protein. We have used one of these mAb to develop a sensitive ELISA with crude phage supernatants. This assay allows rapid screening of large numbers of clones from random peptide phage libraries. Some of the anti-phage mAb described here can interfere with wild-type phage propagation, while phage carrying modifications in their coat proteins are insensitive to growth inhibition. We have exploited this observation as a tool to favour the growth of phage displaying peptides fused to pVIII, with respect to vector phage.

Epitope discovery using peptide libraries displayed on phage

Peptides displayed on phage, which mimic continuous and discontinuous epitopes, can be selected using purified antibodies or preparations of polyclonal serum. This review describes recent advances in this field, discusses the application of phage-display technology to the diagnosis of human diseases, and presents new ideas for the preparation of vaccines directed against specific epitopes on a pathogen.

A database system for handling phage library-derived sequences

We have implemented a system for creating and maintaining nucleotide and amino acid sequence databases especially suited for the handling of phage library-derived sequences. The system is currently used in our laboratory and has already proven to be useful for the comparison of sequences obtained by different investigators. We believe that the availability of this system will encourage and simplify the exchange of sequence data among different laboratories.

Mimicking of discontinuous epitopes by phage-displayed peptides, II. Selection of clones recognized by a protective monoclonal antibody against the Bordetella pertussis toxin from phage peptide libraries

We have screened phage peptide libraries to establish if clones binding to a monoclonal antibody (mAb), specific for a discontinuous epitope, could be isolated and if the selected phage particles would be able to elicit an in vivo immuno-response against the original antigen. Two phage peptide libraries, consisting of 9 random amino acids inserted in the major coat protein (pVIII), were independently screened with a mAb which is capable of neutralizing the Bordetella pertussis toxin (PTX) in in vitro and in vivo assays. The epitope of PTX recognized by this and other protective mAb has been shown to be discontinuous. Six different positive phage clones were selected; their binding to the mAb could be competed for by PTX, showing that these clones bind to the antigen-binding site of the mAb. Three of the clones were used (alone or as a mixture) to immunize BALB/c mice. The sera showed a good immunoresponse both against the phage bearing the epitopes and against synthetic multiple-antigen peptides of the same sequence. The immune sera, however, showed no detectable signal against PTX and no capacity to neutralize the CHO-cell-clustering activity of the toxin. The results show that the selected recombinant phage are capable of mimicking the discontinuous epitope as far as binding to the corresponding mAb, but they are unable to elicit a detectable production of antibodies specific for the original antigen.

Mimicking of discontinuous epitopes by phage-displayed peptides, I. Epitope mapping of human H ferritin using a phage library of constrained peptides

We have constructed a random nonapeptide library in the N-terminal region of the major coat protein VIII of bacteriophage f1, with two cysteines flanking the insert, and preliminary data suggest that many of the clones display at least some of their peptides in cyclized form. This library was used to select oligopeptides binding to the monoclonal antibody (mAb) H107, recognising the assembled native conformation of recombinant human H-subunit ferritin (H Fer), whose three-dimensional structure is known. Comparison of the selected oligopeptides with one another allowed us to derive two consensus sequences characterized by conserved amino acid (aa) residues. Analysis of the distribution of the aa side chains exposed on the surface of H Fer reveals that most of the aa defining both consensus sequences are present either at the end of the big loop or at the end of the A helix. These two regions of the H Fer, though separated in the linear sequence, are very close in the folded molecule. Interestingly, each consensus sequence derived from the selected phage-displayed peptides is characterized by aa present both at the end of the big loop and at the end of the A helix. These two H Fer regions are good candidates for mimicry by the selected peptides and therefore for constituting part of the H107 epitope. To provide support to this hypothesis, we constructed several H Fer mutants carrying point mutations in different positions of these two regions.(ABSTRACT TRUNCATED AT 250 WORDS)

Loop mutations can cause a substantial conformational change in the carboxy terminus of the ferritin protein

Although some protein folding theories sustain that the peptides (loops) that connect elements of more compact secondary structure may be important in the folding process, most of the data accumulated until now seems to contradict this notion. To approach this problem we have isolated and characterized a number of mutants in which the amino acid sequence of the peptide that connects helix D and helix E in the H-chain of human ferritin has been randomized. Our results indicate that, though no single loop residue is absolutely required for ferritin to attain the native conformation, most of the mutants that we have obtained by random regional mutagenesis, affect its folding/assembly process. This conclusion was reached utilizing a sensitive test that associates the color formed by a colony synthesizing a hybrid ferritin-beta-galactosidase protein to the ability of the ferritin domain to fold and assemble as the native protein. The characterization of the folding/assembly properties of our collection of mutants and the comparison of the mutant loop sequences, have allowed us to draw the following conclusions. Mutants that have positively charged residues at position 159, 160 or 161 fail to assemble into the native protein shell and form an insoluble aggregate. Interestingly some loop amino acid sequences cause the E-helix to reverse direction and to expose its COOH group, normally hidden inside the protein cavity, to the solvent. The propensity of a given ferritin mutant to fold into this "non-native" conformation can be attenuated by the introduction of Gly at position 159 and 164, as in the natural ferritin.

A mutational analysis of the epitopes of recombinant human H-ferritin

Murine monoclonal antibodies were elicited by the recombinant human H-ferritin overexpressed in Escherichia coli. They had a specificity analogous to that of the antibodies elicited by natural human H-chain, and all of them showed low additivity in binding the recombinant ferritin. Four antibodies of each group were challenged with four H-ferritin mutants overexpressed in E. coli, altered in different accessible areas of the molecule. They consisted of deletions of the first 13 and last 22 amino acids, a duplication of an 18 amino acid sequence in the loop region, and a substitution of a 5 amino acid stretch in the three-fold symmetry axis region. Double diffusion, immunodot analyses and inhibition plots indicated that: (1) all the mutants were recognized by at least one antibody; (2) the deletion of the N-terminus and the duplication in the loop region had the strongest effect on antibody binding; and (3) epitope boundaries of the various antibodies could not be recognized. The antibodies were tested with H-containing ferritins from rat and hen hearts, and showed low or absent reactivities despite their high structural homology with human ferritin. Comparison of the amino acid sequences of human, mouse, rat and hen H-chains, together with mutational data, suggested that; (i) ferritin epitopes are large, probably encompassing a large portion of the subunit surface and (ii) Thr-5 and Cys-90 have a role in H-ferritin immunogenicity.

Mutational analysis of the channel and loop sequences of human ferritin H-chain

Human ferritin H-chain mutants were obtained by engineering the recombinant protein expressed by Escherichia coli. The mutagenesis were directed to the C-terminal sequence forming the hydrophobic channel, to the hydrophilic channel and to the loop sequence. The mutants were analysed for extent of expression, for stability, for capacity to incorporate iron and for kinetics of iron uptake and iron oxidation. Of the 22 mutants analysed only two with deletions of single residues in the loop sequence and one with deletion of the last 28 amino acid residues did not assemble into ferritin-like proteins. The other mutants assembled correctly and showed similar chemical/physical properties to the wild-type; they included duplication of an 18-amino acid-residue stretch, deletion of the last 22 and the last seven residues and various mutations of single amino acid residues. Two mutants with extensive alteration in the C-terminal sequence had a diminished thermostability associated with incapability to incorporate iron though they still catalysed iron oxidation. The mutants with alterations of the sequence around the hydrophilic channel showed diminished iron uptake and oxidation kinetics, together with a slightly larger apparent molecular size. The results indicate (i) that two of the sequences are important for ferritin assembly/stability, (ii) that the presence of the hydrophobic channel is essential for formation of the iron core and (iii) that the sites of iron interaction and the path of iron penetration into ferritin remain unidentified.

Identification of the ferroxidase centre in ferritin

Ferroxidase activity in human H-chain ferritin has been studied with the aid of site-directed mutagenesis. A site discovered by X-ray crystallography has now been identified as the ferroxidase centre. This centre is present only in H-chains and is located within the four-helix bundle of the chain fold.

A strategy to optimize translation initiation in recombinant mRNA: application to the Rop gene

Using the Escherichia coli Rop gene, we demonstrate a strategy that could be applied generally to optimize the initiation of translation of recombinant genes in E. coli. This involves cloning of the gene encoding the protein of interest in a suitable expression vector between an "efficient" ribosome binding site and the gene for the alpha-peptide of beta-galactosidase. By oligonucleotide-directed deletion mutagenesis, the two coding sequences are then fused in the correct frame. A second oligonucleotide is then used to place the initiator AUG (or GUG) at the correct distance from the Shine-Dalgarno sequence. In this step, however, we use an oligonucleotide that has a degenerate sequence. That is, on the basis of the "efficient" ribosome binding site sequence, we introduce random substitutions at various positions, both upstream and downstream from the initiator ATG, to obtain, after the mutagenesis experiment, a collection of random ribosome binding sites fused to the coding sequence. The development of blue colonies on indicator plates permits selection of clones in which an efficient ribosome binding site has been created for the specific gene of interest. We discuss the results obtained by applying the method to the Rop gene.

Recombinant H-chain ferritins: effects of changes in the 3-fold channels

Human H-chain ferritins bearing sequence changes in the 3-fold channels have been expressed in E. coli to investigate the role of these channels in iron-storage processes. The proteins assemble into shells resembling those of native ferritins. Iron uptake measurements indicate that residues in the 3-fold channels are involved neither in initial Fe(II)-oxidation nor in iron-core nucleation.

Isolation of point mutations that affect the folding of the H chain of human ferritin in E.coli

We have approached the problem of folding and assembly of the heavy (H) chain of human ferritin by isolating point mutations that affect this process. Apoferritin is an ideal model system to approach the problem of protein folding and assembly into multimeric structures. We have developed a recombinant hybrid molecule that allows us to select for ferritin mutants in which the folding-assembly process is altered or completely impaired. The selection procedure is based on a recombinant protein which consists of a fusion between the H chain of human ferritin and the alpha-peptide of beta-galactosidase. In the wild type situation, the alpha-peptide domain is segregated inside the apoferritin shell upon assembly and is unable to interact with the substrate and perform its enzymic function. We show that by selecting for mutations that restore beta-galactosidase activity we are able to identify ferritin mutations that affect the folding-assembly process. The selective procedure was applied to the analysis of the amino acid side chains that are important for the attainment of the correct conformation of the carboxy-terminal E helix in the 4-fold axis.

Mechanism of ferritin iron uptake: activity of the H-chain and deletion mapping of the ferro-oxidase site. A study of iron uptake and ferro-oxidase activity of human liver, recombinant H-chain ferritins, and of two H-chain deletion mutants

To study the functional differences between human ferritin H- and L-chains and the role of the protein shell in the formation and growth of the ferritin iron core, we have compared the kinetics of iron oxidation and uptake of ferritin purified from human liver (90% L) and of the H-chain homopolymer overproduced in Escherichia coli (100% H). As a control for iron autocatalytic activity, we analyzed the effect of Fe(III) on the iron uptake reaction. The results show that the H-chain homopolymer has faster rates of iron uptake and iron oxidation than liver ferritin in all the conditions analyzed and that the difference is reduced in the conditions in which iron autocatalysis in high: i.e. at pH 7 and in presence of iron core. We have also analyzed the properties of two engineered H-chains, one lacking the last 22 amino acids at the carboxyl terminus and the other missing the first 13 residues at the amino terminus. These mutant proteins assemble in ferritin-like proteins and maintain the ability to catalyze iron oxidation. The deletion at the carboxyl terminus, however, prevents the formation of a stable iron core. It is concluded that the ferritin H-chain has an iron oxidation site which is separated from the sites of iron transfer and hydrolysis and that either the integrity of the molecule or the presence of the amino acid sequences forming the hydrophobic channel is necessary for iron core formation.

Immunochemical characterization of human liver and heart ferritins with monoclonal antibodies

A library of 27 murine monoclonal antibodies was obtained by using human liver and heart ferritins as immunogens. The specificity of the antibodies for the two ferritins and their subunits was studied with five different methods. The antibodies elicited by the liver ferritin bound preferentially the immunogen and were specific for the L subunit. Some antibodies elicited by the heart ferritin had characteristics similar to the anti-liver antibodies, other ones bound preferentially the heart over the liver ferritin and were specific for the H subunit. Only two antibodies were able to bind both ferritins and subunits. Some anti-H and anti-L chain antibodies were used to develop and compare four types of immunoassay to quantitate isoferritins. The results indicate that heart ferritin is immunologically more heterogeneous than liver, the H and L subunits having large immunological differences with few, if any, identical epitopes; and that that the architecture of the immunoassays have a strong influence on the crossreactivity of the antibodies with the two isoferritins, probably because H and L chains are not arranged randomly in the assembled protein.