Patent Application
20150346192
The present invention relates to a method for identifying a cellular target involved in a cell phenotype comprising identifying an intrabody which can modify a cell phenotype and identifying a direct or indirect cellular target of the intrabody. The present invention also relates to intrabodies which are capable of inhibiting the degranulation reaction triggered by an allergic stimulus on mast cells, particularly to intrabodies 3H2-1, 3H2-VH and 5H4, and especially to intrabodies 3H2-1 and 5H4 which are capable of directly or indirectly targeting a protein of the ABCF1 family and of the C12ORF4 family respectively. The present invention also relates to ABCF1 and C12ORF4 inhibitors for use in therapy, in particular for treating allergic and/or inflammatory conditions.
Identifying cellular targets associated with a given phenotype is an essential prerequisite to a better understanding of cellular mechanisms underlying that phenotype. In particular, identifying cellular targets associated with a medical condition is of great interest for the pharmaceutical industry as it allows the design of new therapeutics which have an effect on these targets and can be used to treat or diagnose the medical condition. Frequently, a cell phenotype may be associated with a given pathology. An assay which would allow the identification of molecules that modify a cell phenotype (phenotypic screening) could thus be of great help to drug design. The identification of new cellular targets which may be associated with a given phenotype is also important to other sectors, for example in cosmetics or in the plant and food industry.
Understanding the role played by a known or unknown cellular molecule in a given phenotype is not an easy task in particular because cellular pathways are highly complex and because many intracellular molecules play a part in a number of different cellular pathways. Thus inactivating a target protein, for example by RNA interference, can have a modulatory effect on several cellular pathways and generate as a result several concomitant phenotypic effects which may be difficult to distinguish from one another. Moreover, many cell proteins can assume different conformations and/or may comprise post-translational modifications such as phosphorylated amino acids, and as a consequence may interact or not with another molecule as a function of their conformation or of post-translational modifications. A technique which would allow screening for cellular targets responsible for a given phenotype, and even for targets in a specific conformation and/or modified or non-modified state, without completely inactivating these cellular targets would thus be a highly valuable tool towards a better understanding of a number of cellular pathways.
Antibodies can be seen as precision tools as they are highly specific for their target and can be raised against virtually any part of a protein and in particular against large planar zones where protein interactions often take place and which are more difficult to target with small organic molecules or peptides. However, natural antibodies are not adapted to the intracellular environment because of their large size and because the reducing environment of the cytoplasm prevents the formation of disulfur bridges and thus prevents proper folding of the antibodies. Recombinant antibodies, among which scFv, diabodies and sdAb, have been developed for expression in an intracellular environment. They are referred to as “Intracellular antibodies” or “intrabodies”.
An application recently described by the inventors' team, is the use of intrabodies preselected for their modulating effect of a target antibody (or “Ab”) to screen a library of small molecules (European Patent EP1743178B1). The goal is to isolate drug candidates capable of mimicking the intracellular effects of the scFv of interest and that could be used in vivo. The inventors applied this ELISA displacement of Ab in a model of allergy. Previous work of the same team had led to the isolation of a scFv directed against the SH2 domains of tyrosine kinase Syk, involved in the early stages of mast cell activation. It had been shown that intracellular expression of this scFv in a mast cell line inhibits the release of allergic mediators induced by activation of IgE receptors, without disrupting the kinase activity of Syk (Dauvillier et al., 2002). An Ab displacement assay was developed to isolate molecular mimics of the anti-Syk scFv. In that manner, the screening of a chemical library of 3000 molecules led to the identification of a molecule, designated C-13. This molecule has a high anti-allergic potential because further studies demonstrated its ability to inhibit mast cell activation in vitro as well as anaphylactic shock in vivo in mice (Mazuc et al., 2008). The site of interaction of C-13 with Syk was also identified by a structural analysis and directed mutagenesis. The identified cavity at the interface between the two SH2 domains of Syk was used for a second screen in silico of a library of 500,000 molecules. In this manner, 85 non-enzymatic inhibitors of Syk were selected for their ability to inhibit the release of allergic mediators (Villoutreix et al., 2011).
Inoue et al., 2011 describes a loss-of-function screening using a library of randomized intracellular camelid VHH. The authors consider that conventional VH domains are insoluble in the intracellular environment. In the corresponding Japanese patent application JP2008-136455, the authors of Inoue et al., also consider scFvs as useless in the context of intracellular loss-of-function screening, because of a poor intracellular stability and functionality.
The inventors have developed a novel phenotypic screening technique based on the use of intracellular antibodies (or “intrabodies”), for the identification of cellular targets which can act on a cellular pathway and modulate a phenotype of interest. The present work describes the first phenotypic selection using intracellular human antibody fragments in mammalian cells.
The novel method developed by the inventors allows the identification of intracellular targets associated with a given phenotype, and in particular unknown intracellular targets or known targets not previously associated with a given phenotype. The inventors used this technique to identify two RBL-2H3 cellular clones named 3H2 and 5H4 which expressed respectively antibody fragments 3H2-1 and 3H2-VH, and 5H4. When expressed in mast cells, these antibody fragments inhibit the degranulation reaction which is normally triggered by an allergic challenge. The inventors have also identified two proteins, ABCF1 and LOC297607, which can be precipitated with 3H2-1 and 5H4 respectively. The inventors have demonstrated the involvement of LOC297607, for which no known function has been described in the art, in mast cell degranulation.
The choice of intracellular Ab for the realization of a phenotypic screening at the scale of the proteome, offers many advantages. Indeed, when expressed in cells, they are able to interfere with protein functions and thereby generate a given phenotype. Moreover, with their high affinity and specificity properties, they are potentially capable of targeting any molecule in a cell, and in particular proteins, and more precisely certain sub-domains or post-transcriptional modifications which may be involved in a particular signalling. In the context of disease models, another advantage offered by the intracellular antibodies, is that once their target is identified, they can be used to guide drug development.
More specifically, the use of a combinatory scFv library of high diversity and optimised for intracellular expression has allowed the inventors to perform a large scale phenotypic screen. Applying this screen to the model of mast cell activation, they have identified a new molecular player involved in this signalling pathway.
This approach was initially developed with a plasmid system allowing expression of the scFv library in a mast cell line. In order to retain as much as possible the initial diversity of the library, transfection conditions imposing a multicopy expression mode have been selected. Phenotypic selection has been performed and has led to enrichment in cells presenting the phenotype of interest, and expressing the scFv which inhibit the studied signalling pathway. The expression of two particular antibody (Ab) fragments, 3H2-1 and 5H4, in the mast cell line RBL-2H3, strongly inhibited the ability of cells to release allergic mediators. After having produced them, these antibodies have been used in vitro to identify their protein targets by immunoprecipitation experiments coupled to mass spectrometry. Proteins ABCF1 and LOC297607 have thereby been identified as the targets of 3H2-1 and 5H4 respectively.
LOC297607 (or “LOC”) is a protein of previously unknown function which seems to be involved in mast cell activation via FcεRI. Indeed, preliminary results with shRNA experiments suggest it intervenes in regulation both of early events following receptor activation and of later events of allergic mediators' release. It could be interesting to identify more precisely the molecular partners of the protein LOC in the mast cells, by evaluating amongst other things the activity of proteins which intervene in transduction of the signal mediated by the FcεRI.
In addition, several genomic data suggest a link between protein LOC and inflammation. A study using the technique of chromosome mapping by hybrid radiation in pigs showed that the chromosomal locus of LOC may be associated with inflammation, and more specifically with a particular form of arthritis (Genini et al. 2006). On analyzing this chromosome locus in Genebank, the inventors found that the LOC gene was colocalized with the gene encoding the protein tyrosine phosphatase PTPN6. Furthermore, this chromosomal co-location is conserved between rat, mouse and man. This information suggests that these proteins may be involved in the same signalling network. Indeed, new approaches for identifying protein-protein interactions are based on the comparison, between different species, of the chromosomal distances between the sequences of associated proteins (Pazos and Valencia 2001).
To maximize this phenotypic screening approach and expand its scope, the inventors have adapted it using a retroviral system for the intracellular expression of the scFv library. Compared with the previous plasmid system, this mode of expression allows the transfection of a wider repertoire of cell types, and provides stable expression which allows the study of phenotypes over a longer term.
During the plasmid selection, 2000-2500 copies of scFv were expressed per cell, which allowed the exploration of a wide diversity of scFv, while handling a relatively small number of cells. However, the fact that so many scFv were expressed per cell, imposed a strong competition between each of them in terms of intracellular effects, and thus probably maintained a high background noise. In this regard, the expression of the retroviral library that did not exceed more than 3 copies per cell, had the advantage of lowering the selection pressure. These conditions also allow the screen to be carried out in more physiological conditions, where the integrity of the protein network is not disrupted by overexpression.
High throughput sequencing analyses revealed that the inclusion of recloning steps every two rounds, during the retroviral selection allowed a better enrichment of scFv. Cell clones from the same selection and analysed independently showed a significant inhibition of β-hexosaminidase degranulation, which suggests that this selection effectively allowed an enrichment in inhibitory scFv.
Finally, the main data that enables the comparison of the three selections to each other, is the evolution of the Annexin V marking phenotype after cell stimulation (
The present inventors have proven that a scFv-based library could successfully be used in a phenotypic intracellular screening. By using a high diversity library of scFvs, based on a constant framework having intracellular stability, the inventors have succeeded in finding and identifying unknown intracellular targets to a given phenotype, for example mast cell degranulation.
The inventors have also discovered that truncated scFvs, including intrabodies 3H2-VH, 5H4-VH could be stable and functional in the intracellular environment. In particular, intrabody 5H4-VH, restricted to a full VH domain was found stable in said intracellular environment, and successfully inhibited mast cell degranulation.
A first aspect of the present invention is a method for identifying a cellular target which is involved in a cell phenotype, comprising:
Preferably, the intrabody comprises a full VH and/or VL domain of an immunoglobulin.
Antibodies are recognition proteins which are capable of binding specifically to a unique epitope on an antigen. The antigen binding site of an antibody is referred to as the paratope. Antibodies belong to the Immunoglobulin (Ig) class of proteins. Immunoglobulins are a family of proteins which have in common a specific structural domain referred to as the “Ig fold”. Most naturally occurring antibodies (or conventional antibodies) are globular proteins of about 150 kDa, which comprise two light chains (L) and two heavy chains (H) joined together by disulfide bonds. Non-conventional antibodies characterized by the absence of light chains are found in camelids and in non-cartilaginous fishes (e.g. sharks). Light and Heavy chains comprise a constant region (CL and CH respectively) and a variable region (VL and VH respectively). In mammals, there are five types of Ig heavy chains (α, δ, ε, γ, μ), which have a length of about 450 to 550 amino acids, and two types of Ig light chains (κ,λ) which have a length of about 210 to 220 amino acids. An Ig chain is composed of structural domains referred to as “Ig domains”. α, δ, and γ heavy chains are composed of one variable domain, three constant domains and a hinge region which increases flexibility. ε and μ heavy chains are composed of one variable domain and four constant domains. κ and λ light chains are composed of one variable domain and one constant domain. Ig domains generally have a length of about 110 amino acids. Each variable domain comprises hypervariable regions or loops which are responsible for epitope binding and which are referred to as complementary determining regions (CDR), and less variable regions between the CDRs referred to as framework regions (FR). Full mammalian VH and VL domains comprise three CDRs, referred to as CDR1, CDR2 and CDR3 and four framework regions, referred to as FR1, FR2, FR3 and FR4. By contrast, a truncated VH and VL domain lacks at least one of the CDRs or at least one of the framework regions, for example via the occurrence of a stop codon within the nucleotide sequence encoding said truncated VH and VL domain. Similarly, a truncated scFv lacks at least one of the CDRs or at least one of the framework regions of the VH and/or VL domain. A truncated scFv can consist in a single full VH or VL domain. The heavy chain CDR3 is the main contributor to the interaction with the antigen. All Ig domains have a characteristic compact globular structure, referred to as the “Ig-Fold”. The structure of the Ig-fold is well known to the skilled person. Briefly, it consists of a two-layer sandwich of 7 to 9 antiparallel β-strands arranged in two β-sheets with a Greek key topology. The backbone switches repeatedly between the two β-sheets. Typically, the pattern is (N-terminal β-hairpin in sheet 1)-(β-hairpin in sheet 2)-(β-strand in sheet 1)-(C-terminal β-hairpin in sheet 2). The two β-sheets are also connected to each other by a disulfide bridge. The CDR regions form loops which are held on the outside of the sandwich structure.
A conventional antibody digested by papain yields three fragments: two Fab fragments (fragment antigen binding) and one Fc fragment (Fragment crystalizable). A Fab fragment is a region of an antibody that binds to antigens. It is about 50 kDa and is composed of one constant and one variable domain of each of the heavy and the light chain. The two variable domains (VL and VH) shape the antigen binding site (paratope). The Fc region is the tail region of an antibody which interacts with cell surface receptors (Fc receptors). A conventional antibody digested by pepsin yields two fragments: a F(ab′)2 fragment (fragment antigen binding) and a pFc′ fragment. The F(ab′)2 fragment comprises two antigen binding sites and can be split in two Fab′ fragments by mild reduction. The variable regions of a heavy and a light chain can be fused together to form a single chain variable fragment (scFv), which retains the specificity of a Fab fragment while being only half its size (about 25 kDa). The VL and VH domains of an scFv are generally connected to each other by a short peptide linker of about 15 amino acids. The linker is usually rich in glycine for flexibility, as well as serine or threonine for solubility. Bi-valent or trivalent scFv can be engineered by connecting two or three scFvs together. Bi-valent scFvs (a form of diabodies) are of interest because they have a very high affinity for their target. Single-domain antibodies (sdAb) are antibodies of about 12-15 kDa which are composed of a single variable domain (a VL or VH domain). sdAbs can for example be composed of a camelid VHH domain, of a VH or a VL domain of a conventional antibody, or of a recombinant VH or a VL domain. One advantage of scFvs, diabodies and sdAbs is that they can be expressed as a single peptide and can for example be produced in bacteria and displayed on phages.
According to the present invention, an antibody (or “Ab”) is a protein which comprises at least one VL or VH domain of an immunoglobulin.
According to the present invention, an intrabody (or “intracellular antibody”) is an antibody, for example an scFv, a diabody, an sdAb, which is suitable for use in an intracellular environment. In particular, the antibody is able to fold in the reducing conditions of the cell cytosol and/or nucleus and is stable in an intracellular environment. The intrabody can in particular comprise a full VH domain, a full VL domain, or both. The intrabody can in particular be an scFv, a truncated scFv comprising at least a full VH or VL domain, a diabody or a VH or VL domain. Preferably, the intrabody is an scFv or a truncated scFv comprising at least a full VL or VH domain, most preferably a full VH domain.
According to the present invention and in accordance with the common meaning of the term (Muyldermans, 2001) a VH or VL domain refers to a conventional VH or VL domain, namely to a VH or VL domain of an antibody which is not a camelid antibody. In other terms, the term “VH domain” excludes VHH fragments, especially camelid VHH fragments. Thus, in a specific embodiment of the present invention, the VH domain and, preferably the VL domain is a non-camelid antibody, or is derived from a non-camelid antibody, preferably is a mammalian non-camelid antibody or is derived from a mammalian non-camelid antibody. In another embodiment, the VH and, preferably, the VL domain is not a non-cartilagenous fish antibody or is not derived from a non-cartilagenous fish antibody.
The distinction between conventional VH or VL domains and camelid VHH finds a basis in a number of sequence and structural differences (Muyldermans, 2001). For example, CDR1 and CDR2 loops of camelid VHH adopt structures that fall outside the canonical structures described for VH or VL domains of conventional antibodies. Moreover, VH domains comprise 4 canonical hydrophobic amino acids in their FR2 region, which take part in the interface with the VL domain. These amino acids are mutated in VHHs. These mutations, Val37Phe (or Tyr), Gly44Glu (or Gln), Leu45Arg (or Cys), and Trp47Gly (or Ser, Leu, or Phe), (Kabat numbering, reference in Kabat et al., 1991 and in Kontermann and Dübel, 2010) are highly conserved within the camel antibodies and are a key feature distinguishing them from a conventional VH domain. The CDR3 loop of VHH antibodies is also longer (16 to 18 amino acids, commonly 17 for the camel VHH) than the CDR3 loop of conventional antibodies (12 amino acids for human antibodies, 9 amino acids for mouse antibodies).
VHHs have been recognized as having an ability to bind recessed antigenic sites, which has been attributed to their smaller size and the ability of the extended CDR3 loop to penetrate into such sites. However, the single-domain nature of VHHs can be a disadvantage for binding to small antigens because these can bind in a groove or cavity at the VH-VL interface.
In a particular embodiment, a VH domain according to the invention comprises one or more, preferably all of the amino acids V37, G44, L45 and W47, with reference to the Kabat numbering scheme.
Alternatively, a VH domain according to the invention comprises the motif VNNNNNNGLNW, preferably in the FR2 region, wherein each N is an amino acid, independently selected from each other, preferably selected among the 20 canonical amino acids of the genetic code.
In addition, or alternatively, a VH domain according to the invention can also comprise a variant of the motif VNNNNNNGLNW, preferably in the FR2 region, wherein each N is an amino acid, independently selected from each other, preferably selected among the 20 canonical amino acids of the genetic code, and wherein the variant comprises one or more, such as one to two or one to three, amino acid additions and/or one or more, such as one to two or one to three amino acid deletions of any “N” amino acid.
The CDR3 loop of the VH domain and/or VL domain preferably comprises from 1 to 15 amino acids, still preferably from 1 to 12 amino acids.
The use of a high diversity library of intrabodies, containing not only full scFvs but also truncated scFvs, gives access to a high diversity of epitopes. While full scFvs preferably interact with the surface of a target protein or with small antigens, truncated scFvs, such as sdAbs, because of their smaller size, are more adapted to interact with protein cavities. Full scFvs are thus more adapted to disrupt protein-protein interactions, whereas truncated scFvs, such as sdAbs are more adapted to interact with enzymatic sites or with small molecule binding sites on their target antigen. The presence of both types of intrabodies is therefore highly advantageous in a library used for screening an unknown target.
In a specific embodiment, the intrabody comprises a VL and/or VH domain derived from a human antibody. Still preferably, the intrabody is a scFv derived from a human antibody, a truncated scFv comprising at least a full VL or VH domain derived from a human antibody, a diabody derived from a human antibody or a VL or a VH domain derived from a human antibody. More generally, the intrabody can be derived from a human antibody.
As meant herein, a “VL or VH domain derived from a human antibody” signifies a VL or VH domain which is identical or essentially identical to a VL or VH domain of a human immunoglobulin. By “essentially identical”, it is meant that the VL or VH domain can consist in a VL or VH domain of a human immunoglobulin in which modifications have been introduced into one or more CDR regions, preferably into the CDR3 region. The modifications can also be introduced into the framework regions, provided that they do not adversely affect the intracellular folding and/or the intracellular stability of the VL or VH domain.
The modifications can consist of point mutations, additions and/or deletions of one or more amino acids, for example of one to twenty amino acids, especially of one to fifteen amino acids, more particularly of one to twelve amino acids. When the modifications are introduced into one CDR region, from one to all amino acids of the CDR can be modified. Preferably, the modifications are introduced so as to have the same representation of the amino acids as that observed in natural human CDRs or framework regions corresponding to the modified CDR or framework region. The modifications are also introduced so that the “VL or VH domain derived from a human antibody” is still recognized as human when using the below-mentioned test for assessing the species of origin of an antibody. In other terms, when the sequence of a “VL or VH domain derived from a human antibody” is aligned with a library of immunoglobulins, or fragments thereof, of various species, including homo sapiens, the “best hit” corresponds to a human immunoglobulin or a fragment thereof.
As meant herein, a scFv derived from a human antibody comprises a VL domain derived from a human antibody and a VH domain derived from a human antibody. Similarly, a diabody derived from a human antibody comprises two sdAbs derived from a human antibody.
Mutatis mutandi, the term “derived from an antibody of a specific species or group of species” (for example “derived from a non-camelid antibody”) should be understood in a similar way as “derived from a human antibody”.
In another embodiment, the intrabody comprises a humanized VL and/or VH domain. Still preferably, the intrabody is a humanized scFv, a truncated scFv comprising at least a full humanized VL or VH domain, a humanized diabody or a humanized VL or a VH domain. More generally, the intrabody can be a humanized intrabody.
By “humanized” intrabody, it is meant that the intrabody has been obtained by engineering a non-human antibody so that it is less immunogenic for humans. Preferably, the intrabody has been obtained by “CDR grafting”, i.e. it comprises human framework regions, and one or more CDR regions originating from a non-human species, preferably a non-camelid non-human species. The framework can be a “fixed framework”, i.e. the same framework for all humanized antibodies, or can be chosen using the “best fit” approach, i.e. it is the human framework with matches at best the non-human framework to be replaced, when performing a sequence alignment.
The species of origin of an antibody (for example a human antibody) can be assessed by analyzing the sequence homology of said antibody with immunoglobulins of different species, and by determining the species of the immunoglobulin(s) with which said antibody shows the highest homology. This analysis can be carried out, for example, by consulting the database of the International Immunogenetics Information System (www.igmt.org).
By using a library of intrabodies derived from human antibodies, a screening method according to this embodiment is particularly well adapted for identifying targets in human cells. It can also identify intrabodies which are directly valuable for a therapeutic use in human, because of their non-immunogenicity.
According to a specific embodiment the intrabody is a 3H2-1 intrabody, i.e. an intrabody comprising the CDR3 sequence of the VH domain of scFv 3H2-1 as set forth in
According to another specific embodiment, the intrabody is a 3H2-VH intrabody, i.e. an intrabody comprising the CDR3 sequence of the VH domain of 3H2-VH as set forth in
According to another specific embodiment the intrabody is a 5H4 intrabody, i.e. an intrabody comprising the CDR3 sequence of the VH domain of scFv 5H4 as set forth in
According to another specific embodiment the intrabody comprises one of the VH CDR3 sequences set forth in SEQ ID NO:7 to SEQ ID NO: 18. Preferably, the intrabody comprises one of the VL CDR3 sequences set forth in SEQ ID NO:19 to SEQ ID NO: 29 or comprises a glutamine as VL CDR3 sequence. Preferably still, the intrabody comprises one of the combinations of VH and VL CDR3 sequences (also referred to as H3 and L3 respectively) set forth in Table 1. Preferably still, the intrabody comprises one or more of the CDR1 and CDR2 sequences of the VH and/or VL domain of scFv 13R4 as set forth in
As used herein, the term “SEQ ID NO:7 to SEQ ID NO: 18” may refer to each single sequence of the group, namely SEQ ID NO:7, SEQ ID NO: 8, SEQ ID NO:9 etc. . . . , each considered individually. The term “SEQ ID NO:7 to SEQ ID NO: 18” may also refer to any combination of sequences in the group, such as “SEQ ID NO:8 and SEQ ID NO: 13”, or “SEQ ID NO:9 and SEQ ID NO:12 and SEQ ID NO:16”, these examples being purely illustrative and non limiting. The same applies to the terms SEQ ID NO:19 to SEQ ID NO:29.
In a particular embodiment of the method of the invention, step a) comprises the screening of an intracellularly expressed intrabody library, to identify an intrabody which can induce, modify or suppress the cell phenotype. Preferably, each intrabody in the library comprises a full VH and/or VL domain.
An Intrabody library can for example be obtained by selecting an intrabody which is functional inside a cell. An intrabody functional inside a cell is preferably an intrabody that, when expressed inside the cell, is able to bind an intracellular target, especially to modify a phenotype of the cell. On average, only about 0.1 to 1% intrabodies are suitable for intracellular use. This proportion can be significantly increased in particular by using the method described in WO2008/110914 and in Philibert et al., 2007. Briefly, this method first comprises the selection of an intrabody, for example a scFv such as 13R4 (SEQ ID NO: 6), which is stable in vitro and shows optimal folding proprieties in the cell cytosol, and then the introduction of modifications in one or more of its CDR regions, preferably in the CDR3 region of the VH and/or VL chain (also referred to as H3 and L3 respectively), more particularly in the heavy chain CDR3 (H3). The modifications can consist of point mutations, additions and/or deletions of one or more amino acids, for example of one to twelve amino acids. Mutations can also be introduced into the framework regions.
In a specific embodiment, said intrabody library and/or said library of molecules is obtained by selecting an intrabody which is functional inside a cell, and then introducing modifications in one or more of its CDR regions, especially the CDR3 region of the VH and/or VL domain.
In another embodiment, the intrabody library comprises scFvs and truncated scFvs comprising at least a full VH or VL domain.
Preferably, the intrabody library comprises at least 105, more preferably at least 106, 107, 108, 109 or 1010 different intrabodies.
Using a high diversity library of intrabodies increases the number of possible targets. A high number of intrabodies can not only give access to the whole diversity of intracellular targets, including the various post-translational variants of a protein, but it can also give access to various epitopes on the same target. Different mechanisms of action of a single target can be targeted, thus increasing the success rate of the screening method.
In a preferred embodiment, step a) comprises:
According to a specific embodiment, steps iii) to iv), forming a selection round, are repeated more than once. This means that the cells selected from step iv) are recultured according to step iii) for a further selection step. For example 2 to 10 selection rounds, preferably 5 to 8, may be performed prior to step vi). Further, the method may comprise a recloning step after one or more selection rounds, i.e. after step iv) or v). The recloning step may for example be performed after each selection round or after some of the rounds, for example each two rounds. Thereafter the recloned cells may be submitted to one or more selection rounds. The recloning procedure may comprise the extraction of the DNA from the selected cells after step iv), the amplification of the scFv sequences by PCR and their recloning into the expression vector and their transfection again into cells. This allows for a more marked phenotype at each round of selection and for enrichment in scFv sequences which generate the studied phenotype.
Preferably, the molecules encoding the intrabodies are vectors. Preferably the vectors are suitable for expression in eukaryotic cells, for example in mammalian cells, such as human, rodent and primate cells. The vectors may be non-integrative vectors which allow for transient expression, for example plasmids or adenoviral vectors, or integrative vectors such as retroviral vectors. The vector may also encode a signal which directs the intrabody to a specific cellular compartment or to the cell membrane.
In particular when using a non integrative vector, a high number of copies, for example between 100 and 100000 copies, preferably between 500 and 20000 copies, for example about 10000 copies may be transfected per cell. According to this embodiment, less cells are required. Additional subcloning steps are performed subsequently to identify the intrabody of interest. This embodiment is in particular suitable for screening for modifications in a dominant phenotype.
When using an integrative vector, it is preferred that less than five copies, more preferably one, two or three copies are transfected per cell.
The cell can be transfected in vitro or in vivo as part of a whole organism. If the cell is transfected in vivo, the cell phenotype can be a phenotype which may be determined in the whole organism.
The intrabody may be fused to another protein, for example a protein which causes a detectable signal such as the GFP protein.
The cellular target may be a fully intracellular or partially intracellular molecule. In the case of a partially intracellular molecule, it can be a membrane protein, which may comprise an intracellular and/or an extracellular domain. The cellular target may in particular be a protein, a peptide, a carbohydrate, a lipid or even a nucleic acid. If the target is a protein, it may be for example an enzyme such as a kinase or a protein phosphatase, a transmembrane receptor, a transcription factor, a scaffolding protein such as the tubulin, metabolic enzymes, proteins implicated in protein synthesis and turnover such as ribosomal proteins, chaperones and proteases . . . . The cellular target may be endogenous or exogenous to the cell. Exogenous targets include in particular proteins, nucleic acids, particles of viral origin as well as whole viruses, bacterial antigens. The cellular target may also be an unknown protein or a known protein whose function is unknown.
The cellular target is a target which interacts with the intrabody either directly by binding to it (direct target), or indirectly in the sense that in does not bind to the intrabody but that the intrabody has an indirect effect on it. A direct target may be a target for which the dissociation constant (Kd) corresponding to its binding with the intrabody in vitro is at least 1 nM. An indirect target can also be a molecule which interacts with another cellular molecule with which the intrabody interacts also, or can be a molecule which is part of a cellular pathway which is modulated by the intrabody. For example, an indirect target may be a molecule which can be precipitated with the intrabody because it binds to another cellular molecule to which the intrabody also binds.
The interaction, whether direct or indirect, between the intrabody and the target may have a modulatory effect on or more several cellular pathways. This interaction can for example change the conformation and/or the phosphorylation status of the target, and thereby induce, enhance or repress its ability to interact with one or more other molecules, in particular proteins, inside the cell. This interaction may completely or partially activate or inactivate the target and/or another protein.
The cell phenotype can be any observable or measurable trait of the cell, including a morphological feature, a developmental stage, a biochemical or physiological property. For example the phenotype may be cell death, the onset of senescence, a resistance to an antibiotic, a resistance to a viral or bacterial attack, the expression or loss of expression of a receptor, the release of a compound outside the cellular environment. The phenotype may also be any type of signal, for example a signal, such as fluorescence, triggered by the expression of a recombinant reporter gene. Preferably, the cell phenotype is a phenotype associated with an animal or plant disease, preferably a pathology which affects a human being. Such a phenotype is thus a phenotype of therapeutic or diagnostic interest. In a particular embodiment, the phenotype is a phenotype associated with allergy or inflammation, in particular with a type I allergic reaction. The phenotype may be a trait which can be observed or measured in vitro on a cell culture or in vivo on a whole being.
The modification in the cell phenotype may be a modulation, such as an increase or a decrease in intensity, the induction (or the occurrence) or the suppression (or repression) of the cell phenotype.
The cell may be a prokaryote cell or a eukaryote cell. Preferably, the cell is a eukaryotic cell, such as a yeast cell or a higher eukaryote cell, for example an animal cell or a plant cell. In a specific embodiment, the cell is a mammalian cell, such as a mouse, rat or human cell. In a preferred embodiment, the cell is a human cell. According to a specific embodiment, the cell is a eukaryotic cell involved in allergy, inflammation, or both, and the cell phenotype is a phenotype associated with an allergic reaction, an inflammatory reaction, or both. For example the cell is a mast cell and the cell phenotype is degranulation and release of pro-inflammatory mediators.
The intrabody which is responsible for the phenotype induction, modification or suppression may be identified by any suitable means. For example, the cells which show the phenotype induction, modification or suppression are cloned and the DNA from each clone is extracted. The genes encoding the scFv are amplified by PCR and sequenced.
The direct or indirect target of the intrabody may be isolated by any suitable means. For example, the direct or indirect target of the intrabody can be isolated by a binding assay wherein the intrabody is marked with a detectable label. Alternatively, the direct or indirect target is immunoprecipitated together with the labelled intrabody. Once the target has been isolated, Mass spectrometry (MS-MS) can for example be used to identify it.
According to a specific embodiment, the method further comprises a step of target validation using RNA interference technology or a known inhibitor of said target such as an antibody. The method may also comprise a step of identification of the epitope of the target, and/or a step of characterisation of the target. The identification of the epitope is of particular interest as it may guide towards the identification of a site of interest on the target. In yet another embodiment, the method may comprise a step of identification of a molecule which competes with the binding of the intrabody identified in step a) to the protein identified in step b), and which is capable of modifying said cell phenotype.
Another aspect of the invention is an intrabody comprising the CDR3 sequence of the VH domain of scFv 3H2-1 as set forth in
A further aspect of the invention is an intrabody comprising the CDR3 sequence of the VH domain of 3H2-VH as set forth in
A further aspect of the invention is an intrabody comprising the CDR3 sequence of the VH domain of the intrabody 5H4 as set forth in
Another aspect of the invention is an intrabody comprising one of the VH CDR3 sequences set forth in Table 1 for use in therapy, especially for use in treating allergy and/or inflammation. Preferably, the antibody further comprises one of the VL CDR3 sequences set forth in Table 1. Still preferably, it comprises one of the combinations of VH and VL CDR3 sequences set forth in Table 1. Preferably, the intrabody comprises the framework sequences and CDR1 and CDR2 sequences of scFv 13R4 as set forth in
Another aspect of the invention is a nucleic acid sequence comprising or consisting in a sequence encoding an intrabody according to the invention as defined above, or a vector containing said nucleic acid sequence, or an eukaryotic cell containing said nucleic acid sequence or said vector.
Preferably, the vector is suitable for introduction and expression of the nucleic acid in mammalian cells, in mammalian tissue, or in a mammalian organ, for example a retroviral vector, a lentiviral vector or a plasmid.
Another aspect of the invention consists in the use of a 3H2-1 intrabody as defined above for identifying a molecule which is capable of competing with the binding of said intrabody with a protein of the ABCF1 family, preferably the ABCF1 protein, and of modifying a phenotype associated with an allergic reaction, an inflammatory reaction or both, in a cell involved in allergy, inflammation, or both.
The ABCF1 (ATP-binding cassette sub-family F member 1) protein is a member of the ABC protein superfamily. It is part of the subfamily F or GCN20 of the superfamily. The ABC sub-family F comprises other members, such as the ABCF2 and ABCF3 proteins.
A protein of the ABCF1 family can consist in a protein having an amino acid sequence with at least 60% identity, preferably at least 70% identity, most preferably at least 80% or at least 90% or least 95% or least 99% identity with the amino acid sequence set forth in SEQ ID NO:41. It can also consist in a protein encoded by a polynucleotide with at least 60% identity, preferably at least 70% identity, most preferably at least 80% or at least 90% or at least 95% or least 99% identity with the nucleotide sequence set forth in SEQ ID NO:40. It can especially consist into the human protein ABCF1 or into any ortholog of the human ABCF1 protein. It can also consist into any homolog or paralog of the human ABCF1 protein. It is to be noted that the gene encoding the ABCF1 protein is highly conserved in many species (for example in chimpanzee, Rhesus monkey, dog, cow, mouse, rat, zebrafish, fruit fly, mosquito, C. elegans, S. pombe, A. Thaliana and rice).
A further aspect of the invention consists in the use of a 5H4 intrabody as defined above for identifying a molecule which is capable of competing with the binding of said intrabody with a protein of the C12ORF4 or “LOC” family, and of modifying a phenotype associated with an allergic reaction, an inflammatory reaction or both, in a cell involved in allergy, inflammation, or both.
A protein of the C12ORF4 or “LOC” family can consist in a protein having an amino acid sequence with at least 60% identity, preferably at least 70% identity, most preferably at least 80% or at least 90% or at least 95% or at least 99% identity with one of the amino acid sequences set forth in SEQ ID NO: 33, SEQ ID NO:35, SEQ ID NO:37 and SEQ ID NO:39, preferably with the amino sequence set forth in SEQ ID NO:35. It can also consist in a protein encoded by a polynucleotide with at least 60% identity, preferably at least 70% identity, most preferably at least 80% or at least 90% or at least 95% or at least 99% with one of the nucleotide sequences of set forth in SEQ ID NO: 32, SEQ ID NO:34, SEQ ID NO:36, SEQ ID NO:38, preferably with SEQ ID NO: 34. It can especially consist into one of the proteins C12ORF4, LOC297607, LOC57102 or LOC28040, preferably into the human C12ORF4 protein or into any ortholog of the human C12ORF4 protein. It can also consist into any homolog or paralog of the human C12ORF4 protein. It is to be noted that the gene encoding the protein C12ORF4 is highly conserved in many species (for example in chimpanzee, dog, cow, mouse, rat, chicken, zebrafish, fruit fly, mosquito, and C. elegans).
The phenotype associated with an allergic reaction, an inflammatory reaction or both may be related to a disease or condition where allergic reaction or inflammatory reaction happen, especially Type I, II, III and IV hypersensitivity, rheumatoid arthritis, ankylosing spondylitis, and systemic lupus erythematosus (SLE).
The cells involved in allergy, inflammation, or both may be selected among mast cells and basophil granulocytes, phagocytic cells or phagocytes (granulocytes—mainly neutrophil- and scavenger cells), and lymphocytes.
According to a particular embodiment of one or the other of the uses above, the molecule which is capable of competing with the binding of the intrabody with the target protein is an organic molecule having a molecular weight of 100 to 2500 Da.
In yet another aspect, the invention is directed to an inhibitor of a protein of the ABCF1 family, in particular of ABCF1, for use in a method for treating allergy, inflammation, or both. Such an inhibitor is herein referred to as an “ABCF1 inhibitor”.
In a specific embodiment, the ABCF1 inhibitor according to the invention is for use in a method for treating allergy, inflammation, or both, especially for treating a disease or condition where allergic reaction and/or inflammatory reaction happen especially Type I, II, III and IV hypersensitivity, rheumatoid arthritis, ankylosing spondylitis, and systemic lupus erythematosus (SLE).
In yet another aspect, the invention relates to a method for manufacturing a medicament or a pharmaceutical composition for treating allergy, inflammation, or both, comprising:
In a particular embodiment, the ABCF1 inhibitor is
According to the invention, an antigen-binding fragment of an intrabody may comprise at least one CDR loop of a VH or VL domain, preferably the CDR3 loop of a VH or VL domain, still preferably of a VH domain.
In yet another aspect, the invention is directed to an inhibitor of a protein of the C12ORF4 family, also referred as “LOC” family, in particular of C12ORF4, LOC297607, of LOC57102 or of LOC28040, most preferably of C12ORF4, for use in therapy. Such an inhibitor is herein referred to as a “LOC inhibitor”. In a specific embodiment, the LOC inhibitor according to the invention is for use in a method for treating allergy, inflammation, or both, especially for treating a disease or condition where allergic reaction and/or inflammatory reaction happen especially Type I, II, III and IV hypersensitivity, rheumatoid arthritis, ankylosing spondylitis, and systemic lupus erythematosus (SLE).
In yet another aspect, the invention relates to a method for manufacturing a medicament or a pharmaceutical composition for treating allergy, inflammation, or both, comprising:
In a particular embodiment, the LOC inhibitor is
The heavy chain variable domain is shown in black. The light chain variable domain is shown in grey.
a) is a schematic view of an IgG. Disulfide bonds are indicated with dotted lines
b) is the representation of some monovalent and bivalent antibody fragments.
The scFv library was cloned in an expression vector (Step 1) and was transfected in a suitable cell line (Step 2). The cells were sorted according to a given phenotype in Step 3. In the described case, sorting was performed by FACS after cell staining with fluorescent Annexin V. After a suitable number of selection rounds, the enriched scFv population was used to generate stable clones and the scFv genes were sequenced (Step 4). The best scFv were produced in E. coli and they were used to identify their target by capturing it from cellular extracts and analyzing the captured proteins by Mass spectrometry (Step 5). The implication of the target in the given phenotype was then confirmed using usual techniques (Step 6).
In (a), the expression of the antibody fragments was measured by RT-PCR; and in (b) by immunofluorescence using anti-myc tag 9E10 monoclonal antibody, following the indicated selection rounds.
For each round of selection, the cell population transfected with the plasmidic intrabody library was analyzed by FACS after staining with Annexin-V-APC. Cells were either stimulated with IgE/DNP (+) or non-stimulated by omitting the DNP (−). The value of the shift is represented in (a) by the horizontal marker labeled M1. (b) shows the value of the Annexin-V-APC shift for rounds 1 to 7 of the selection.
Analysis of 24 stables RBL-2H3 clones generated from the selection round 7 following plasmidic intrabody library transfection; The RBL-2H3 cell line was used as a positive control.
Some stable RBL-2H3 clones generated from the selection round 7 were analyzed in parallel with an irrelevant non-inhibitory intrabody used as a positive control. IgE loaded cells were stimulated by adding the antigen DNP after 20 seconds and the rise of calcium was evaluated by FACS using Fluo 3 dye. The names of the tested clones are indicated in the figure.
a) Schematic view of 5H4-VH, 5H4-VL, 3H2-1 and 3H2-VH fragments.
b) The sequence of the CDR3 loops are indicated in comparison with the original scFv 13R4.
The arrow indicates the non-stimulated cell population (+IgE/−DNP); FcεRI stimulation-dependent Annexin V shift follows the incubation of cells with IgE and the antigen DNP (+IgE/+DNP).
To measure the calcium flux, the antigen DNP was added to IgE loaded cells and the rise of intracellular calcium was measured by FACS using the Fluo 3 dye.
Mass spectrometry analysis of the proteins captured from a cellular extract using (a) the scFv 3H2-1 and (b) 5H4-VH fragment.
The band corresponding to the protein with the highest Protscore is indicated with an arrow on the gels.
The sign “*” indicates the percentage of sequence identity using the human protein as reference.
The identity of the protein identified in
The identity of the protein identified in
The localization (in bold in the sequence) and the sequences of the two shRNA are indicated.
An RBL-2H3 cell population transfected with an irrelevent shRNA specific to Luciferase (shLUC) was used as a control.
FcεRI-mediated activation of the transfected RBL-2H3 cell populations was evaluated by the measure of: (a) β-hexosaminidase release; b) TNF-α secretion; c) rise of intracellular calcium flux. The RBL-2H3 cell line was used in parallel with the cell populations transfected either with shRNA specific to the Luciferase (shLUC) as a control, or with shRNA specific to LOC297607 (sh1).
Cell extracts were prepared 0, 3 and 10 minutes after FcεRI stimulation (+IgE/+DNP). Antibodies specific to the indicated targets were used.
The arrow indicates the non stimulated cell population (+IgE/−DNP); FcεRI stimulation-dependent Annexin V shift follows the incubation of cells with IgE and the antigen DNP (+IgE/+DNP). The percentages indicate the value of the shift.
a) selection including cloning steps
b) selection without cloning steps
The arrow indicates the non stimulated cell population (+IgE/−DNP); FcεRI stimulation-dependent Annexin V shift follows the incubation of cells with IgE and the antigen DNP (+IgE/+DNP).
One representative scFv gene from each cluster was transfected in the RBL-2H3 cell line and the cellular phenotype was studied. The indicated percentage is calculated in comparison to the results obtained with an irrelevant scFv fragment.
The phenotypic screening approach developed by the inventors is based on the selection, from a combinatorial library, of scFvs capable of inducing a phenotype of therapeutic interest. A phenotype associated with allergy was studied, using a cellular model of mast cell degranulation. The steps of the functional screening developed by the inventors is described in
Two parallel approaches were developed, using either a plasmid expression system, or a retroviral system. This generic approach may be used for the study of multiple cellular phenotypes.
Step 1: Cloning of the scFv Library
The present strategy uses a combinatory library of human scFvs of high diversity (109), allowing in theory the coverage of all the proteins expressed in cells (about 106, when taking into account alternative splicing and posttranslational modifications). Moreover, this library was built using a constant framework allowing for a great stability in the cytosol, so as to optimise intracellular expression of the scFvs in the cytoplasm of eukaryote cells. The framework antibody is the 13R4 human scFv, having the sequence set forth in
These two sub-libraries were assembled by PCR and cloned into the eukaryote expression vector pEF/myc/cyto (Persic et al., 1997). This vector comprises a strong promoter of the type EF1a, and is adapted to cytoplasmic expression of the scFvs. It also allows the inclusion of a C-terminal c-Myc tag (EQKLISEEDL), which is recognised by monoclonal antibody 9E10.
After bacterial transformation, a library having a diversity of 109 was generated.
Mast cell line RBL-2H3 (a line derived from a subcloning of RBL, ‘rat basophilic leukaemia’) was used for this study. This line is commonly used in the field of allergy and its signalling pathways have been well studied.
These cells were transfected by electroporation with the previously cloned vector pEF/myc/cyto. The voltage, capacitance, and biochemical conditions relative to the electric shock were chosen such as to maintain as much as possible the initial diversity of the scFv library. The parameters to be considered therefor are:
In this manner, multicopy transfection conditions were attained, where each cell was transfected with 2000-2500 recombinant plasmids. Reverse transcription experiments, followed by qPCR (RT-qPCR) showed that in these conditions, the transcription of a single gene copy of one scFv among 2500 could be detected. Thus, in the first selection round, 3.108 cells were electroporated with 3 mg DNA. This allowed in theory for a representation of about 50 times the scFv library while taking into account the previously cited parameters.
Intracellular expression of the scFv library was controlled by performing an RT-qPCR analysis of the sequences expressed by the cells obtained after each selection round (see
For each selection round, transfected cells were stimulated via the FcεRI (high affinity IgE Receptor) by addition of IgE coupled to its Ag, dinitrophenyl (DNP). Immediately after the stimulus, the cells were labelled with Annexin V coupled to APC (allophycocyanin). This marker specifically binds to the phosphatidylserines exposed at the plasma membrane during exocytosis proportionally to the intensity of degranulation (Demo et al. 1999). The fluorescent Annexin V allowed a distinction to be made with a flow cytometer between cells which degranulate (strongly marked) and cells whose activation is inhibited (faintly marked). Marking with Propidium iodide (PI) allowed the exclusion of apoptotic cells. When sorting by FACS, a population of cells transfected with an empty vector was used as control.
A region corresponding to 0-10% of the least degranulating cells which a priori express the inhibiting scFvs was sorted. After each selection, the plasmid DNA of the sorted cells was extracted and the sequences of the scFvs were amplified by PCR and recloned into the same expression vector in order to transfect RBL-2H3 cells and carry out a new selection round. 8 enrichment rounds were thus performed. The analysis of the Annexin shift along selection revealed a marked decrease of cell degranulation (
After selection, the DNA extracted from the cells was used to generate stable transfectants of rat mast cell line RBL-2H3. The stable transfectants were obtained by electroporation of the extracted DNA, followed by a selection with the antibiotic G418 (Geneticine) at the final concentration of 2 mg/mL. The stable cell transfectant clones were obtained by limited dilutions. 132 stable clones were isolated and their phenotype associated with activation of the FcεRI was analysed by the measurement of β-hexosaminidase release, which is one of the preformed allergic mediators released by exocytosis.
Phenotype of the Stable Clones
The cellular test was performed in duplicate.
After the first screen, 34 clones were selected and analysed using a second cellular test consisting in the measurement of the intracellular calcium flux. The calcium flux is triggered a few seconds after activation of the FcεRI by the release of intracellular stocks of Ca2+, followed by a calcium influx. Cell marking was performed with a calcium sensor, Fluo 3-AM, whose fluorescence detected by FACS, is proportional to the amount of intracellular Ca2+.
Sequence Analysis
The genomic DNA of 34 clones presenting an inhibition of β-hexosaminidase was extracted, and the genes of the scFv expressed by these cells were amplified by PCR, and sequenced. This analysis revealed that 65% of the cellular clones presented a mixture of 2-3 sequences. In addition, 74% of the scFvs from the phenotypic selection presented complete sequences. However, 14% of the analysed were truncated at the CDR3 of the VL (L3), 3% at the CDR3 of the VH (H3), and 9% had neither H3 nor L3.
Two clones, 3H2 and 5H4, presenting pronounced and reproducible degranulation inhibition phenotypes, were retained for further study (
Clone 3H2
The three cellular tests allowing the evaluation of the degranulation of this clone were performed in parallel, and showed that β-hexosaminidase release was inhibited by 59% (
New stable transfectants of the RBL-2H3 line were generated and clones expressing scFv 3H2-1 and 3H2-VH were analyzed. Three cellular tests allowing the evaluation of the degranulation of these clones were performed, and showed that β-hexosaminidase release was strongly inhibited in cells expressing 3H2-VH (
Clone 5H4
The analysis of the sequence of the scFv expressed by this clone revealed the presence of a stop codon at the beginning of the L3 (
Therefore, two 5H4 formats were recloned into the pEF/myc/cyto vector: 5H4-VL corresponding to the sequence of the original clone; and 5H4-VH which comprises only the VH domain (
New stable transfectants of the RBL-2H3 line were generated. The cellular tests for these populations revealed that the stimulation of the FcεRI resulted in a strong reduction of calcium mobilisation (
Step 5: Production of Selected scFvs and Identification of their Target
Production of the scFvs
So as to use the selected Ab fragments for identifying their target, the three sequences 3H2-1, 5H4-VH and 5H4-VL, were cloned into two prokaryote expression vectors allowing their production from E. Coli cytoplasmic (pET23 vector) or periplasmic (pHEN2 vector) extracts.
They were also cloned into a eukaryote expression vector (ps1119) allowing their production fused with a mouse Fc, in the form of dimers (Moutel et al. 2009).
Target Identification
The three Ab fragments, as well as the appropriate irrelevant Abs, were purified and immobilised on magnetic nickel beads thanks to their 6×His tag. They were used to immunoprecipitate protein lysates from RBL-2H3 cells stimulated or not beforehand via the FcεRI. After the immunoprecipitations, the beads were washed in low stringency conditions in order to preserve low affinity bindings.
After electrophoresis on acrylamide gel (SDS-PAGE), and staining with Coomassie blue, bands present when immunoprecipitated by Ab fragments 3H2-1 and 5H4 and whose intensity was either higher as compared to the irrelevant Ab, or who were absent with the irrelevant Ab, were cut. After digestion with trypsin, the bands were analysed by mass spectrometry MS/MS using the MALDI-TOF (matrix-assisted laser desorption/ionization time of flight) technique.
The Mass spectrometer which was used is composed of a laser which ionises previously digested peptides; an analyser which measures the mass/charge (m/z) ratios of the ionised peptides as a function of their flight time; and a detector which records the number of ions for each given m/z ratio. The obtained spectra of m/z ratios allow the identification of peptides and thus proteins present in the samples, by interrogating the Genebank rat database, as well as the mouse database for comparison (the mice Genebank data being more complete). A confidence index or ProtScore is attributed for each protein as a function of the number of identified peptides which are part of its sequence and of the nature of these peptides (composition, size). A ProtScore higher than 3 corresponds to 0.1% probability of identifying a false positive, in other words the probability that this protein is present is 99.9%. The proteins with Protscores the highest for the tested Ab fragments and absent or insignificant for the irrelevant Ab were kept.
In the case of scFv 3H2-1, 10 proteins were selected at the end of this analysis, with Protscores ranging from 4 to 51 against 0 to 2 for the control. 95 kDa protein ABCF1 (ATP binding cassette sub-family F member 1) which had the highest Protscore was retained (
In the Table, the columns are the following: Column 1, Protscore of the identified proteins using the intrabody. Column 2, Protscore of the identified proteins using an irrelevant intrabody. Column 3, Genbank gi number of the identified protein. Column 4, name of the protein in Genbank. Column 5, predicted molecular weight.
This protein is a particular member of the ABC transporters family because it is the only one which does not comprise a transmembrane domain. Several studies indicate that it could have a role in translation initiation by interacting with the eIF2 factor (eucaryotic initiation factor 2) (Paytubi et al., 2009, 2008; Tyzack et al., 2000). Previous work reported that the gene encoding ABCF1 was regulated by TNF-α in synoviocytes. In this context, this protein may take part in the increase of protein synthesis and in the inflammatory process (Richard et al., 1998). In addition, the locus of the gene coding for ABCF1, located in the HLA region (human leukocyte antigen), may be associated with a predisposition for an autoimmune disease (Ota et al., 2007).
In the case of 5H4, the comparison of the results obtained for 5H4-VH and 5H4-VL showed that the proteins immunoprecipitated by the two formats were the same. Therefore, only 5H4-VH was used in the further experiments. The MS/MS results allowed the selection of 8 proteins with Protscores ranging from 4 to 22 versus 0 to 2 in the control. The 64 kDa protein LOC297607 (hereafter “LOC”), with a ProtScore of 16 to 22 for three independent experiments was retained (
In the Table, the columns are the following: Column 1, Protscore of the identified proteins using the intrabody. When several experiments have been performed, the 3 values separated by a “/” are reported. Column 2, Protscore of the identified proteins using an irrelevant intrabody. Column 3, Genbank gi number of the identified protein. Column 4, name of the protein in Genbank. Column 5, predicted molecular weight.
This protein whose function was unknown until now is found in higher eukaryotes, and is ubiquitously expressed. Its sequence, identified using cDNA libraries, is highly conserved among species (
3H2-1 Potential Target: ABCF1
In order to confirm the protein identification, the IP (immunoprecipitation) experiment conducted for mass spectrometry was repeated and followed by an immunoblot revealed by a commercial Ab directed against the ABCF1 protein (
An IF marking of RBL-2H3 cells with 3H2-1 in a dimerised form (3H2-Fc), as well as with the commercial Ab directed against ABCF1 was performed. These analyses revealed similar cytoplasmic markings (
5H4 Potential Target: LOC297607
In the same manner as for the mass spectrometry analysis, an IP of the RBL-2H3 lysates was performed with 5H4-VH, and revealed by WB with a commercial Ab directed against the LOC protein (
This result was confirmed by a double marking by IF analysis with a confocal microscope. This experiment revealed a cytosolic marking of the LOC protein, and a colocalisation of signals corresponding to 5H4-VH and the commercial Ab directed against LOC (
5H4 Target: LOC297607, Confirmation by shRNA
Inhibition of the LOC protein expression by shRNA (short hairpin RNA), followed by a study of the associated phenotype was performed in order to validate the implication of the LOC protein in mast cell activation by an independent approach. Therefore, two shRNA, sh1 (SEQ ID NO: 30) and sh2 (SEQ ID NO:31) directed against the LOC protein were constructed using the Dharmacon software (http://www.dharmacon.com/designcenter); as well as one control irrelevant shRNA: shLUC directed against luciferase. The corresponding sequences of the shRNA are presented in
The shRNA were cloned into the retroviral expression vector pSIREN to infect the RBL-2H3 line. A selection with puromycin allowed the recovery of only the infected cells expressing the different shRNA. Cellular tests detailed below were performed 5, 10 and 15 days from the beginning of the puromycin selection and on three independent series of shRNA infected cells.
In order to control the extinction of the expression of the protein, an RT-qPCR was first carried out on the RNA extracted from the different types of cells.
a shows the results obtained with the infected cells, after normalisation with the housekeeping gene HPRT encoding hypoxanthine-guanine phosphoribosyltransferase. This experiment revealed that sh1 expression extinguishes 85% of the messenger of this protein at day 5, and from 66 to 73% at days 10 and 15 of selection. The extinction of the LOC messenger by sh2 showed similar results, ranging from 58 to 64% inhibition between days 5 and 15 of selection.
The extinction of the protein was also confirmed by WB. Protein extracts from cells expressing the different shRNA were revealed with the commercial Ab directed against LOC. After normalisation of charges between the tracks, the loss of intracellular LOC protein when sh1 was expressed was evaluated at 50-71% (depending upon the series) after 5 days, and up to 88% after 15 days of selection (
Throughout the puromycin selection, cellular tests were performed to evaluate the impact of the shRNA on mast cell activation. The measurement of β-hexosaminidase revealed that after 5 days of selection, the cells expressing sh1 present a significant inhibition of 28% (
Following mast cell activation, signal transduction is mediated by a balance between the activity of protein tyrosine kinases (PTK) and protein tyrosine phosphatases (PTP). This results from an increase in phosphorylation of a number of intracellular substrates. The protein extracts of cells non stimulated or stimulated via the FcεRI were analysed by WB with monoclonal Ab 4G10 directed against the P-Tyr. It was noted that a reduction in LOC expression affected phosphorylation of certain intracellular proteins migrating to around 130, 75, 50-60 kDa (
These preliminary results confirm the involvement of the protein LOC in mast cell activation following FcεRI activation. Indeed, during the first week following infection by the shRNA, a partial extinction of the protein in the cell is at the origin of a phenotype of inhibition of the calcium flux and the release of preformed and neosynthesised allergic mediators (β-hexosaminidase and TNF-α).
The structure of the LOC protein being unknown, a search for conserved motifs was carried out, which revealed potential phosphorylation sites of tyrosine and serine residues. We also identified two adjacent <<ITIM-like>> motifs similar to the motifs that are present in inhibitory receptors of immune cells, and that are capable of recruiting potential partners via their SH2 domains (amino acids 319-342: SLYSTSLCGLVLLVDNRINSYSGI)
The phenotypic screen was adapted using a retroviral system for expression of the scFv library in the same mast cell line. The objective was to optimize this generic approach to apply it to the study of other phenotypes. The principle is the same as for the approach described in example 1, with the exception of selection: indeed, the viral approach a priori allows the avoidance of the prior recloning steps of the scFv library, and the use of the same population throughout the selection.
Step 1: Cloning of the scFv Library
The same scFv library as for the plasmid selection (Philibert et al., 2007), was cloned into the pMSCV-hygro-GFP vector. This vector confers on the one hand, hygromycin resistance when cells are infected, and on the other hand, allows the production of the scFv fused to the GFP protein. Infection efficiency can thus be evaluated, and intracellular scFv expression can also be monitored and controlled throughout the entire selection. After the cloning of the library, it was amplified by transformation of E. Coli. The putative initial diversity was 2.108 clones.
The retroviral supernatants used were produced after co-transfection of 293T cells with genes coding for the scFv library, the capsid proteins, as well as for the amphotropic envelop proteins. This system allows the infection of the RBL-2H3 line with an efficiency of over 95%.
The number of copies per cell was checked by qPCR, and showed that each cell had between 1 and 3 scFv copies. In order to remain in manageable experimental conditions, 4.107 RBL-2H3 cells were infected. Thus, under these conditions, the diversity of the putative initial scFv library was approximately 108.
4 days after infection, cells were activated with IgE/DNP, labelled with Annexin V-APC and Propidium Iodide and analyzed by FACS (
A population corresponding to 10% of the least fluorescent cells was then sorted and put directly into culture, so as to amplify the cells for the following selection a week later. Eight rounds of selection were performed in this manner.
Inhibition of the phenotype appeared in the 6th round of selection. Indeed, the shift in Annexin V-APC between the non-stimulated and stimulated with IgE/DNP cell populations, went from 51% in the first round to 16% in round 6, and 15% in round 7.
In order to control that the appearance of the phenotype came from the enrichment in inhibitory scFv and was not due to cell bias, such as a population drift, or an enrichment of scFv according to their genomic insertion site, a second selection of the retroviral library was performed including recloning steps.
Starting from the DNA of cells derived from selection round 3, the scFv sequences were amplified by PCR and cloned into the pMSCV new-hygro-GFP vector. After transformation into the virus, the infection was carried out on a new cell population. This step of recloning was repeated every 2 rounds of selection, in rounds 5′ and 7′. The results of the marking with Annexin V-APC show that this method of selection also allowed the convergence towards an inhibitory phenotype, which demonstrates that it was not a phenotypic drift of the cell population used (
The evolution of the degranulation phenotype of the cell populations derived from different rounds of selection was analysed. Aliquots of cells from different sorting steps were thawed and subcultured in order to achieve a measure of the release of β-hexosaminidase on entire populations. The results show a decrease of 75% of the degranulation of cells infected with the scFv library from round 7 (
Phenotype of the Stable Transfectants
An aliquot of cells from the retroviral selection including the recloning steps was seeded at limiting dilution on five 96-well plates. After a hygromycin selection, 50 cell clones were analysed in duplicate by measuring the release of β-hexosaminidase (
This analysis revealed that the degranulation of the 50 clones tested, was inhibited by 54% on average compared to 10 irrelevant clones.
High Throughput Sequencing
The evolution of diversity of the retroviral library during the two modes of phenotypic selection (with or without recloning steps) was analysed by high throughput sequencing (or NGS, “next generation sequencing”) using Illumina technology.
The sequences of the analysed scFv were amplified from 106 cells infected before the first screen (naive library), and from 5.105 cells from each of rounds 3, 5 and 7, with and without recloning. Ultimately 33.106 scFv were sequenced
From the 106 cells infected with the naive library and analysed, approximately 107 sequences of VH were obtained, among which 4-5.105 sequences were unique. This result suggests that the diversity of the library has not been affected significantly by retroviral infection. Indeed, the difference could be due to the fact that during the 3-4 days between the retroviral infection and the extraction of genomic DNA, cells would have had time to double, thus lowering the diversity by a factor of 2.
The sequencing revealed that over 99% of the read sequences all had H3 or L3 without frame shift or stop codon.
The translation of the unique nucleotide sequences generated 2-3.105 unique VH amino acid sequences and 3. 105 VL sequences. These results reflect the actual diversity of the library from the 106 cells analyzed. This diversity potentially increases with the number of possible combinations between VH and VL.
These sequencing analyses also revealed that starting with the same number of analyzed cells, the number of unique sequences is comparable between the direct and recloned retroviral selections, which suggests that the cloning steps do not seem to affect the diversity of the library.
Statistic Analysis
After the sequencing, statistical analysis of data was performed with the SAMSeq software (Significance analysis of microarrays) (Li and Tibshirani, 2011), which calculates a statistical score of enrichment for each sequence, and thus determines the subset of sequences significantly enriched during the selection. By setting a FDR (False Discovery Rate) of 0.05, 2500-10000 sequences appear to be enriched. The next step of the statistical analysis consisted in the comparison of the two selections in terms of enrichment of scFv sequences between different rounds, or Fold Change (FC) and frequency of occurrence, which corresponds to the presence of the same sequence in each library. The figures herein refer to the analysis of VH sequences. Similar results were obtained with VL libraries.
The sequences considered as enriched corresponded to sequences with an increase throughout the selections, and whose FC was greater than or equal to 100 between the naive library and round 7. In the end, 110 VH sequences were enriched with the recloned selection, and 107 with the direct selection.
Comparison of these sequences with each other allowed their grouping into clusters or families, with at least 60% identity between them. 71 out of the 110 sequences from the recloned selection were regrouped into 14 clusters (each containing 2-25 sequences), and 54 out of the 107 sequences of the direct selection were grouped into 17 clusters (2-7 sequences per cluster). Among these 31 clusters, we conserved those present at least at 0.1% in the final enriched library after 7 rounds. Finally, 10 clusters from the recloned selection and 3 clusters from the direct selection were selected and further studied (
Comparison of Retroviral Selections
Comparison of the FC and frequency of the VH sequences enriched after two selections revealed that:
In conclusion, in the light of this analysis, it appears that the recloned selection has a better enrichment than the direct selection, both in terms of frequency of occurrence and fold change. One hypothesis would be that without recloning, the selection could be biased by an enrichment in insertion sites favouring the appearance of the phenotype of interest. Thus the inclusion of recloning steps in the selection would optimise the enrichment in scFv inhibitors by limiting this bias.
Comparison of Retroviral and Plasmid Selections
Finally, 178 scFv sequences corresponding to the plasmid selection (stable clones and sequences stitched at random) were included in a second analysis in order to compare the scFv sequences enriched using plasmid and retroviral selections, in clusters rather than individually.
It is interesting to note that there is only one sequence in common to both selections, and that it is the VH domain of 5H4 (
This sequence presented an enrichment of 150 times using the retroviral selection.
Validation of Retroviral Clones
The most frequent sequence of each of the 13 clusters was cloned in the pMSCV-hygro-GFP retroviral vector and the cellular phenotype of retrovirus-infected cell populations were analyzed. As shown in
All antibodies used were obtained from Santa Cruz Biotechnology (Santa Cruz, Calif., USA), with the exception of phospho-specific Antibodies which were obtained from Cell Signaling Technologies (Danvers, Mass., USA). The anti-LOC (anti-C12orf4) antibody and all reagents were obtained from Sigma-Aldrich (St Louis, Mo., USA).
Rat Mast Cell Line RBL-2H3:
RBL-2H3 cells (ATCC, Manassas, Va.) were cultured in DMEM medium supplemented by 15% foetal calf serum (FCS) and antibiotics, at 37° C. in a humid incubator with 5% CO2. Adherent cells were passaged 3 times a week. The cells were detached by trypsin treatment during 5 minutes at 37° C., and subsequently inactivated by addition of two volumes of cell culture medium.
Line 293T (or HEK-T):
These cells were maintained in culture in DMEM medium supplemented with 10% FCS and antibiotics, at 37° C. in a humid incubator with 5% CO2.
Hybridoma.
The murine monoclonal antibody producing hybridoma were maintained in culture in DMEM medium supplemented with 10% FCS and antibiotics, at 37° C. in a humid incubator with 5% CO2. The culture supernatants containing the antibodies were filtered and preserved at 20° C. The antibody concentration was determined by an ELISA technique.
3.1. Cloning of the scFv Library
The eukaryote cytoplasmic expression vector pEF/myc/cyto (Invitrogen) was used to express the scFv library in RBL-2H3 cells. This vector comprised a promoter EF1a, followed by a NcoI site containing the initiatory ATG. The cloning was carried out between the NcoI and NotI sites, the latter being followed a c-Myc epitope detected by the 9E10 monoclonal antibody. In order to clone the scFv library into this vector, VHpool and VLpool sub-libraries, which are the source of the diversity of the CDR3 loops (P. Philibert et al., 2007), were assembled by PCR using the Pfu ADN-polymerase (Promega, Madison, Wis., USA) and the M13rev-49 and scFvCAT2.rev primers (35 cycles, 55° C.). Following purification, the PCR was digested with NcoI (Fermentas, Thermo Fisher Scientific, Waltham, Mass., USA) and NotI (New England Biolabs, Ipswich, Mass., U.S.) enzymes, and then purified and quantified on gel. The pEF/myc/cyto vector was also digested with NcoI et NotI enzymes, and subsequently purified and quantified on gel. The linearised vector was used for the equimolar ligation with the T4 DNA ligase insert. The ligation was inactivated 20 minutes at 65° C. and purified. It was used to electroporate C-Max5αF′ competent bacteria (Sidhu et al., 2000), which were subsequently plated on 600 cm2 square dishes of LB containing 100 μg/ml ampicillin. After 16 h incubation at 37° C., the bacteria were recovered with a rake in LB containing 10% glycerol and preserved at −80° C. in aliquots. An aliquot corresponding to 40 times the diversity of the library was used to prepare DNA with the Nucleobond Xtra Maxi kit (Macherey Nagel). This DNA corresponds to the library used subsequently for the transfection of the RBL-2H3 line. Ampicillin resistant bacteria were counted in order to estimate the initial diversity of the scFv library. In these conditions, the estimated size of the library cloned in the pEF vector was of 109 clones. For expression of the scFv library by retroviral infection, the pMSCV-hygro-GFP vector (Clontech) comprising the packaging sequence Ψ was used. The cloning was performed between the Sfi et NotI sites. For this cloning, the estimated size of the library cloned into the pMSCV vector was of 2.108 clones. For the recloning steps during the phenotypic selection, the scFv sequences were amplified from genomic DNA of the sorted cells, by PCR with pMSCV.for and EGFP-N.back primers, using Taq Phusion. The insert corresponding to the library was digested by Sfi and NotI enzymes, purified on gel, and cloned again into the pMSCV-hygro-GFP vector by ligation. As previously, the plasmid DNA was amplified by CMax5αF′ bacterial culture, and purified by kit.
3.2. Plasmid Transfection of the scFv Library
The trypsinised RBL-2H3 cells were washed once by centrifugation with culture medium, and subsequently resuspended at 107 cells/ml in culture medium supplemented with 1 mM sodium pyruvate (Invitrogen). 500 μl of this cell suspension was mixed with 50 μg plasmid and transferred in a 4 mm electroporation cuvette (BioRad, Hercules, Calif., U.S.). The cuvette was placed on ice for 5 min and subsequently the electroporation was performed with Gene Pulser I (BioRad) at 310V and 960 μF capacitance. After the pulse, the cell suspension was placed in a culture flask containing culture medium pre-heated in an incubator with 5% CO2. After 24 h, the culture medium was changed. In the first selection round, 3.108 cells were electroporated, and then only 107 cells for each subsequent round. The selection medium of the stable des transfectants contained culture medium complemented by the addition of geneticin (G418, Gibco) at a final concentration of 1 mg/ml the first week following transfection, and at 2 mg/ml subsequently.
3.3. Retroviral Infection of the scFv Library
Initially, the retroviral supernatants were produced by 293T cells. One day after plating, a transient double transfection was performed with the cationic agent JetPRIME® (Polyplus, Illkirch, France), with the DNA of interest, as well as the DNA coding for the amphotropic envelop gene (vsv-g) and the gag/pol genes. 48 h later, the culture supernatant were collected, filtered, and used for the infection of RBL-2H3 cells. Therefore, the cells were plated in the morning and infected with ⅓ volume of retroviral supernatant, ⅔ volume culture medium and 8 μg/ml polybrene. 48 h post-infection, the culture medium was replaced by fresh medium supplemented with selection agent. The expression of the transgene can be detected from days 4-5 post-infection. The stable transfectants were selected by addition of 1 mg/ml hygromycin B (Invitrogen) to the culture medium.
The cells were incubated one night at 37° C. with anti-DNP (dinitrophenyl) IgE hybridome supernatant at a final IgE concentration of 0.5 μg/ml. On the activation day, the cells were washed once with RPMI and once with Tyrode buffer (10 mM HEPES pH 7.4, 130 mM NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 5.6 mM glucose, et 0.01% BSA). The cells were subsequently activated in the Tyrode buffer with 100 ng/ml DNP-KLH (keyhole limpet hemocyanin conjugated dinitrophenyl, Sigma-Aldrich, St Louis, Mo., USA) at 37° C., in the dark, for 45 minutes. The activation was stopped by placing the cells on ice or by centrifugation at 4° C. Once the cells were stimulated, they were immediately marked with Annexin V-APC (Becton Dickinson Biosciences, San Jose, Calif., USA) as follow: 100 μl Annexin V-APC were added to 2.106 cells (contained in 500 μl), placed 30 min on ice in the dark glace. The cells were then marked with 20 μg/ml Propidium Iodide 3 minutes prior to FACS analysis. The analysis and the sorting by flow cytometry were performed immediately after the marking, by means of a FACS Aria cell sorter (BD) on the Montpellier Rio Imaging platform.
All the oligonucleotides were synthesised and purified by HPLC by MWG (Ebersberg, Allemagne). The plasmid DNA extraction and purification, PCR and ligation kits, were from Macherey Nagel (Duren, Germany). The genomic DNA and RNA extraction were performed with kits made by Qiagen (Germantown, Md., USA).
RBL-2H3 cells were transfected as described above and seeded in limiting dilution on 96 wells culture plates at 1 cell every 5 wells. The transfectants were selected 2 days after transfection by adding the antibiotic corresponding to the resistance provided by the plasmid which was used.
The day before the experiment, 105 cells per well were seeded on 96-well culture plates. After 24 hours, adherent cells were activated by the addition of anti-DNP IgE overnight, and then by 50 to 200 ng/ml DNP-KLH for 45 minutes at 37° C., as previously described. After collecting the supernatant of each well (S1), cells were lysed with lysis buffer (Tyrode buffer, 0.5% Triton, 50 μg/ml aprotinin, 50 μg/ml leupeptin, 50 μg/ml pepstatin, 2 mM PMSF) for 20 minutes on ice. The plate was then centrifuged for 5 minutes at 2000 rpm and the supernatants corresponding to the cell lysates (S2) were harvested. The β-hexosaminidase dosage was carried out on 20 μl of each S1 and S2 supernatants, by addition of 50 μl of the β-hexosaminidase substrate (p-dinitrophenyl-N-acetyl-β-D-glucosaminidase, SIGMA) at a final concentration of 1.3 mg/ml for 1 h 30 at 37° C. The β-hexosaminidase substrate is freshly prepared or stored at −20° C. in a solution of 0.1 M citric acid, pH4.5. The reaction was stopped by addition of 75 μl per well of 0.4 M glycine, pH10.7, and the intensity of staining was evaluated by measuring the optical density at 405 nm. The percentage of β-hexosaminidase release was calculated according to the ratio: S1/(S1+S2)×100.
Annexin V coupled to Fitc (25 μg/ml stock, Assay Designs, Ann Arbor, Mich.) was used for analyses of degranulation. 10 μl of Annexin V-Fitc were added in 106 cells (contained in 1 ml), placed 30 min on ice away from light. Then, in order to distinguish degranulating cell populations from cells undergoing apoptosis, Propidium Iodide at a final concentration of 20 μg/ml was added to cells 3 minutes before analysis by flow cytometry.
To measure the intracellular calcium flux, 106 cells were stimulated in suspension by the addition of anti-DNP IgE for 2-3 hours at 37° C. with stirring (60-70 rpm). After washing in RPMI (centrifugation for 5 minutes at 1000 rpm), cells were labelled with Fluo 3-AM (Invitrogen) at a final concentration of 5 μM for 30 minutes at room temperature and away from light. After marking, washing in RPMI was realized, and then the cells were resuspended at 106 cells/ml, and kept on ice. The marking is stable 1 h. Prior to their activation by addition of DNP and their sorting by flow cytometer, the cells were warmed 10 minutes at 37° C. Two aliquots of cells were used for the analysis of calcium flux in FL1 in function of time. A first tube of cells allowed setting the cytometer to exclude debris and establish the basal fluorescence of the cells. Then a second batch of cells was activated by the addition of DNP-KLH at a final concentration of 200 ng/ml. The intracellular Ca2+ flow increases within seconds, and the analysis is performed over 2-3 minutes. Subsequently, analysis of the mean fluorescence of the population is carried out on Excel.
The day before the experiment, 8.105 cells per well were seeded on 12-well culture plates. The next day, after washing with RPMI and washing with Tyrode, cells are activated as described earlier, with 200 μl per well of 50 ng/ml DNP-KLH for 2 h at 37° C. The supernatants were harvested and used for the quantification of released TNF-α, using an ELISA kit (BD). The cell monolayer was washed once with cold PBS and lysed in PBS supplemented with 0.1% Triton and protease inhibitors (Complete Mini EDTA free tablet, Roche) for 15 minutes at 4° C. The amount of cellular protein was measured using BC Assay kit (Uptima) to normalize the results.
The genomic DNA of 106 cells for the naive library, or 5.106 cells for the following libraries, was extracted and amplified by PCR with primers bordering the VH and VL scFv domains. Then the DNA to be analysed were prepared according to sequencing procedures.
The analysis was performed on the MGX-Montpellier sequencing platform.
The data were analyzed using the SAM software.
Production of scFv Fused to a Mouse Fc
The antibody fragments were previously cloned into the vector ps1119 allowing the expression of scFv fused to a mouse Fc (mufc), of the type IgG1, by insertion at the BglII and EcoRI sites. 293T cells were transiently transfected with JetPEI® (Polyplus), with DNA encoding the Ac fragments, and the vector ps1119. 6 h after transfection, the culture medium was replaced by OPTIMEM (Gibco). Culture supernatants enriched in scFv in the muFc format, were harvested after 6 days, filtered on 0.2□m and stored at −80° C.
Production of Antibody Fragments from Bacteria Cytoplasmic Extracts
The antibody fragments were cloned into the vector pET23NN (modified vector from Novagen, allowing the expression of a Myc tag and a 6×His tag at the C-terminus of the scFv product) between the NcoI and NotI sites. The protocol followed for the expression of scFv in the cytoplasm of E. Coli has been described by the inventor's team (Guglielmi and Pierre Martineau, 2009).
Production of Antibody Fragments from Bacteria Periplasmic Extracts
The antibody fragments were inserted into the vector pHEN2 at sites NcoI and NotI. HB2151 bacteria were thus transformed. After 16 hours of pre-culture at 16° C. in 2×TY containing 100 μg/ml ampicillin and 1% glucose, the growth of bacteria was relaunched at 37° C. At an OD 600 nm of 0.8, induction was triggered by the addition of IPTG at a final concentration of 1 mM for 3-4 h at 30° C. The bacteria were then centrifuged for 20 minutes at 3500 rpm and the pellets were lysed 15 minutes on ice in buffer: 30 mM Tris, pH 8, 20% sucrose, 1 mM EDTA, 1 mM PMSF. After centrifugation, 5 mM MgCl2 and MgSO4 were added to the supernatant periplasm. The extracts were then stored at −20° C.
Purification on Nickel Resin
The antibody fragments were purified on nickel NTA resin from Qiagen. The procedure of the corresponding Ni-NTA spin kit was followed.
Purification on Nickel Magnetic Beads
For each immunoprecipitation the scFv were purified from bacterial cytoplasmic extracts corresponding to 50 ml of culture, with 20 μl of magnetic nickel beads (Ademtech, Pessac, France), according to the kit specification.
The cell layer was washed twice in RPMI. The activated cells were stimulated with 50-100 ng/ml DNP-KLH in RPMI for 3-10 minutes at 37° C. in the dark. After washing with cold PBS containing phosphatase inhibitors (100 mM NaF, 5 mM orthovanadate), cells were lysed for 15 minutes on ice with lysis buffer containing PBS supplemented with: 0.5% sodium deoxycholate, 1% NP-40, 0.1% SDS. After scratching the dishes with a rake, the cell lysates were clarified by centrifugation for 15 minutes at 4° C. at 13000 rpm. An aliquot of the supernatant containing the protein extracts was assayed using a BC Assay kit (Uptima), the remaining lysate was supplemented with loading buffer (2% SDS, 10% glycerol, 2.5% 3-mercaptoethanol, 0.01% bromophenol blue, 30 mM Tris pH6.8).
For immunoprecipitation, 3 mg of protein lysates were used for immunoprecipitation with scFv previously purified from bacteria cytoplasmic extracts on magnetic nickel beads for 2 h at 4° C. Before elution, three 10 minutes washes were performed with the previously described lysis buffer supplemented with 10 mM imidazole. The elution was performed by adding 500 mM imidazole, the eluate was then complemented with loading buffer and loaded on gel for electrophoresis by SDS-PAGE.
The immunoprecipitated proteins were separated by electrophoresis on 10% acrylamide gel. After staining with Coomassie Blue Brilliant Blue, bands of interest were excised, discoloured, reduced, alkylated and digested with trypsin treatment as described by previous work (Shevchenko et al., 2007). The method used for the mass spectrometry analysis was the nano reversed-phase LC-MALDI MS/MS. The data were then analyzed using the software ProteinPilot. The MS/MS analysis was performed on the Proteomics Imaging and Molecular Interaction platform of Montpellier, France.
For these analyses, proteins were transferred on a 0.2 μm nitrocellulose membrane. Before each hybridization, the membrane was blocked in 5% BSA in TBS-T buffer (10 mM Tris pH 7.4, 150 mM NaCl, and 0.1% Tween) for 1 h at room temperature. After hybridization with specific and secondary antibodies coupled to peroxidase (HRP), the proteins were visualized thanks to the ECL-Plus chemiluminescent substrate (PerkinElmer, Waltham, Mass., United States) and using the chemiluminescence camera G-Box (Syngene, Cambridge, UK). Quantification of the intensities of the signals detected was evaluated using the software supplied by the same manufacturer.
The cells were seeded on glass slides in Labtek chambers. All stages of these experiments were performed at room temperature. After two washes in PBS, cells were fixed with 3.7% formaldehyde (Sigma) for 10 minutes, and then permeabilised with PBS containing 0.05% saponin and 0.2% BSA for 10 minutes. The antibodies were then incubated 1-2 h. After washing, the slides were mounted in Mowiol and observed after 3-4 h.
5.6. shRNA
The sequences of the shRNA used (list in
Retroviral supernatants were produced as described previously after double transfection of 293T cells with pSIREN vector containing the sequences of the different shRNA and viral DNA. The infected RBL-2H3 cells were selected by addition of 2.5 g/ml of puromycin (HyClone, ThermoScientific) to the culture medium two days after retroviral infection.
Total RNA was extracted and purified from 105-106 cells using the Qiagen kit. 1 μg of RNA was reverse transcribed with 100 ng of random primer, with the M-MuLV Reverse Transcriptase (Invitrogen). Subsequently the qPCR was performed on the cDNA thus obtained by using the SYBR Green I Master mix, and detected on a Light Cycler 480 (Roche). The data were analysed using the software supplied by the same manufacturer.
SEQ ID NO: 1: Intracellular antibody 3H2-1
SEQ ID NO: 2: Intracellular antibody 3H2-VH
SEQ ID NO: 3: Intracellular antibody 5H4
SEQ ID NO: 4: Intracellular antibody 5H4-VH
SEQ ID NO: 5: Intracellular antibody 5H4-VL
SEQ ID NO: 6: Intracellular antibody 13R4
SEQ ID NO: 7: CDR3 of the VH chain of an intracellular antibody of cluster R—1
SEQ ID NO: 8: CDR3 of the VH chain of an intracellular antibody of cluster R—2
SEQ ID NO: 9: CDR3 of the VH chain of an intracellular antibody of cluster R—3
SEQ ID NO: 10: CDR3 of the VH chain of an intracellular antibody of cluster R—4
SEQ ID NO: 11: CDR3 of the VH chain of an intracellular antibody of cluster R—5
SEQ ID NO: 12: CDR3 of the VH chain of an intracellular antibody of cluster R—6
SEQ ID NO: 13: CDR3 of the VH chain of an intracellular antibody of cluster R—7
SEQ ID NO: 14: CDR3 of the VH chain of an intracellular antibody of cluster R—9
SEQ ID NO: 15: CDR3 of the VH chain of an intracellular antibody of cluster R—10
SEQ ID NO: 16: CDR3 of the VH chain of an intracellular antibody of cluster D—1
SEQ ID NO: 17: CDR3 of the VH chain of an intracellular antibody of cluster D—2
SEQ ID NO: 18: CDR3 of the VH chain of an intracellular antibody of cluster D—3
SEQ ID NO: 19: CDR3 of the VL chain of an intracellular antibody of cluster R—1
SEQ ID NO: 20: CDR3 of the VL chain of an intracellular antibody of cluster R—2
SEQ ID NO: 21: CDR3 of the VL chain of an intracellular antibody of cluster R—3
SEQ ID NO: 22: CDR3 of the VL chain of an intracellular antibody of cluster R—4
SEQ ID NO: 23: CDR3 of the VL chain of an intracellular antibody of cluster R—5
SEQ ID NO: 24: CDR3 of the VL chain of an intracellular antibody of cluster R—6
SEQ ID NO: 25: CDR3 of the VL chain of an intracellular antibody of cluster R—9
SEQ ID NO: 26: CDR3 of the VL chain of an intracellular antibody of cluster R—10
SEQ ID NO: 27: CDR3 of the VL chain of an intracellular antibody of cluster D—1
SEQ ID NO: 28: CDR3 of the VL chain of an intracellular antibody of cluster D—2
SEQ ID NO: 29: CDR3 of the VL chain of an intracellular antibody of cluster D—3
SEQ ID NO: 30: sense strand of sh1 shRNA to LOC297607
SEQ ID NO: 31: sense strand of sh2 shRNA to LOC297607
SEQ ID NO: 33: amino acid sequence of LOC297607 (NP—001100093.1 GI:157820037)
SEQ ID NO: 35: amino acid sequence of C12ORF4 (Accession: NP—065107.1 GI:9966847)
SEQ ID NO: 37: amino acid sequence of LOC57102 (Accession: AFI37796.1 GI:384948382)
SEQ ID NO: 39: amino acid sequence of LOC28040 (NP—613060.1 GI:20070406)
SEQ ID NO: 40: human ABCF1 CDS (Accession: NM—001025091.1 GI:69354670)
SEQ ID NO: 41: amino acid sequence of human ABCF1 (Accession: NP—001020262.1 61:69354671)
SEQ ID NO: 42: CDR3 sequence of the VH domain of intrabody 3H2-1
SEQ ID NO: 43: CDR3 sequence of the VH domain of intrabody 3H2-VH
SEQ ID NO: 44: CDR3 sequence of the VH domain of intrabody 5H4
SEQ ID NO: 45: antisense strand of sh1 shRNA to LOC297607
SEQ ID NO: 46: antisense strand of sh2 shRNA to LOC297607
| Number | Date | Country | Kind |
|---|---|---|---|
| 13305003.9 | Jan 2013 | FR | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2014/050032 | 1/2/2014 | WO | 00 |
