Patent Application
20180036401
Vaccines with an immunologically protective effect against HIV infection.
A major goal of HIV-1 vaccine development is to discover immunogens able to elicit broadly neutralizing antibodies (bNAbs). After more than 25 years of effort, none of the vaccines developed to date have been able to elicit antibodies of this type. For the last decade, the inability to elicit bNAbs has been assumed to result from our lack of success in accurately replicating the trimeric envelope proteins (gp120 and gp41) that compose viral spikes on the virus surface.
Over the last few years, data has emerged suggesting that broadly neutralizing antibodies recognize glycan-dependent epitopes in the HIV-1 envelope protein, gp120 rather than epitopes consisting only of amino acids. More recently, several 3-D structures of broadly neutralizing antibodies binding to HIV-1 envelope proteins have been solved. These have shown that as much as 77% of the binding surface recognized by selected broadly neutralizing monoclonal antibodies (bN-MAbs) is composed of carbohydrate.
Moreover these studies show that the epitope recognized by the prototypic PG9 MAb (16) was located in the V1/V2 domain of gp120 and dependent on mannose-5 for binding. Similarly, another epitope recognized by the prototypic MAb PGT128 (15) is located at the stem of V3 loop and depended on mannose-9 for binding.
Since these bN-MAbs exhibit little if any binding to monomeric gp120, but exhibit robust binding to trimeric envelopes on the surface of cells or viruses, these antibodies were thought to recognize epitopes dependent on both carbohydrate and on the quaternary structure of the envelope trimers (8, 15). Studies of the ontogeny of PG9-like antibodies (3) demonstrated the germline precursor of PG9 and PG9-like antibodies are unable to bind to most monomeric gp120 proteins.
However, one notable exception was gp120 from the A244 strain of HIV-1. Our lab originally determined the sequence of this virus in the early 1990s (7) and has worked with this protein for many years. Indeed this protein was a major component of the AIDSVAX B/E vaccine developed at Genentech and then licensed to VaxGen (1, 2). This vaccine failed to provide protection (11) in a Phase 3 trial (VAX003) carried out in Thailand (1998-2003). However modest but significant protection (31.4% efficacy) was achieved when this vaccine was combined with another vaccine vCP1521 in the RV144 trial involving more than 16,000 volunteers that ended in 2009 (12).
The inventors have discovered that almost any monomeric gp120 will bind to PG9 provided it has the glycosylation sites at positions 156 and 160 that are not glycosylated or are only partially glycosylated, and the right amino acids in the area around amino acids 166-173. The inventors have data from several isolates and novel synthetic constructs showing this to be the case. The present consensus among HIV scientists is that gp120 binds better to trimers because of quaternaty interactions. The inventors have shown this is incorrect—almost any gp120 will bind if its produced in GNT1(−) cells. Our data suggests that PG9 binds better to trimers because the formation of trimers inside the cell shields positions 156 and 160 from interactions with glycoprocessing enzymes resulting in incomplete glycosylation at positions 156 and 160. Most monomeric proteins are not shileded from these enzymes and hence acquire the fully mature complex type of carbohdrate that is not recognized by PG9. Thus the incomplete glycosylation resulting from trimerization accounts for the preferential binding of PG9 to trimers and not the presence of an epitope dependent on quaternary interactions.
The present invention is based upon the reasoning that the best way to improve the immune response to PG9-like epitopes would be to develop small properly glycosylated glycoprotein fragments, or ‘scaffolds’ that possess the carbohydrate structures that the inventors discovered are required for the PG9 binding, and combine these with other Env proteins or other scaffolds to create a multivalent vaccine cocktail.
The invention encompasses small glycosylated fragments of HIV gp120 V1/V2 antigens (scaffolds) that possess the carbohydrate structures required for the PG9 binding. The invention additionally comprises combinations of such scaffolds with other Env proteins or other scaffolds to create a multivalent vaccine cocktail.
The invention encompasses a vaccine comprising a V1/V2 glycoprotein fragment scaffold produced in a cell line, for example A244 strain of HIV-1, or in other embodiments, two or more strains of HIV, where glycosylation is limited to glycosylation of mannose-5, and where the scaffold possesses an epitope in the V1/V2 domain recognized by the PG9 MAb.
The V1/V2 scaffold may be expressed in any strain of HIV-1 produced in a cell line where glycosylation is limited to mannose-5, and wherein the scaffold possesses the epitope in the V1/V2 domain recognized by the PG9 MAb.
The vaccine may comprise V1/V2 scaffolds from two or more strains of HIV-1 produced in a cell line where glycosylation is limited to mannose-5 and wherein the scaffold possesses the epitope in the V1/V2 domain recognized by the PG9 MAb.
In various embodiments the scaffolds may be conjugated to an immunogenic carrier protein.
In another embodiment a vaccine may comprise a mixture of gp120 and the V1/V2 scaffolds where both components are produced under conditions where mannose-5 is incorporated in the V1/V2 domain, and both components are capable of binding to the PG9 MAb.
Additionally, one or both components are derived from the A244 strain of HIV-1.
Where two or more immunogenic components are employed, for example a mixture of gp120 and the V1/V2 scaffolds, the components may be introduced to the subject simultaneously and both components may be contained in the same formulation and immunized at the same time.
Alternatively the components may be formulated separately and immunizations may be given at different times (prime/boost protocol).
Another embodiment comprises an envelope gene from MN-rgp120 mutated to incorporate N-linked glycosylation sites at positions 289, 301, 332 and/or 334 that possesses the epitopes recognized by both the PG9 and PGT128 MAbs when expressed in normal 293 cells. (MN-rgp120 is a recombinant gp120 that was a major component of the AIDSVAX B/E vaccine used in the RV144 trial).
In another embodiment the vaccine formulation is based on recombinant gp120 from the envelope gene from MN-rgp120 mutated to incorporate N-linked glycosylation sites at positions 289, 301, 332 and/or 334 that possesses the epitopes recognized by both the PG9 and PGT128 MAbs when expressed in normal 293 cells.
A related embodiment further comprises a V1/V2 scaffold produced under conditions where mannose-5 is incorporated in the V1/V2 domain, and both components are capable of binding to the PG9 MAb.
In an alternative embodiment the vaccine comprises a mixture of gp120 from the envelope gene from MN-rgp120 mutated to incorporate N-linked glycosylation sites at positions 289, 301, 332 and/or 334 that possesses the epitopes recognized by both the PG9 and PGT128 MAbs when expressed in normal 293 cells and a V1/V2 scaffold from the A244 strain of HIV-1 capable of binding the PG9 MAbs and produced under conditions where mannose-5 is incorporated in the V1/V2 domain and both components are capable of binding to the PG9 MAb and administered concurrently, or alternatively, sequentially, in a prime/boost strategy.
In another embodiment a subject would be immunized with monomeric gp120s possessing both the PG9 and PGT128 epitopes (for example the MN gp120 glycosylation mutant UCSC468) followed by boosting with a V1/V2 fragment that binds PG9 and a V3 fragment that binds PGT128.
Certain embodiments include a composition comprising a monomeric gp120 with glycosylation sites at positions 156 and 160. In various embodiments gp120 is produced in GNT1(−) cells.
Incorporation by reference: U.S. provisional application No. 61/699,680 filed 11 Sep. 2012 is hereby incorporated by reference for all purposes, as are all documents referred to in the provisional application and in the present disclosure.
The terms “complementary” and “complementarity” refer to the natural binding of polynucleotides by base pairing. For example, the sequence “5′ A-G-T 3′” bonds to the complementary sequence “3′ T-C-A 5′.” Complementarity between two single-stranded molecules may be “partial,” such that only some of the nucleic acids bind, or it may be “complete,” such that total complementarity exists between the single stranded molecules. The degree of complementarity between nucleic acid strands has significant effects on the efficiency and strength of the hybridization between the nucleic acid strands.
The term “antisense” refers to any composition containing a nucleic acid sequence which is complementary to the “sense” strand of a specific nucleic acid sequence. Antisense molecules may be produced by any method including synthesis or transcription. Once introduced into a cell, the complementary nucleotides combine with natural sequences produced by the cell to form duplexes and to block either transcription or translation. The designation “negative” or “minus” can refer to the antisense strand, and the designation “positive” or “plus” can refer to the sense strand.
“Conservative amino acid substitutions” are those substitutions that, when made, least interfere with the properties of the original protein, i.e., the structure and especially the function of the protein is conserved and not significantly changed by such substitutions. The table below shows amino acids which may be substituted for an original amino acid in a protein and which are regarded as conservative amino acid substitutions.
The term “derivative” refers to the chemical modification of a polypeptide sequence, or a polynucleotide sequence. Chemical modifications of a polynucleotide sequence can include, for example, replacement of hydrogen by an alkyl, acyl, hydroxyl, or amino group. A derivative polynucleotide encodes a polypeptide which retains at least one biological or immunological function of the natural molecule. A derivative polypeptide is one modified by glycosylation, pegylation, or any similar process that retains at least one biological or immunological function of the polypeptide from which it was derived.
A “fragment” is a unique portion of a parent sequence which is identical in sequence to but shorter in length than the parent sequence. A fragment may comprise up to the entire length of the defined sequence, minus one nucleotide/amino acid residue. For example, a fragment may be at least 5, 10, 15, 20, 25, 30, 40, 50, 60, 75, 100, 150, 250 or at least 500 contiguous nucleotides or amino acid residues in length. Fragments may be preferentially selected from certain regions of a molecule. For example, a polypeptide fragment may comprise a certain length of contiguous amino acids selected from the first 250 or 500 amino acids (or first 25% or 50% of a polypeptide) as shown in a certain defined sequence. Clearly these lengths are exemplary, and any length that is supported by the specification, including the Sequence Listing, tables, and figures, may be encompassed by the present embodiments.
The phrases “percent identity” and “% identity,” as applied to polynucleotide sequences, refer to the percentage of residue matches between at least two polynucleotide sequences aligned using a standardized algorithm. Such an algorithm may insert, in a standardized and reproducible way, gaps in the sequences being compared in order to optimize alignment between two sequences, and therefore achieve a more meaningful comparison of the two sequences. Percent identity between polynucleotide sequences may be determined using the default parameters of the CLUSTAL V algorithm as incorporated into the MEGALIGN version 3.12e sequence alignment program. This program is part of the LASERGENE software package, a suite of molecular biological analysis programs (DNASTAR, Madison Wis.). CLUSTAL V is described in Higgins, D. G. and P. M. Sharp (1989) CABIOS 5:151-153 and in Higgins, D. G. et al. (1992) CABIOS 8:189-191. For pairwise alignments of polynucleotide sequences, the default parameters are set as follows: Ktuple=2, gap penalty=5, window=4, and “diagonals saved”=4. The “weighted” residue weight table is selected as the default. Percent identity is reported by CLUSTAL V as the “percent similarity” between aligned polynucleotide sequence pairs. Alternatively, a suite of commonly used and freely available sequence comparison algorithms is provided by the National Center for Biotechnology Information (NCBI) Basic Local
Alignment Search Tool (BLAST) (Altschul, S. F. et al. (1990) J. Mol. Biol. 215:403-410). The “BLAST 2 Sequences” tool can be used for both blastn and blastp (discussed below). BLAST programs are commonly used with gap and other parameters set to default settings. For example, to compare two nucleotide sequences, one may use blastn with the “BLAST 2 Sequences” tool Version 2.0.9 (May 7, 1999) set at default parameters. Such default parameters may be, for example: Matrix: BLOSUM62; Reward for match: 1; Penalty for mismatch: −2; Open Gap: 5 and Extension Gap: 2 penalties; Gap x drop-off: 50; Expect: 10; Word Size: 11; .Filter: on.
Percent identity may be measured over the length of an entire defined sequence, for example, as defined by a particular SEQ ID number, or may be measured over a shorter length, for example, over the length of a fragment taken from a larger, defined sequence, for instance, a fragment of at least 20, at least 30, at least 40, at least 50, at least 70, at least 100, or at least 200 contiguous nucleotides. Such lengths are exemplary only, and it is understood that any fragment length supported by the sequences shown herein, in the tables, figures, or Sequence Listing, may be used to describe a length over which percentage identity may be measured.
The phrases “percent identity” and “% identity,” as applied to polypeptide sequences, refer to the percentage of residue matches between at least two polypeptide sequences aligned using a standardized algorithm. Methods of polypeptide sequence alignment are well-known. Some alignment methods take into account conservative amino acid substitutions. Such conservative substitutions, explained in more detail above, generally preserve the hydrophobicity and acidity at the site of substitution, thus preserving the structure (and therefore function) of the polypeptide.
Percent identity between polypeptide sequences may be determined using the default parameters of the CLUSTAL V algorithm as incorporated into the MEGALIGN version 3.12e sequence alignment program (described and referenced above). For pairwise alignments of polypeptide sequences using CLUSTAL V, the default parameters are set as follows: Ktuple=1, gap penalty=3, window=5, and “diagonals saved”=5. Percent identity may be measured over the length of an entire defined polypeptide sequence, for example, as defined by a particular SEQ ID number, or may be measured over a shorter length, for example, over the length of a fragment taken from a larger, defined polypeptide sequence, for instance, a fragment of at least 15, at least 20, at least 30, at least 40, at least 50, at least 70 or at least 150 contiguous residues. Such lengths are exemplary only, and it is understood that any fragment length supported by the sequences shown herein, in the tables, figures or Sequence Listing, may be used to describe a length over which percentage identity may be measured.
The phrases “nucleic acid” and “nucleic acid sequence” refer to a nucleotide, oligonucleotide, polynucleotide, or any fragment thereof. These phrases also refer to DNA or RNA of genomic or synthetic origin which may be single stranded or double stranded and may represent the sense or the antisense strand, to peptide nucleic acid (PNA), or to any DNA-like or RNA-like material.
“Operably linked” refers to the situation in which a first nucleic acid sequence is placed in a functional relationship with the second nucleic acid sequence. For instance, a promoter is operably linked to a coding sequence if the promoter affects the transcription or expression of the coding sequence. Generally, operably linked DNA sequences may be in close proximity or contiguous and, where necessary to join two protein coding regions, in the same reading frame.
A “variant” of a particular nucleic acid sequence is defined as a nucleic acid sequence having at least 40% sequence identity to the particular nucleic acid sequence over a certain length of one of the nucleic acid sequences using blastn with the “BLAST 2 Sequences” tool Version 2.0.9 (May 7, 1999) set at default parameters. Such a pair of nucleic acids may show, for example, at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95% or at least 98% or greater sequence identity over a certain defined length. A variant may be described as, for example, an “allelic” (as defined above), “splice,” “species,” or “polymorphic” variant. A splice variant may have significant identity to a reference molecule, but will generally have a greater or lesser number of polynucleotides due to alternate splicing of exons during mRNA processing. The corresponding polypeptide may possess additional functional domains or lack domains that are present in the reference molecule. Species variants are polynucleotide sequences that vary from one species to another. The resulting polypeptides generally will have significant amino acid identity relative to each other. A polymorphic variant is a variation in the polynucleotide sequence of a particular gene between individuals of a given species. Polymorphic variants also may encompass “single nucleotide polymorphisms” (SNPs) in which the polynucleotide sequence varies by one nucleotide base. The presence of SNPs may be indicative of, for example, a certain population, a disease state, or a propensity for a disease state.
A “variant” of a particular polypeptide sequence is defined as a polypeptide sequence having at least 40% sequence identity to the particular polypeptide sequence over a certain length of one of the polypeptide sequences using blastp with the “BLAST 2 Sequences” tool Version 2.0.9 (May 7, 1999) set at default parameters. Such a pair of polypeptides may show, for example, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, or at least 98% or greater sequence identity over a certain defined length of one of the polypeptides.
A major goal in HIV vaccine research is the identification of antigens able to elicit the production of broadly neutralizing antibodies (bNAbs) effective against primary isolates of HIV. The applicant has investigated the molecular features of the HIV-1 envelope glycoproteins, gp160, gp120 and gp41, particularly the epitopes recognized by the prototypic PG9 MAb located in the V1/V2 domain, that confer sensitivity and resistance of viruses to neutralization.
Experimental results. Based on reports that A244-rgp120 was able to bind PG9, the inventors analyzed archival specimens of A244 and MN-rgp120 from the AIDSVAX B/E vaccine to see if they may have failed to provide protection because they lacked the epitopes and carbohydrate required for the binding of the PG9 and PGT128 MAbs. The inventors found weak binding to A244-rgp120 and no binding to MN-rgp120. The inventors also found that both proteins were highly heterogeneous with respect to glycosylation, with the majority of carbohydrate being the complex, sialic acid-containing form that should not bind either MAb (17). To see if the inventors could improve the binding to monomeric gp120, the inventors produced monomeric gp120 from 3 strains of virus (MN, TR011 and A244) in a cell line that lacks the enzyme N-acetylglucosaminyltransferase I (GnTI) that results in mannose-5 glycans being incorporated at all of the predicted N-linked glycosylation sites (PNGS). When grown in normal 293 cells, MN-rgp120 and TRO11-rgp120 were unable to bind to PG9, and A244-rgp120 exhibited weak binding to PG9 (
There has been a considerable amount debate about the heterogeneity and nature of glycosylayion of V1/V2 proteins. Our protein produced in GNTI-cells show 3 major binds and all are due to differences in the occupancy of glycosylation sites. Thus all 3 bands collapse to a single band when deglycosylated by treatment with PNGase which removes all of the N-linked carbohydrate or Endo H that removes all of the high mannose carbohydtare (including mannose 5 glycans). The inventors know from mass spectroscopy studies that the 3 bands are due to differences in glycosylation site occupancy. These studies show that all 3 bands possess the carbohydrate at positions 156 and 160 required for the binding of PG9. These studies, carried out in our lab, used LC-MS/MS. The inventors also carried out disulfide mapping studies of the A244 V1/V2 fragment expressed in GNTI-cells. The inventors found that there was considerable disulfide heterogeneity, which in some preparations may be as high as 50%. The inventors would imagine that this would effect PG9 binding, but have not yet studies this.
Development of scaffolds from the V1/V2 domain that binds glycan-dependent bN-MAbs. While expression of gp120 in GnTI-cells can greatly improve the binding by PG9-like antibodies, this method of production results in major changes in the biophysical and pharmacokinetic profile that could jeopardize the protective immunity achieved in the RV144 trial. Ideally, the inventors would like to add to the existing efficacy of AIDSVAX B/E rather than begin the development of a new vaccine from scratch. See
One way the inventors could accomplish this is to supplement the AIDSVAX vaccine with other Env proteins produced in GnTI-cells. However, the antibody response to gp120 is very complex and only a small percentage of the overall antibody response to any Env protein produced in GnTI-cells would be directed to the PG9 epitope. Moreover, production of Envs in GnTI-cells would destroy other important neutralizing epitopes such as those recognized by PGT128 and 2G12 that depend on mannose-9 glycosylation for binding (10, 15). Therefore the inventors reasoned that the best way to improve the immune response to PG9-like epitopes would be to develop small properly glycosylated fragments, or scaffolds, that possessed the carbohydrate structures required for the PG9 binding, and combine these with other Env proteins or other scaffolds to create a multivalent vaccine cocktail.
To determine if the inventors could identify scaffolds able to bind PG9-like antibodies, the inventors first tested a series of gp120 fragments from MN-rgp120 recently used to map a panel of mouse MAbs to the V2 domain of MN-rgp120 (9). These fragments were expressed as fusion proteins with the signal sequence and amino-terminal flag epitopes of HSV glycoprotein D (gD). When the inventors tested the fragments produced in normal 293 cells, PG9 was unable to bind to any of the fragments. However when the fragments were produced in GnTI-cells, the inventors found three fragments (7-15, 4-27, and 1-3) that bound PG9 (
Encouraged by the success of these studies, the inventors attempted to produce V1/V2 scaffolds from other proteins as well. The next scaffold the inventors tried was one made from the CF01_AE TH023 Env that was contained in the vCP1521 pox virus vector used in the RV144 trial. Interestingly, the inventors found that this scaffold exhibited good binding to the PG9 MAb when expressed in GnTI-cells, but little or no binding to the CH01, CH03, or PGT145 MAbs (data not shown). This result confirmed reports suggesting that the PG9, CH01 and CH03 MAbs bind to non-identical overlapping epitopes (8). Our present work involves identifying scaffolds from other clades that the inventors can use for the sequential immunization studies described below. The inventors first screen envelope proteins produced in normal 293 cells for binding to PG9 as observed with A244-rgp120, and then use the best of these to make V1/V2 scaffolds that the inventors express in GnTI-293 cells. Results from a study where the inventors screened Env proteins
from a panel of clade C viruses produced in our lab (13) are shown in
Characterization of the structure of V1/V2 scaffolds. The inventors next wanted to characterize the structure of the A244-V1/V2 scaffold. The inventors first examined the size of the scaffolds produced in normal and GnTI-293 cells by polyacrylamide gel electrophoresis (PAGE). The results of these studies are shown in
Because the V1/V2 domain is a 4-stranded anti-parallel β-sheet, this fragment should have a distinctive absorbance profile when measured by circular dichroism (CD). The results of these studies are shown in
The mechanism responsible for the incorporation of mannose-5 glycans at the PG9 epitope and mannose-9 at the PGT128 epitopes. The discovery of glycan-dependent epitopes in gp120 was intriguing, and the inventors wanted to investigate the mechanism by which HIV-1 is able to selectively incorporate intermediates in the N-linked glycosylation pathway (e.g. mannose-5 and mannose-9) at specific epitopes recognized by the PG9 and PGT128/2G12 MAbs. Several groups have suggested that glycosylation sites are occluded from glycan processing enzymes during the process of gp160 timer formation, leading to incomplete glycosylation at selected sites (5, 6). However, this could not explain the fact that monomeric A244-rgp120 appeared to possess the glycans required for PG9 binding when expressed in normal 293 cells, while gp120 from the MN strain was unable to bind PG9. To account for this phenomenon, the inventors compared the sequences and structural features of MN-rgp120 with A244-rgp120. The inventors noticed that the structure on MN-rgp120 differed from A244 gp120 in the number of glycosylation sites in the V3 stem region where MN-rgp120 possessed 2 PNGS and A244-rgp120 possessed 4 PNGS (
To determine whether the presence of glycosylation sites in the V3 stem might affect the type of glycosylation at the PG9 epitope in the V1/V2 domain, the inventors carried out site-directed mutagenesis to incorporate PNGS (predicted N-linked glycosylation sites) present in the V3 stem of A244-rgp120 into the structure of MN-rgp120 that lacked these sites (
Previous studies have reported that incomplete glycosylation can result from steric hindrance either by the polypeptide chain or by dense clusters of N-glycans (14). It is likely that both types of interactions result in the incorporation of the glycosylation pathway intermediates required for the binding of the PG9 and PGT128 MAbs. These results suggest that the glycan shield, thought to have evolved in HIV-1 to prevent the binding of neutralizing antibodies, also prevents access by carbohydrate processing enzymes.
Thus PNGS (predicted N-linked glycosylation sites) sometimes interfere with each other, resulting in the incorporation of intermediate structures (e.g. mannose-5 and mannose-9). The development of the monomeric MN-rgp120 glycosylation mutants described above, able to bind two of the most potently neutralizing MAbs described to date (PG9 and PGT128), represents a major advance that should be particularly useful in the development of an improved gp120 subunit vaccine.
Further recent data show that fragments of gp120 from the 108060 isolate of HIV-1 are able to bind the PGT128 MAb that recognizes a mannose 9 dependent glycan epitope in the stem of the V2 domain when the fragments are grown in the presence of kefunensine. PGT128 is the most potent neutralizing monoclonal antibody ever discovered. Kefunensine is a drug that limits N linked glycosylatin to the mannose 9 form.
In one embodiment a subject would be immunized with monomeric gp120s possessing both the PG9 and PGT128 epitopes as with the MN gp120 glycosylation mutant (UCSC468) that the inventors described followed by boosting with a V1/V2 fragment that binds PG9 and a V3 fragment that binds PGT128.
The V1/V2 fragment can be A244-V1/V2 produced in GNTI-cells and the V3 fragment can be one of several produced from 108060 envelope protein grown in cells treated with kefunensine. This data shows that use of multiple fragments provides neutralizing monoclonal antibodies. See
Additional data shows that V1/V2 scaffolds can be made from viruses of other strains and other clades when grown in GNTI-cells. However the binding of PG9 to these appears weaker than to the A244-V1/V2 fragment. Therefore the inventors have made scaffolds from a total of 7 different envelope proteins A244, TH023, BAL, CAP45, ZM109, ZM197, and ZM53. See
Sequences of Scaffold Proteins.
Further Synthetic Engineered Scaffold Proteins
The inventors further engineered synthetic scaffold proteins. These V1/V2 domain scaffolds were all expressed as fusion proteins with the signal sequence and 27 amino acid flag epitope of herpes simplex virus glycoprotein D. In addition to the flag epitope the engineered proteins posess a 3 amino acid linker (LLE).
In most cases the V1/V2 sequence begins with VPL. Experimental data show that these fragments are all able to bind PG9 when grown in GnTI-cells, but bind PG9 poorly when grown in normal 293 cells. However, the ZM233 fragment appears to be unique in that there is good binding to PG9 when this scaffold is produced in normal 293 cells. This characteristic will provide a big advantage in manufacturing since the protein can be produced in some normal cell lines acceptable for the production of human pharmaceuticals.
These synthetic fragments have been successfully expressed using signal sequences from human tissue plasminogen activator and ICAM-1.
Methods for Creating Synthetic Engineered Scaffold Proteins
gp160
Plasmid DNA is available through the AIDS Reagent program. ZM233 (Catalog #1131; Accession #DQ388517); ZM109 (Catalog #11314; AY424138); CAP45 (Catalog #11316; Accession #DQ435682); Ba1.01 (Catalog #11445; Accession #DQ318210); ZM53 (Catalog #11313; Accession #AY423984); ZM197 (Catalog #11309; Accession #DQ388515)
Construction of V1/V2 Fragments:
The V1/V2 fragments were PCR amplified from gp160's using primers containing Kpn1 and Not1 restriction sites. The fragments were digested with the corresponding restriction enzymes and ligated into an expression vector (pRK) containing the HSV signal sequence, HSV gD “tag” (amino-terminal flag epitopes of HSV glycoprotein D), and a short peptide linker, LLEVPL, that is used to separate the “tag” from the start of the V1/V2 fragment. The fragments start at amino acid 117 (HXB2 numbering) and end at amino acid 207; a stop codon was placed at position 208. All fragments contain 3 disulfide bonds.
Expression
The V1/V2 fragments were expressed in 293F™ cells (Invitrogen: #R790-07) or 293-GnT1-cells (ATCC: #CRL-3022). Tranfection was done using Polyethylenimine (PEI). Briefly, 1×108 293F or GnT1-cells were centrifuged for 10 min. at 1200 rpm. The media was removed and the cells were resuspended in 5 ml of media containing 0.1% pluronic acid; 250 ug of plasmid DNA and 800 ug of PEI was added. After 3 hr incubation at 37° C., the cells were brought up to a final volume of 100 ml. 3-days post-transfection, the cell supernatant was collected, centrifuged, and filtered through a 0.45 um PES membrane.
ELISA data
Indirect ELISA was used to measure PG9 binding to the V1/V2 fragments. Briefly, Maxisorp ELISA plates (Nunc) were coated with 2 ug/ml of anti-gD antibody (34.1) in PBS overnight at 4° C. The following day the plates were washed 4× with PBS+0.05% Tween-20. The plates were blocked using PBS+1% BSA for 2 hrs at room temperature. Cell supernatant (100 ul) was added for 1 hr. PG9 was added at 10 ug/ml, 1 ug/ml, 0.5 ug/ml, 0.1 ug/ml, 0.05ug/ml, 0.01 ug/ml, and 0.001 ug/ml. A blank well was used to measure background absorbance. A secondary antibody, anti-human Fc Peroxidase-conjugated, was used to detect PG9. Substrate (OPD) was added for 10 min. and stopped using 3M H2SO4. All dilutions, (except coating) were done in PBS+1% BSA. Wash steps were done after every incubation.
CONCLUSION. An important conclusion from this work is that the inventors have discovered that almost any monomeric gp120 will bind to PG9 provided it has the glycosylation sites at positions 156 and 160 and the right amino acids in the area around amino acids 166-173. The inventors have data from several other isolates and novel synthetic constructs showing this to be the case. The present consensus among HIV scientists is that gp120 binds better to trimers because of quaternaty interactions. The inventors have shown this is incorrect—almost any gp120 will bind if its produced in GNT1(−) cells.
Our data suggests that PG9 binds better to trimers because the formation of trimers inside the cell shields positions 156 and 160 from interactions with glycoprocessing enzymes resulting in incomplete glycosylation at these positions. Most monomeric proteins are not shileded from these enzymes and hence acquire the fully mature complex type of carbohdrate that is not recognized by PG9. Thus the incomplete glycosylation resulting from trimerization accounts for the preferential binding of PG9 to trimers and not the presence of an epitope dependent on quaternary interactions.
This international application claims the benefit of and priority to U.S. provisional application No. 61/699,680 filed 11 Sep. 2012, titled “HIV-1 envelope proteins and fragments that possess glycan dependent epitopes recognized by broadly neutralizing antibodies” which is hereby incorporated by reference for all purposes.
| Number | Date | Country | |
|---|---|---|---|
| 61699680 | Sep 2012 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 14427393 | Mar 2015 | US |
| Child | 15635064 | US |
