Prediction of tumor-reactive T cell receptors from scRNA-seq data for personalized T cell therapy.

The identification of patient-derived, tumor-reactive T cell receptors (TCRs) as a basis for personalized transgenic T cell therapies remains a time- and cost-intensive endeavor. Current approaches to identify tumor-reactive TCRs analyze tumor mutations to predict T cell activating (neo)antigens and use these to either enrich tumor infiltrating lymphocyte (TIL) cultures or validate individual TCRs for transgenic autologous therapies. Here we combined high-throughput TCR cloning and reactivity validation to train predicTCR, a machine learning classifier that identifies individual tumor-reactive TILs in an antigen-agnostic manner based on single-TIL RNA sequencing. PredicTCR identifies tumor-reactive TCRs in TILs from diverse cancers better than previous gene set enrichment-based approaches, increasing specificity and sensitivity (geometric mean) from 0.38 to 0.74. By predicting tumor-reactive TCRs in a matter of days, TCR clonotypes can be prioritized to accelerate the manufacture of personalized T cell therapies.

Non-viral S/MAR vectors replicate episomally in vivo when provided with a selective advantage.

The ideal gene therapy vector should enable persistent expression without the limitations of safety and reproducibility. We previously reported that a prototype plasmid vector, containing a scaffold matrix attachment region (S/MAR) domain and the luciferase reporter gene, showed transgene expression for at least 6 months following a single administration to MF1 mice. Following partial hepatectomy of the animals, however, we found no detectable vector replication and subsequent propagation in vivo. To overcome this drawback, we have now developed an in vivo liver selection strategy by which liver cells transfected with an S/MAR plasmid are provided with a survival advantage over non-transfected cells. This allows an enrichment of vectors that are capable of replicating and establishing themselves as extra-chromosomal entities in the liver. Accordingly, a novel S/MAR plasmid encoding the Bcl-2 gene was constructed; Bcl-2 expression confers resistance against apoptosis-mediated challenges by the Fas-activating antibody Jo2. Following hydrodynamic delivery to the livers of mice and frequent Jo2 administrations, we demonstrate that this Bcl-luciferase S/MAR plasmid is indeed capable of providing sustained luciferase reporter gene expression for over 3 months and that this plasmid replicates as an episomal entity in vivo. These results provide proof-of-principle that S/MAR vectors are capable of preventing transgene silencing, are resistant to integration and are able to confer mitotic stability in vivo when provided with a selective advantage.

Persistent episomal transgene expression in liver following delivery of a scaffold/matrix attachment region containing non-viral vector.

An ideal gene therapy vector should enable persistent transgene expression without limitations of safety and reproducibility. Here we report the development of a non-viral episomal plasmid DNA (pDNA) vector that appears to fulfil these criteria. This pDNA vector combines a scaffold/matrix attachment region (S/MAR) with a human liver-specific promoter (alpha1-antitrypsin (AAT)) in such a way that long-term expression is enabled in murine liver following hydrodynamic injection. Long-term expression is demonstrated by monitoring the longitudinal luciferase expression profile for up to 6 months by means of in situ bioluminescent imaging. All relevant control pDNA constructs expressing luciferase are unable to sustain significant transgene expression beyond 1 week post-administration. We establish that this shutdown of expression is due to promoter methylation. In contrast, the S/MAR element appears to inhibit methylation of the AAT promoter thereby preventing transgene silencing. Although this vector appears to be maintained as an episome throughout, we have no evidence for its establishment as a replicating entity. We conclude that the combination of a mammalian, tissue-specific promoter with the S/MAR element is sufficient to drive long-term episomal pDNA expression of genes in vivo.

Characterisation of LMD virus-like nanoparticles self-assembled from cationic liposomes, adenovirus core peptide mu and plasmid DNA.

Liposome:mu:DNA (LMD) is a ternary nucleic acid delivery system built around the mu peptide associated with the condensed core complex of the adenovirus. LMD is prepared by precondensing plasmid DNA (D) with mu peptide (M) in a 1:0.6 (w/w) ratio and then combining these mu:DNA (MD) complexes with extruded cationic liposomes (L) resulting in a final lipid:mu:DNA ratio of 12:0.6:1 (w/w/w). Correct buffer conditions, reagent concentrations and rates of mixing are all crucial to success. However, once optimal conditions are established, homogeneous LMD particles (120 +/- 30 nm) will result that each appear to comprise an MD particle encapsulated within a cationic bilammellar liposome. LMD particles can be formulated reproducibly, they are amenable to long-term storage (>1 month) at -80 degrees C and are stable to aggregation at a plasmid DNA concentration up to 5 mg/ml (15 mM nucleotide concentration). Furthermore, LMD transfections are significantly more time and dose efficient in vitro than cationic liposome-plasmid DNA (LD) transfections. Transfection times as short as 10 min and plasmid DNA doses as low as 0.001 microg/well result in significant gene expression. LMD transfections will also take place in the presence of biological fluids (eg up to 100% serum) giving 15-25% the level of gene expression observed in the absence of serum. Results from confocal microscopy experiments using fluorescent-labelled LMD particles suggest that endocytosis is not a significant barrier to LMD transfection, although the nuclear membrane still is. We also confirm that topical lung transfection in vivo by LMD is at least equal in absolute terms with transfection mediated by GL-67:DOPE:DMPE-PEG(5000) (1:2:0.05 m/m/m), an accepted 'gold-standard' non-viral vector system for topical lung transfection, and is in fact at least six-fold more dose efficient. All these features make LMD an important new non-viral vector platform system from which to derive tailor-made non-viral delivery systems by a process of systematic modular upgrading.

A novel peptide, THALWHT, for the targeting of human airway epithelia.

Targeting gene vectors to human airway epithelial cells may help to overcome the current inefficiency of gene transfer as the major problem confronting cystic fibrosis gene therapy. To elucidate novel ligands targeting abundant, apically located receptors on airway epithelial cells, a phage display library was screened for peptides binding with high affinity to such cells. This screening yielded a selectively enriched amino acid sequence, Thr-His-Ala-Leu-Trp-His-Thr (THALWHT). Subsequent binding studies confirmed that THALWHT-displaying phages bound much stronger than phages displaying control peptides to human airway epithelial cells. In contrast, no significant binding differences were observed on a variety of non-airway-derived human cell lines suggesting selective binding of the THALWHT motif to airway epithelia. Confocal microscopy of such cells after exposure to labelled synthetic THALWHT peptide indicated that its binding is followed by specific internalisation via endocytosis. A synthetic peptide comprising a cyclic CTHALWHTC domain and a DNA binding moiety enabled efficient targeted gene delivery into human airway epithelial cells. Competition assays with free THALWHT peptide confirmed the specificity of gene delivery. Thus, the THALWHT motif may prove a useful targeting moiety for both non-viral and viral gene therapy vectors.

Cell delivery, intracellular trafficking and expression of an integrin-mediated gene transfer vector in tracheal epithelial cells.

The mechanism of cell entry and intracellular fate of a gene transfer vector composed of a receptor-targeting, DNA-condensing peptide, RGD-oligolysine, a luciferase encoding plasmid DNA (pDNA) and a cationic liposome was examined. We demonstrate by confocal microscopy, electron microscopy and subcellular fractionation that the major mechanism of entry of the vector is endocytic. The vector complex rapidly (5 min) internalizes into early endosomes, then late endosomes and lysosomes. Entry involves, at least in part, clathrin-coated pit-mediated endocytosis since different conditions or drugs known to influence this pathway modify both uptake of pDNA and its expression. The observed increase in expression with addition of a lip some correlated with an increase in the rate of transfer of the pDNA to lysosomes, a decrease in intracellular recycling and exocytosis of the pDNA and an increase in the amount of pDNA in the nuclear fraction. Trafficking within the cell involved endosome fusion and the acid environment of the endosomes-lysosomes was beneficial for expression. After 30 min both the peptide and pDNA localized to the nucleus and the amount of intact pDNA in the nuclear fraction was highest with liposome and peptide. A better understanding of the cellular mechanisms by which vectors transfer to and traffic in cells should help design improved vectors.

Targeted gene delivery into alpha9beta1-integrin-displaying cells by a synthetic peptide.

We have investigated the usefulness of two small synthetic peptides comprising either a linear or a cyclic PLAEIDGIEL domain and a DNA-binding moiety of 16 lysine residues to mediate gene transfer selectively into alpha9beta1-integrin-displaying cells. Such specific gene delivery could only be achieved with the peptide containing the cyclic PLAEIDGIEL domain. However, inclusion of the cationic liposome Lipofect-AMINE into the peptide/DNA complexes resulted for both peptides in efficient gene transfer with significant targeting specificity. Naturally, the integrin alpha9beta1 is present only in a few highly specialised tissues and abundant throughout the human airway epithelia in vivo. Targeting gene vectors to this integrin therefore appears a useful approach to gene therapy of lung diseases such as cystic fibrosis. As the integrin alpha9beta1 is associated with tissue differentiation during foetal development and may cause resurgence of the foetal phenotype in colon cancers, such vectors may also be applicable for prenatal and cancer gene therapy.

A novel peptide, PLAEIDGIELTY, for the targeting of alpha9beta1-integrins.

Targeting gene therapy vectors to abundant receptors on airway epithelia may allow a significant enhancement of gene delivery and thereby be of particular importance for the gene therapy of cystic fibrosis. Alpha9beta1-integrins are highly expressed throughout the human airway epithelia in vivo, irrespective of any particular clinical status. Aiming to improve the targeting of our non-viral integrin-mediated gene transfer systems to airway epithelia, we searched for a short tenascin C-derived peptide which would bind to these integrins. By utilizing recombinant bacteriophages that display overlapping regions of the third fibronectin type III repeat of tenascin C (TNfn3), we were able to localize its alpha9beta1-integrin binding site to the B-C loop of TNfn3. A synthetic Pro-Leu-Ala-Glu-Ile-Asp-Gly-Ile-Glu-Leu-Thr-Tyr peptide (PLAEIDGIELTY) was shown to displace alpha9beta1-integrin-expressing cells completely from binding to TNfn3. This peptide, therefore, may prove useful both for the examination of the functional importance of alpha9beta1-integrins in vivo and the development of gene therapy vectors or drugs targeting these integrins.

An RGD-oligolysine peptide: a prototype construct for integrin-mediated gene delivery.

We have synthesized a linear, bifunctional peptide that comprises an integrin-targeting domain containing an arginine-glycine-aspartic acid tripeptide motif and a DNA-binding moiety consisting of a short stretch of 16 lysine residues. This peptide can form distinctive, condensed complexes with DNA and is capable of mediating its delivery and expression in a variety of mammalian cells in culture. Internalization is mediated by cell surface integrin receptors via a mechanism that is known to be phagocytic. We have analyzed the relationship between DNA and peptide and have investigated the conditions suitable for optimal gene delivery. The formation of condensed peptide DNA complexes leads to resistance to nuclease degradation. The level of reporter gene expression obtained is dependent on the peptide-to-DNA ratio and is enhanced in the presence of the endosomal buffer chloroquine, polyethyleneimine, and deactivated adenovirus during gene delivery. Under optimal conditions the levels of reporter gene expression obtained approach or even exceed those obtained with DNA delivered with the commercial liposome Lipofectamine. The ability to produce an efficient gene delivery system using small, easily modified, and well-defined constructs that have no constraint of particle size demonstrates the advantages of integrin-targeting peptides for gene transfer.

Lipid-mediated enhancement of transfection by a nonviral integrin-targeting vector.

Nonviral vectors consisting of integrin-targeting peptide/DNA (ID) complexes have the potential for widespread application in gene therapy. The transfection efficiency of this vector, however, has been limited by endosomal degradation. We now report that lipofectin (L) incorporated into the ID complexes enhances integrin-mediated transfection, increasing luciferase expression by more than 100-fold. The transfection efficiency of Lipofectin/Integrin-binding peptide/DNA (LID) complexes, assessed by beta-galactosidase reporter gene expression and X-gal staining, was improved from 1% to 10% to over 50% for three different cell lines, and from 0% to approximately 25% in corneal endothelium in vitro. Transfection complexes have been optimized with respect to their transfection efficiency and we have investigated their structure, function, and mode of transfection. Both ID and LID complexes formed particles, unlike the fibrous network formed by lipofectin/DNA complexes (LD). Integrin-mediated transfection by LID complexes was demonstrated by the substantially lower transfection efficiency of LKD complexes in which the integrin-biding peptide was substituted for K16 (K). Furthermore, the transfection efficiency of complexes was shown to be dependent on the amount of integrin-targeting ligand in the complex. Finally, a 34% reduction in integrin-mediated transfection efficiency by LID complexes was achieved with a competing monoclonal antibody. The role of lipofectin in LID complexes appears, therefore, to be that of a co-factor, enhancing the efficiency of integrin-mediated transfection. The mechanism of enhancement is likely to involve a reduction in the extent of endosomal degradation of DNA.

Liposomes enhance delivery and expression of an RGD-oligolysine gene transfer vector in human tracheal cells.

Nonviral gene delivery systems consist predominantly of lipoplexes or receptor-targeting and nontargeting polyplexes. We examined integrin-mediated gene delivery using an Arg-Gly-Asp/oligo-L-lysine ([K]16RGD) cyclic peptide and investigated its gene transfer efficiency when associated with a cationic liposome. We demonstrated that human cystic fibrosis and noncystic fibrosis tracheal epithelial cells in culture express integrins that recognise the RGD integrin-binding motif. We found a 10-fold (P < 0.01) increased expression of a luciferase encoding plasmid in these cells when complexing the plasmid to the [K]16RGD peptide as compared with plasmid alone. This increase was specific to the [K]16RGD peptide since neither a [K]16RGE nor a [K]16 peptide gave a comparable increase. Expression was further enhanced 30-fold (P < 0.01) with lipofectamine and the ratio of DNA/peptide/lipofectamine was critical for specificity and expression. Fluorescence and radioactive labelling of the complex showed that the [K]16RGD peptide increased the endocytic uptake of DNA into cells. The cell association of both DNA and peptide increased even further with lipofectamine. Confocal microscopy showed that the [K]16RGD peptide and the DNA internalised together within 30 min and localised to vesicles in the perinuclear region. These results show that an integrin-binding ligand can deliver genetic material to airway cells and that a cationic liposome can enhance the efficacy of this nonviral vector system.

A cell adhesion peptide from foot-and-mouth disease virus can direct cell targeted delivery of a functional enzyme.

The G-H loop of foot-and-mouth disease virus is a disordered protrusion of the VP1 protein exposed on the virion surface. This short stretch includes an arginine-glycine-aspartic acid tripeptide, a recognized integrin-binding motif, which is responsible for cell attachment and infection. Eight copies of a peptide reproducing the amino acid sequence of this FMDV ligand have been displayed in solvent-exposed regions on an enzymatically active recombinant beta-galactosidase. This viral peptide segment enables the chimeric enzyme to bind mammalian cell lines with different efficiencies, probably depending on the number of suitable cell receptors present on each of them. Moreover, it also promotes the internalization of the attached enzyme, which is transiently active inside the cells. These results suggest further exploration of the potential use of short adhesion peptides of viral origin as cell attachment tags to direct the targeted delivery of both genes and enzymes, instead of whole, infectious viruses.

Recombinant HMG1 protein produced in Pichia pastoris: a nonviral gene delivery agent.

This paper describes the production of a recombinant protein from the expression system based on the methylotrophic yeast Pichia pastoris. Efficient production of rat high-mobility-group 1 (HMG1) protein was obtained using the system. Two forms of HMG1 were secreted into the culture medium: a 24.5-kDa species corresponding to the native HMG1 and a 32-kDa glycosylated derivative. Non-glycosylated recombinant HMG1 was purified easily and shown to possess the same DNA-binding properties as HMG1 purified from calf thymus. Plasmid DNA complexed to the recombinant HMG1 is taken up by a variety of mammalian cells in culture. Transient expression of a luciferase reporter gene was observed. Under selective conditions, stable expression of a neomycin gene was established as a result of integration into the genome. HMG1-mediated gene delivery was as efficient as calcium phosphate-mediated transfection but without associated cell damage. In addition, stable transfectants obtained after selection for G418 resistance usually integrated only one copy of the transfected DNA in contrast to the high unpredictable number obtained by the calcium phosphate method. HMG1 transfection complexes were not toxic to cultured cells, even at high concentrations.

A recombinant, arginine-glycine-aspartic acid (RGD) motif from foot-and-mouth disease virus binds mammalian cells through vitronectin and, to a lower extent, fibronectin receptors.

The cell-binding abilities of a recombinant, RGD-containing peptide from foot-and-mouth disease virus (FMDV) have been characterized in HeLa and BHK cells. This peptide represents the aa sequence of the solvent-exposed G-H loop of protein VP1 which is involved in cell recognition and infection. The efficiency of the viral motif in promoting cell attachment and spreading is comparable to that shown by fibronectin or vitronectin. Cell binding is inhibited by a monoclonal antibody directed against a viral, RGD-involving B-cell epitope and also by sera against vitronectin (alpha V beta 3/beta 5) and fibronectin (alpha 5 beta 1) receptors. In addition, a synthetic RGD peptide, which is a ligand for both integrins, prevents the cell binding mediated by the FMDV domain. These data demonstrate that the FMDV RGD motif is a potent ligand for cell-receptor integrins and sufficient to promote cell attachment to susceptible cells mainly through the vitronectin receptor.

Gene delivery and expression mediated by an integrin-binding peptide.

The ability to transfer sufficient DNA to specific target cells remains one of the main limitations to the development of gene therapy. For this reason much attention is being paid to the development of new gene delivery systems, both viral and non-viral. We describe gene transfer with a polycation-DNA complex which contains an integrin-binding domain. Integrin-mediated gene delivery has several potential advantages. Such complexes are less likely than other receptor-mediated gene delivery complexes to be constrained by the size of the complex. The ligands are small peptides, resembling naturally occurring integrin ligands, which minimises the possibility of complexes inducing an immune response in vivo.

Cell binding and internalization by filamentous phage displaying a cyclic Arg-Gly-Asp-containing peptide.

Ligands that bind mammalian cell surface integrins with high affinity can mediate cellular internalization. We show that particles of the bacteriophage fd that display the cyclic integrin-binding peptide sequence GGCRGDMFGC in a proportion of their major coat protein subunits bind to cells and are efficiently internalized. In the displayed peptide the conformation of the RGD motif is restricted within a hairpin loop formed by a disulfide bridge between the 2 cysteine residues. Cellular internalization of phage was demonstrated by confocal and non-confocal immunofluorescence microscopy of tissue-cultured cells incubated with phage particles. The phage were contained in juxtanuclear vesicles in the same serial sections as transferrin receptor but were not colocalized with the cell surface marker alkaline phosphatase. Cell binding and internalization was inhibited by preincubation of cells with the integrin-binding peptide GRGDSP, whereas the control peptide GRGESP had no inhibitory effect. These results indicate that cyclic integrin-binding peptides can be used to target and enter cells and that it should be possible to exploit such peptides for the introduction of DNA, drugs, or other macromolecules.