Inhibition of tumor growth in vivo by in situ secretion of bispecific anti-CEA x anti-CD3 diabodies from lentivirally transduced human lymphocytes.

Infiltrating T lymphocytes are found in many malignancies, but they appear to be mostly anergic and do not attack the tumor, presumably because of defective T-cell activation events. Recently, we described a strategy for the tumor-specific polyclonal activation of tumor-resident T lymphocytes based on the in situ production of recombinant bispecific antibodies (bsAbs) by transfected nonhematological cell lines. Here, we have constructed a novel HIV-1-based lentiviral vector for efficient gene transduction into various human hematopoietic cell types. Several myelomonocytic and lymphocytic cell lines secreted the anti-carcinoembryonic antigen (CEA) x anti-CD3 diabody in a functionally active form with CD3(+) T-cell lines being the most efficient secretors. Furthermore, primary human peripheral blood lymphocytes (PBLs) were also efficiently transduced and secreted high levels of functional diabody. Importantly gene-modified PBLs significantly reduced in vivo tumor growth rates in xenograft studies. These results demonstrate, for the first time, the utility of lentiviral vectors for sustained expression of recombinant bsAbs in human T lymphocytes. Such T lymphocytes, transduced ex vivo to secrete the activating diabody in autocrine fashion, may provide a promising route for a gene therapy strategy for solid human tumors.

Directed evolution of polymerase function by compartmentalized self-replication.

We describe compartmentalized self-replication (CSR), a strategy for the directed evolution of enzymes, especially polymerases. CSR is based on a simple feedback loop consisting of a polymerase that replicates only its own encoding gene. Compartmentalization serves to isolate individual self-replication reactions from each other. In such a system, adaptive gains directly (and proportionally) translate into genetic amplification of the encoding gene. CSR has applications in the evolution of polymerases with novel and useful properties. By using three cycles of CSR, we obtained variants of Taq DNA polymerase with 11-fold higher thermostability than the wild-type enzyme or with a >130-fold increased resistance to the potent inhibitor heparin. Insertion of an extra stage into the CSR cycle before the polymerase reaction allows its application to enzymes other than polymerases. We show that nucleoside diphosphate kinase and Taq polymerase can form such a cooperative CSR cycle based on reciprocal catalysis, whereby nucleoside diphosphate kinase produces the substrates required for the replication of its own gene. We also find that in CSR the polymerase genes themselves evolve toward more efficient replication. Thus, polymerase genes and their encoded polypeptides cooperate to maximize postselection copy number. CSR should prove useful for the directed evolution of enzymes, particularly DNA or RNA polymerases, as well as for the design and study of in vitro self-replicating systems mimicking prebiotic evolution and viral replication.

CD3X anti-nitrophenyl bispecific diabodies: universal immunotherapeutic tools for retargeting T cells to tumors.

We developed a universal recombinant bispecific molecule (BiMol) that is capable of redirecting cytotoxic T cells to tumor cells via tagged anti-tumor ligands such as antibody fragments or cytokines. A recombinant bispecific diabody with binding specificities for the CD3 molecule on T cells as well as for the hapten nitrophenyl (NIP) was produced. This bispecific molecule is capable of redirecting cytotoxic T cells to kill a series of malignant cells, including B cell lymphoma, Hodgkin's lymphoma, and colon carcinoma via NIP-conjugated ligands to tumor-associated antigens. Cytotoxic activity of the diabody was found to be comparable to tetradoma-derived bispecific antibodies with similar specificities. Our findings demonstrate that universal CD3xanti-NIP diabodies could be used for T cell based cellular immunotherapy in a variety of human malignancies. Additionally, these bispecific molecules allow fast and economic testing of tumor-associated antigens on malignant cells for their potential use as immunotherapeutic target structures if corresponding hapten-conjugated antibodies or ligands are available.

Carcinoembryonic antigen (CEA)-specific T-cell activation in colon carcinoma induced by anti-CD3 x anti-CEA bispecific diabodies and B7 x anti-CEA bispecific fusion proteins.

Two bispecific recombinant molecules, an anti-CD3 x anti-carcinoembryogenic antigen (CEA) diabody and a B7 x anti-CEA fusion protein, were tested for their capacity to specifically activate T cells in the presence of CEA-expressing colon carcinoma cells. T-cell activation by the anti-CD3 x anti-CEA diabody required close contact to CEA-positive cells and resulted in diabody-mediated cytotoxicity against the target cells. Additionally, CD28-mediated costimulation in combination with anti-CD3 x anti-CEA diabodies induced activation of autologous T cells in CEA-positive primary colon carcinoma specimens, as determined by flow cytometry. The high specificity of the bispecific diabody approach could be further enhanced by the use of B7 x anti-CEA fusion proteins because the costimulatory CD28-signaling to the T cells strictly depended on the expression of CEA on the target cells. We demonstrate that displaying engagement sites for the T-cell antigens CD3 and CD28 on the surface of colon carcinoma cells is a suitable way to activate and retarget T cells in a highly tumor-specific manner. For clinical purposes, B7 x anti-tumor-associated antigen (TAA) fusion proteins, which are equally effective but more specific compared with anti-CD28 monoclonal anti-bodies, thus may improve the tumor specificity of anti-CD3 x anti-TAA bispecific antibodies. Furthermore, B7-negative tumors can be converted into B7-positive tumors by B7 x anti-TAA fusion proteins without the need for B7 gene transfer to the malignant cells.

Crystal structure of the two N-terminal domains of g3p from filamentous phage fd at 1.9 A: evidence for conformational lability.

Infection of Escherichia coli by filamentous bacteriophages is mediated by the minor phage coat protein g3p and involves two distinct cellular receptors, the F' pilus and the periplasmic protein TolA. Recently we have shown that the two receptors are contacted in a sequential manner, such that binding of TolA by the N-terminal domain g3p-D1 is conditional on a primary interaction of the second g3p domain D2 with the F' pilus. In order to better understand this process, we have solved the crystal structure of the g3p-D1D2 fragment (residues 2-217) from filamentous phage fd to 1.9 A resolution and compared it to the recently published structure of the same fragment from the related Ff phage M13. While the structure of individual domains D1 and D2 of the two phages are very similar (rms<0.7 A), there is comparatively poor agreement for the overall D1D2 structure (rms>1.2 A). This is due to an apparent movement of domain D2 with respect to D1, which results in a widening of the inter-domain groove compared to the structure of the homologous M13 protein. The movement of D2 can be described as a rigid-body rotation around a hinge located at the end of a short anti-parallel beta-sheet connecting domains D1 and D2. Structural flexibility of at least parts of the D1D2 structure was also suggested by studying the thermal unfolding of g3p: the TolA binding site on D1, while fully blocked by D2 at 37 degrees C, becomes accessible after incubation at temperatures as low as 45 degrees C. Our results support a model for the early steps of phage infection whereby exposure of the coreceptor binding site on D1 is facilitated by a conformational change in the D1D2 structure, which in vivo is induced by binding to the F' pilus on the host cell and which can be mimicked in vitro by thermal unfolding.

Engineering antibodies for the clinic.

In the last ten years recombinant 'protein drugs' such as erythropoietin or tissue plasminogen activator have become widely used in the clinic. After some early setbacks antibodies look well placed to join them. A decade of antibody engineering is finally beginning to pay off with a string of chimeric and humanized antibodies gaining the Food and Drug Administration approval in the last two years. Here we will report on recent developments in the clinical application of antibodies, in particular, in the treatment of malignant lymphoma. We will also discuss some of the current strategies for the engineering of both whole antibodies (IgG) and recombinant antibody fragments for the next generation of antibody therapeutics.

The 2.0-A resolution crystal structure of a trimeric antibody fragment with noncognate VH-VL domain pairs shows a rearrangement of VH CDR3.

The 2.0-A resolution x-ray crystal structure of a novel trimeric antibody fragment, a "triabody," has been determined. The trimer is made up of polypeptides constructed in a manner identical to that previously described for some "diabodies": a VL domain directly fused to the C terminus of a VH domain-i.e., without any linker sequence. The trimer has three Fv heads with the polypeptides arranged in a cyclic, head-to-tail fashion. For the particular structure reported here, the polypeptide was constructed with a VH domain from one antibody fused to the VL domain from an unrelated antibody giving rise to "combinatorial" Fvs upon formation of the trimer. The structure shows that the exchange of the VL domain from antibody B1-8, a Vlambda domain, with the VL domain from antibody NQ11, a Vkappa domain, leads to a dramatic conformational change in the VH CDR3 loop of antibody B1-8. The magnitude of this change is similar to the largest of the conformational changes observed in antibody fragments in response to antigen binding. Combinatorial pairing of VH and VL domains constitutes a major component of antibody diversity. Conformationally flexible antigen-binding sites capable of adapting to the specific CDR3 loop context created upon VH-VL pairing may be employed by the immune system to maximize the structural diversity of the immune response.

The C-terminal domain of TolA is the coreceptor for filamentous phage infection of E. coli.

Filamentous bacteriophages infecting gram-negative bacteria display tropism for a variety of pilus structures. However, the obligatory coreceptor of phage infection, postulated from genetic studies, has remained elusive. Here we identify the C-terminal domain of the periplasmic protein TolA as the coreceptor for infection of Escherichia coli by phage fd and the N-terminal domain of the phage minor coat protein g3p as its cognate ligand. The neighboring g3p domain binds the primary receptor of phage infection, the F pilus, and blocks TolA binding in its absence. Contact with the pilus releases this blockage during infection. Our findings support a sequential two-way docking mechanism for phage infection, analogous to infection pathways proposed for a range of eukaryotic viruses including herpes simplex, adenoviruses, and also lentiviruses like HIV-1.

Retargeting serum immunoglobulin with bispecific diabodies.

Monospecific antibody fragments produced in bacteria lack the Fc portion of antibodies, and are therefore unable to recruit natural effector functions. We describe the use of a bispecific antibody fragment (diabody) to recruit the whole spectrum of antibody effector functions by retargeting serum immunoglobulin (Ig). One arm of the diabody was directed against the target antigen, and the other against the serum Ig. The bispecific diabodies were able to recruit complement, induce mononuclear phagocyte respiratory burst and phagocytosis, and promote synergistic cytotoxicity towards colon carcinoma cells in conjunction with CD8+ T-cells. Further, by virtue of binding to serum Ig their half-life (beta-phase) was increased fivefold compared to a control diabody of the same molecular weight. Such bispecific diabodies may provide an attractive alternative to monoclonal antibodies for serotherapy.

A conserved infection pathway for filamentous bacteriophages is suggested by the structure of the membrane penetration domain of the minor coat protein g3p from phage fd.

BACKGROUND: . Gene 3 protein (g3p), a minor coat protein from bacteriophage fd mediates infection of Escherichia coli bearing an F-pilus. Its N-terminal domain (g3p-D1) is essential for infection and mediates penetration of the phage into the host cytoplasm presumbly through interaction with the Tol complex in the E. coli membranes. Structural knowledge of g3p-D1 is both important for a molecular understanding of phage infection and of biotechnological relevance, as g3p-D1 represents the primary fusion partner in phage display technology. RESULTS: . The solution structure of g3p-D1 was determined by NMR spectroscopy. The principal structural element of g3p-D1 is formed by a six-stranded beta barrel topologically identical to a permutated SH3 domain but capped by an additional N-terminal alpha helix. The presence of structurally similar domains in the related E. coli phages, lke and 12-2, as well as in the cholera toxin transducing phage ctxφ is indicated. The structure of g3p-D1 resembles those of the recently described PTB and PDZ domains involved in eukaryotic signal transduction. CONCLUSIONS: . The predicted presence of similar structures in membrane penetration domains from widely diverging filamentous phages suggests they share a conserved infection pathway. The widespread hydrogen-bond network within the beta barrel and N-terminal alpha helix in combination with two disulphide bridges renders g3p-D1 a highly stable domain, which may be important for keeping phage infective in harsh extracellular environments.

Improved tumour targeting by disulphide stabilized diabodies expressed in Pichia pastoris.

Diabodies are dimeric antibody fragments held together by associated heavy and light chain variable domains present on different polypeptide chains. To improve their stability we have introduced cysteine residues into the V-domains to promote the disulphide crosslinking of the dimer. A crosslinked bivalent diabody against carcinoembryonic antigen (CEA) and a crosslinked bispecific diabody against CEA and the T-cell co-receptor CD3 were expressed from Pichia pastoris and Escherichia coli by secretion. From Pichia (but not E.coli) the chains were almost quantitatively crosslinked. Compared with the parent diabodies both crosslinked diabodies were more stable to heat (by >7 degrees C) and the crosslinked bivalent diabody showed improved localization to CEA+ human tumour xenografts in nude mice.

Mimicking somatic hypermutation: affinity maturation of antibodies displayed on bacteriophage using a bacterial mutator strain.

Human antibodies can now be isolated from antibody repertoires displayed on the surface of filamentous bacteriophage in a process that mimics the primary immune response. Here we have attempted to mimic the secondary response, the natural process of affinity maturation of antibodies occurring in germinal centres, by multiple cycles of random mutation and selection. Phage displaying a human antibody fragment recognising the hapten 2-phenyl-5-oxazolone were grown in a mutator strain of bacteria (Escherichia coli: mutD5) to generate a large repertoire of antibodies that should include the majority of possible single nucleotide point mutations. The repertoire of phage antibody mutants was then selected by binding to hapten. By multiple rounds of growth in the mutator strain, and increasingly stringent selection, we succeeded in isolating mutants with improved binding affinities; furthermore, the distribution of mutations and nucleotide substitution preferences strongly resembled those of somatic hypermutation. We then constructed a genealogical tree from the sequences of mutants taken at different rounds, and identified four sequentially acquired mutations that together improve the binding affinity of the antibody by a factor of 100-fold (from Kd 320 nM to 3.2 nM).

Specific killing of lymphoma cells by cytotoxic T-cells mediated by a bispecific diabody.

Antibody fragments produced by bacterial fermentation lack natural effector functions. Bispecific antibody fragments, however, can be endowed with effector functions, for example cell-mediated killing, by binding to and retargeting of cytotoxic cells. Diabodies are a class of engineered antibody fragments with two antigen binding sites, consisting of two associated chains; each chain consists of heavy and light chain variable domains linked by a short polypeptide linker. In contrast to IgG, or other antibody fragments in which the two binding sites can take up a range of orientations and spacings, the diabody structure is more rigid and compact, with the two binding sites separated by 65 lA (less than half the distance in IgG). To establish whether diabodies could also be used in cell-mediated killing, we have explored the use of a bispecific diabody binding to an idiotypic marker on mouse B-cell lymphoma (BCL-1) and to mouse CD3. The bispecific diabody activated naive T-cells and also mediated the specific killing of the lymphoma cells by cytotoxic T-cells. The diabody was less active in T-cell activation but 10-fold more active (w/v) in killing than an analogous bispecific IgG.

Crystal structure of a diabody, a bivalent antibody fragment.

BACKGROUND: Diabodies are dimeric antibody fragments. In each polypeptide, a heavy-chain variable domain (VH) is linked to a light-chain variable domain (VL) but unlike single-chain Fv fragments, each antigen-binding site is formed by pairing of one VH and one VL domain from the two different polypeptides. Diabodies thus have two antigen-binding sites, and can be bispecific. Direct structural evidence is lacking for the connections and dimeric interactions between the two polypeptides of the diabody. RESULTS: The 2.6 A resolution structure has been determined for a bivalent diabody with a flexible five-residue polypeptide linker between the (amino-terminal) VH and (carboxy-terminal) VL domains. The asymmetric unit of the crystal consists of four polypeptides comprising two diabodies; for one of these polypeptides the linker can be traced between the VH and VL domains. Within each diabody the two associated VH and VL domains make back-to-back interactions through the VH domains, and there is an extensive VL-VL interface between the two diabodies in the asymmetric unit. CONCLUSIONS: The structure of the diabody is very similar to that which had been predicted by molecular modelling. Diabodies directed against cell-surface antigens should be capable of bringing together two cells, such as in cell-targeted therapy, because the two antigen-binding sites of the diabody are at opposite ends of the molecule and separated by approximately 65 A.

Engineering bispecific antibodies.

Bispecific antibodies have immense potential as reagents and in medicine. Until recently, they were made by combining monoclonal antibodies of two different specificities in vitro, or by fusion of the corresponding hybridomas. Protein engineering now offers the chance to construct a range of small 'designer' bispecific antibodies using antibody fragments as building blocks.